[ { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=anthropic_claude-3-5-haiku-20241022", "run_spec": { "name": "helpdesk_call_summarization:model=anthropic_claude-3-5-haiku-20241022", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "anthropic/claude-3-5-haiku-20241022", "model": "anthropic/claude-3-5-haiku-20241022", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1372.1666666666667, "sum_squared": 1882841.3611111112, "min": 1372.1666666666667, "max": 1372.1666666666667, "mean": 1372.1666666666667, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 2.9493638438942993, "sum_squared": 8.698747083670957, "min": 2.9493638438942993, "max": 2.9493638438942993, "mean": 2.9493638438942993, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 460.10493827160496, "sum_squared": 211696.5542219174, "min": 460.10493827160496, "max": 460.10493827160496, "mean": 460.10493827160496, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1372.1666666666667, "sum_squared": 1882841.3611111112, "min": 1372.1666666666667, "max": 1372.1666666666667, "mean": 1372.1666666666667, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1372.1666666666667, "sum_squared": 1882841.3611111112, "min": 1372.1666666666667, "max": 1372.1666666666667, "mean": 1372.1666666666667, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 2.9493638438942993, "sum_squared": 8.698747083670957, "min": 2.9493638438942993, "max": 2.9493638438942993, "mean": 2.9493638438942993, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 2.9493638438942993, "sum_squared": 8.698747083670957, "min": 2.9493638438942993, "max": 2.9493638438942993, "mean": 2.9493638438942993, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 86.29012345679013, "sum_squared": 7445.9854061880815, "min": 86.29012345679013, "max": 86.29012345679013, "mean": 86.29012345679013, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 460.10493827160496, "sum_squared": 211696.5542219174, "min": 460.10493827160496, "max": 460.10493827160496, "mean": 460.10493827160496, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 460.10493827160496, "sum_squared": 211696.5542219174, "min": 460.10493827160496, "max": 460.10493827160496, "mean": 460.10493827160496, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7969821673525349, "sum_squared": 0.635180575077944, "min": 0.7969821673525349, "max": 0.7969821673525349, "mean": 0.7969821673525349, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7969821673525349, "sum_squared": 0.635180575077944, "min": 0.7969821673525349, "max": 0.7969821673525349, "mean": 0.7969821673525349, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7969821673525349, "sum_squared": 0.635180575077944, "min": 0.7969821673525349, "max": 0.7969821673525349, "mean": 0.7969821673525349, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=anthropic_claude-3-5-sonnet-20240620", "run_spec": { "name": "helpdesk_call_summarization:model=anthropic_claude-3-5-sonnet-20240620", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "anthropic/claude-3-5-sonnet-20240620", "model": "anthropic/claude-3-5-sonnet-20240620", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1372.1666666666667, "sum_squared": 1882841.3611111112, "min": 1372.1666666666667, "max": 1372.1666666666667, "mean": 1372.1666666666667, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 3.4916146360797646, "sum_squared": 12.191372766886428, "min": 3.4916146360797646, "max": 3.4916146360797646, "mean": 3.4916146360797646, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 683.5370370370371, "sum_squared": 467222.8810013718, "min": 683.5370370370371, "max": 683.5370370370371, "mean": 683.5370370370371, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1372.1666666666667, "sum_squared": 1882841.3611111112, "min": 1372.1666666666667, "max": 1372.1666666666667, "mean": 1372.1666666666667, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1372.1666666666667, "sum_squared": 1882841.3611111112, "min": 1372.1666666666667, "max": 1372.1666666666667, "mean": 1372.1666666666667, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 3.4916146360797646, "sum_squared": 12.191372766886428, "min": 3.4916146360797646, "max": 3.4916146360797646, "mean": 3.4916146360797646, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 3.4916146360797646, "sum_squared": 12.191372766886428, "min": 3.4916146360797646, "max": 3.4916146360797646, "mean": 3.4916146360797646, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 130.02469135802468, "sum_squared": 16906.420362749577, "min": 130.02469135802468, "max": 130.02469135802468, "mean": 130.02469135802468, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 683.5370370370371, "sum_squared": 467222.8810013718, "min": 683.5370370370371, "max": 683.5370370370371, "mean": 683.5370370370371, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 683.5370370370371, "sum_squared": 467222.8810013718, "min": 683.5370370370371, "max": 683.5370370370371, "mean": 683.5370370370371, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.8072702331961568, "sum_squared": 0.6516852294045773, "min": 0.8072702331961568, "max": 0.8072702331961568, "mean": 0.8072702331961568, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.8072702331961568, "sum_squared": 0.6516852294045773, "min": 0.8072702331961568, "max": 0.8072702331961568, "mean": 0.8072702331961568, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.8072702331961568, "sum_squared": 0.6516852294045773, "min": 0.8072702331961568, "max": 0.8072702331961568, "mean": 0.8072702331961568, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=deepseek-ai_deepseek-v3", "run_spec": { "name": "helpdesk_call_summarization:model=deepseek-ai_deepseek-v3", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/deepseek-v3", "model": "deepseek-ai/deepseek-v3", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1336.8456790123457, "sum_squared": 1787156.3694939797, "min": 1336.8456790123457, "max": 1336.8456790123457, "mean": 1336.8456790123457, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 15.269175598650802, "sum_squared": 233.14772346243308, "min": 15.269175598650802, "max": 15.269175598650802, "mean": 15.269175598650802, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1336.8456790123457, "sum_squared": 1787156.3694939797, "min": 1336.8456790123457, "max": 1336.8456790123457, "mean": 1336.8456790123457, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1336.8456790123457, "sum_squared": 1787156.3694939797, "min": 1336.8456790123457, "max": 1336.8456790123457, "mean": 1336.8456790123457, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 15.269175598650802, "sum_squared": 233.14772346243308, "min": 15.269175598650802, "max": 15.269175598650802, "mean": 15.269175598650802, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 15.269175598650802, "sum_squared": 233.14772346243308, "min": 15.269175598650802, "max": 15.269175598650802, "mean": 15.269175598650802, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.8161865569272949, "sum_squared": 0.6661604957088324, "min": 0.8161865569272949, "max": 0.8161865569272949, "mean": 0.8161865569272949, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.8161865569272949, "sum_squared": 0.6661604957088324, "min": 0.8161865569272949, "max": 0.8161865569272949, "mean": 0.8161865569272949, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.8161865569272949, "sum_squared": 0.6661604957088324, "min": 0.8161865569272949, "max": 0.8161865569272949, "mean": 0.8161865569272949, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=google_gemini-1.5-flash-002", "run_spec": { "name": "helpdesk_call_summarization:model=google_gemini-1.5-flash-002", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "google/gemini-1.5-flash-002", "model": "google/gemini-1.5-flash-002", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1394.8086419753085, "sum_squared": 1945491.1477290045, "min": 1394.8086419753085, "max": 1394.8086419753085, "mean": 1394.8086419753085, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 0.9047453506493274, "sum_squared": 0.8185641495215743, "min": 0.9047453506493274, "max": 0.9047453506493274, "mean": 0.9047453506493274, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1394.8086419753085, "sum_squared": 1945491.1477290045, "min": 1394.8086419753085, "max": 1394.8086419753085, "mean": 1394.8086419753085, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1394.8086419753085, "sum_squared": 1945491.1477290045, "min": 1394.8086419753085, "max": 1394.8086419753085, "mean": 1394.8086419753085, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.9047453506493274, "sum_squared": 0.8185641495215743, "min": 0.9047453506493274, "max": 0.9047453506493274, "mean": 0.9047453506493274, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.9047453506493274, "sum_squared": 0.8185641495215743, "min": 0.9047453506493274, "max": 0.9047453506493274, "mean": 0.9047453506493274, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7764060356652923, "sum_squared": 0.6028063322174951, "min": 0.7764060356652923, "max": 0.7764060356652923, "mean": 0.7764060356652923, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7764060356652923, "sum_squared": 0.6028063322174951, "min": 0.7764060356652923, "max": 0.7764060356652923, "mean": 0.7764060356652923, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7764060356652923, "sum_squared": 0.6028063322174951, "min": 0.7764060356652923, "max": 0.7764060356652923, "mean": 0.7764060356652923, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=google_gemini-1.5-pro-002", "run_spec": { "name": "helpdesk_call_summarization:model=google_gemini-1.5-pro-002", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "google/gemini-1.5-pro-002", "model": "google/gemini-1.5-pro-002", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1394.8086419753085, "sum_squared": 1945491.1477290045, "min": 1394.8086419753085, "max": 1394.8086419753085, "mean": 1394.8086419753085, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 2.7368182208802967, "sum_squared": 7.490173974142392, "min": 2.7368182208802967, "max": 2.7368182208802967, "mean": 2.7368182208802967, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1394.8086419753085, "sum_squared": 1945491.1477290045, "min": 1394.8086419753085, "max": 1394.8086419753085, "mean": 1394.8086419753085, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1394.8086419753085, "sum_squared": 1945491.1477290045, "min": 1394.8086419753085, "max": 1394.8086419753085, "mean": 1394.8086419753085, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 2.7368182208802967, "sum_squared": 7.490173974142392, "min": 2.7368182208802967, "max": 2.7368182208802967, "mean": 2.7368182208802967, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 2.7368182208802967, "sum_squared": 7.490173974142392, "min": 2.7368182208802967, "max": 2.7368182208802967, "mean": 2.7368182208802967, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.785322359396431, "sum_squared": 0.6167312081679772, "min": 0.785322359396431, "max": 0.785322359396431, "mean": 0.785322359396431, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.785322359396431, "sum_squared": 0.6167312081679772, "min": 0.785322359396431, "max": 0.785322359396431, "mean": 0.785322359396431, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.785322359396431, "sum_squared": 0.6167312081679772, "min": 0.785322359396431, "max": 0.785322359396431, "mean": 0.785322359396431, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=meta_llama-3.1-405b-instruct-turbo", "run_spec": { "name": "helpdesk_call_summarization:model=meta_llama-3.1-405b-instruct-turbo", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/llama-3.1-405b-instruct-turbo", "model": "meta/llama-3.1-405b-instruct-turbo", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 3.4933589741035744, "sum_squared": 12.203556921949978, "min": 3.4933589741035744, "max": 3.4933589741035744, "mean": 3.4933589741035744, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": -17.894988236128505, "sum_squared": 320.2306039711776, "min": -17.894988236128505, "max": -17.894988236128505, "mean": -17.894988236128505, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 620.4753086419753, "sum_squared": 384989.6086343545, "min": 620.4753086419753, "max": 620.4753086419753, "mean": 620.4753086419753, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.1655099442793402, "sum_squared": 1.3584134302140307, "min": 1.1655099442793402, "max": 1.1655099442793402, "mean": 1.1655099442793402, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.041608441827498704, "sum_squared": 0.0017312624313123438, "min": 0.041608441827498704, "max": 0.041608441827498704, "mean": 0.041608441827498704, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": -0.028840774140223225, "sum_squared": 0.0008317902530073687, "min": -0.028840774140223225, "max": -0.028840774140223225, "mean": -0.028840774140223225, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 3.4933589741035744, "sum_squared": 12.203556921949978, "min": 3.4933589741035744, "max": 3.4933589741035744, "mean": 3.4933589741035744, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 3.4933589741035744, "sum_squared": 12.203556921949978, "min": 3.4933589741035744, "max": 3.4933589741035744, "mean": 3.4933589741035744, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": -17.894988236128505, "sum_squared": 320.2306039711776, "min": -17.894988236128505, "max": -17.894988236128505, "mean": -17.894988236128505, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": -17.894988236128505, "sum_squared": 320.2306039711776, "min": -17.894988236128505, "max": -17.894988236128505, "mean": -17.894988236128505, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 116.8395061728395, "sum_squared": 13651.470202713, "min": 116.8395061728395, "max": 116.8395061728395, "mean": 116.8395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 620.4753086419753, "sum_squared": 384989.6086343545, "min": 620.4753086419753, "max": 620.4753086419753, "mean": 620.4753086419753, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 620.4753086419753, "sum_squared": 384989.6086343545, "min": 620.4753086419753, "max": 620.4753086419753, "mean": 620.4753086419753, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7976680384087768, "sum_squared": 0.6362742994989058, "min": 0.7976680384087768, "max": 0.7976680384087768, "mean": 0.7976680384087768, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7976680384087768, "sum_squared": 0.6362742994989058, "min": 0.7976680384087768, "max": 0.7976680384087768, "mean": 0.7976680384087768, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7976680384087768, "sum_squared": 0.6362742994989058, "min": 0.7976680384087768, "max": 0.7976680384087768, "mean": 0.7976680384087768, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=meta_llama-3.1-70b-instruct-turbo", "run_spec": { "name": "helpdesk_call_summarization:model=meta_llama-3.1-70b-instruct-turbo", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/llama-3.1-70b-instruct-turbo", "model": "meta/llama-3.1-70b-instruct-turbo", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 1.0229934556984608, "sum_squared": 1.0465156104018787, "min": 1.0229934556984608, "max": 1.0229934556984608, "mean": 1.0229934556984608, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": -24.021595388222757, "sum_squared": 577.0370449954848, "min": -24.021595388222757, "max": -24.021595388222757, "mean": -24.021595388222757, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 553.5987654320987, "sum_squared": 306471.59308794385, "min": 553.5987654320987, "max": 553.5987654320987, "mean": 553.5987654320987, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.260175999917213, "sum_squared": 1.5880435507673478, "min": 1.260175999917213, "max": 1.260175999917213, "mean": 1.260175999917213, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.06260100040826563, "sum_squared": 0.003918885252115674, "min": 0.06260100040826563, "max": 0.06260100040826563, "mean": 0.06260100040826563, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": -0.04339170693322131, "sum_squared": 0.0018828402305785662, "min": -0.04339170693322131, "max": -0.04339170693322131, "mean": -0.04339170693322131, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0229934556984608, "sum_squared": 1.0465156104018787, "min": 1.0229934556984608, "max": 1.0229934556984608, "mean": 1.0229934556984608, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0229934556984608, "sum_squared": 1.0465156104018787, "min": 1.0229934556984608, "max": 1.0229934556984608, "mean": 1.0229934556984608, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": -24.021595388222757, "sum_squared": 577.0370449954848, "min": -24.021595388222757, "max": -24.021595388222757, "mean": -24.021595388222757, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": -24.021595388222757, "sum_squared": 577.0370449954848, "min": -24.021595388222757, "max": -24.021595388222757, "mean": -24.021595388222757, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 103.87654320987654, "sum_squared": 10790.33622923335, "min": 103.87654320987654, "max": 103.87654320987654, "mean": 103.87654320987654, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 553.5987654320987, "sum_squared": 306471.59308794385, "min": 553.5987654320987, "max": 553.5987654320987, "mean": 553.5987654320987, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 553.5987654320987, "sum_squared": 306471.59308794385, "min": 553.5987654320987, "max": 553.5987654320987, "mean": 553.5987654320987, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7894375857338795, "sum_squared": 0.6232117017693363, "min": 0.7894375857338795, "max": 0.7894375857338795, "mean": 0.7894375857338795, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7894375857338795, "sum_squared": 0.6232117017693363, "min": 0.7894375857338795, "max": 0.7894375857338795, "mean": 0.7894375857338795, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7894375857338795, "sum_squared": 0.6232117017693363, "min": 0.7894375857338795, "max": 0.7894375857338795, "mean": 0.7894375857338795, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=meta_llama-3.1-8b-instruct-turbo", "run_spec": { "name": "helpdesk_call_summarization:model=meta_llama-3.1-8b-instruct-turbo", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/llama-3.1-8b-instruct-turbo", "model": "meta/llama-3.1-8b-instruct-turbo", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 0.6208152520803758, "sum_squared": 0.3854115772156205, "min": 0.6208152520803758, "max": 0.6208152520803758, "mean": 0.6208152520803758, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": -35.72712776274848, "sum_squared": 1276.4276581757533, "min": -35.72712776274848, "max": -35.72712776274848, "mean": -35.72712776274848, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 621.4567901234568, "sum_squared": 386208.54199055023, "min": 621.4567901234568, "max": 621.4567901234568, "mean": 621.4567901234568, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.3468546931185368, "sum_squared": 1.814017564375428, "min": 1.3468546931185368, "max": 1.3468546931185368, "mean": 1.3468546931185368, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.08293955568219664, "sum_squared": 0.006878969896760198, "min": 0.08293955568219664, "max": 0.08293955568219664, "mean": 0.08293955568219664, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": -0.057489319178009195, "sum_squared": 0.0033050218195510156, "min": -0.057489319178009195, "max": -0.057489319178009195, "mean": -0.057489319178009195, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1325.1604938271605, "sum_squared": 1756050.3344002438, "min": 1325.1604938271605, "max": 1325.1604938271605, "mean": 1325.1604938271605, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.6208152520803758, "sum_squared": 0.3854115772156205, "min": 0.6208152520803758, "max": 0.6208152520803758, "mean": 0.6208152520803758, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.6208152520803758, "sum_squared": 0.3854115772156205, "min": 0.6208152520803758, "max": 0.6208152520803758, "mean": 0.6208152520803758, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": -35.72712776274848, "sum_squared": 1276.4276581757533, "min": -35.72712776274848, "max": -35.72712776274848, "mean": -35.72712776274848, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": -35.72712776274848, "sum_squared": 1276.4276581757533, "min": -35.72712776274848, "max": -35.72712776274848, "mean": -35.72712776274848, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 119.98148148148148, "sum_squared": 14395.555898491084, "min": 119.98148148148148, "max": 119.98148148148148, "mean": 119.98148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 621.4567901234568, "sum_squared": 386208.54199055023, "min": 621.4567901234568, "max": 621.4567901234568, "mean": 621.4567901234568, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 621.4567901234568, "sum_squared": 386208.54199055023, "min": 621.4567901234568, "max": 621.4567901234568, "mean": 621.4567901234568, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7661179698216716, "sum_squared": 0.5869367436836797, "min": 0.7661179698216716, "max": 0.7661179698216716, "mean": 0.7661179698216716, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7661179698216716, "sum_squared": 0.5869367436836797, "min": 0.7661179698216716, "max": 0.7661179698216716, "mean": 0.7661179698216716, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7661179698216716, "sum_squared": 0.5869367436836797, "min": 0.7661179698216716, "max": 0.7661179698216716, "mean": 0.7661179698216716, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=mistralai_mistral-7b-instruct-v0.3", "run_spec": { "name": "helpdesk_call_summarization:model=mistralai_mistral-7b-instruct-v0.3", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/mistral-7b-instruct-v0.3", "model": "mistralai/mistral-7b-instruct-v0.3", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 0.952246833730627, "sum_squared": 0.9067740323500044, "min": 0.952246833730627, "max": 0.952246833730627, "mean": 0.952246833730627, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": -28.110070563464035, "sum_squared": 790.1760670829273, "min": -28.110070563464035, "max": -28.110070563464035, "mean": -28.110070563464035, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 625.2407407407408, "sum_squared": 390925.9838820302, "min": 625.2407407407408, "max": 625.2407407407408, "mean": 625.2407407407408, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.240677824363209, "sum_squared": 1.539281463866626, "min": 1.240677824363209, "max": 1.240677824363209, "mean": 1.240677824363209, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.06486183122504556, "sum_squared": 0.004207057149866295, "min": 0.06486183122504556, "max": 0.06486183122504556, "mean": 0.06486183122504556, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": -0.04495879543959535, "sum_squared": 0.0020212932873793794, "min": -0.04495879543959535, "max": -0.04495879543959535, "mean": -0.04495879543959535, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.952246833730627, "sum_squared": 0.9067740323500044, "min": 0.952246833730627, "max": 0.952246833730627, "mean": 0.952246833730627, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.952246833730627, "sum_squared": 0.9067740323500044, "min": 0.952246833730627, "max": 0.952246833730627, "mean": 0.952246833730627, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": -28.110070563464035, "sum_squared": 790.1760670829273, "min": -28.110070563464035, "max": -28.110070563464035, "mean": -28.110070563464035, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": -28.110070563464035, "sum_squared": 790.1760670829273, "min": -28.110070563464035, "max": -28.110070563464035, "mean": -28.110070563464035, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 130.34567901234567, "sum_squared": 16989.996037189452, "min": 130.34567901234567, "max": 130.34567901234567, "mean": 130.34567901234567, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 625.2407407407408, "sum_squared": 390925.9838820302, "min": 625.2407407407408, "max": 625.2407407407408, "mean": 625.2407407407408, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 625.2407407407408, "sum_squared": 390925.9838820302, "min": 625.2407407407408, "max": 625.2407407407408, "mean": 625.2407407407408, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7585733882030161, "sum_squared": 0.5754335852898038, "min": 0.7585733882030161, "max": 0.7585733882030161, "mean": 0.7585733882030161, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7585733882030161, "sum_squared": 0.5754335852898038, "min": 0.7585733882030161, "max": 0.7585733882030161, "mean": 0.7585733882030161, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7585733882030161, "sum_squared": 0.5754335852898038, "min": 0.7585733882030161, "max": 0.7585733882030161, "mean": 0.7585733882030161, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=mistralai_mixtral-8x22b-instruct-v0.1", "run_spec": { "name": "helpdesk_call_summarization:model=mistralai_mixtral-8x22b-instruct-v0.1", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/mixtral-8x22b-instruct-v0.1", "model": "mistralai/mixtral-8x22b-instruct-v0.1", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 3.4505287408828735, "sum_squared": 11.906148591658749, "min": 3.4505287408828735, "max": 3.4505287408828735, "mean": 3.4505287408828735, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": -21.759122981938592, "sum_squared": 473.4594329431282, "min": -21.759122981938592, "max": -21.759122981938592, "mean": -21.759122981938592, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 441.15432098765433, "sum_squared": 194617.13492607835, "min": 441.15432098765433, "max": 441.15432098765433, "mean": 441.15432098765433, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.261309934633061, "sum_squared": 1.5909027512040566, "min": 1.261309934633061, "max": 1.261309934633061, "mean": 1.261309934633061, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.07115827121415497, "sum_squared": 0.005063499562187235, "min": 0.07115827121415497, "max": 0.07115827121415497, "mean": 0.07115827121415497, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": -0.049323155065611426, "sum_squared": 0.0024327736256263503, "min": -0.049323155065611426, "max": -0.049323155065611426, "mean": -0.049323155065611426, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 3.4505287408828735, "sum_squared": 11.906148591658749, "min": 3.4505287408828735, "max": 3.4505287408828735, "mean": 3.4505287408828735, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 3.4505287408828735, "sum_squared": 11.906148591658749, "min": 3.4505287408828735, "max": 3.4505287408828735, "mean": 3.4505287408828735, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": -21.759122981938592, "sum_squared": 473.4594329431282, "min": -21.759122981938592, "max": -21.759122981938592, "mean": -21.759122981938592, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": -21.759122981938592, "sum_squared": 473.4594329431282, "min": -21.759122981938592, "max": -21.759122981938592, "mean": -21.759122981938592, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 93.72839506172839, "sum_squared": 8785.01204084743, "min": 93.72839506172839, "max": 93.72839506172839, "mean": 93.72839506172839, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 441.15432098765433, "sum_squared": 194617.13492607835, "min": 441.15432098765433, "max": 441.15432098765433, "mean": 441.15432098765433, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 441.15432098765433, "sum_squared": 194617.13492607835, "min": 441.15432098765433, "max": 441.15432098765433, "mean": 441.15432098765433, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7908093278463623, "sum_squared": 0.6253793930088153, "min": 0.7908093278463623, "max": 0.7908093278463623, "mean": 0.7908093278463623, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7908093278463623, "sum_squared": 0.6253793930088153, "min": 0.7908093278463623, "max": 0.7908093278463623, "mean": 0.7908093278463623, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7908093278463623, "sum_squared": 0.6253793930088153, "min": 0.7908093278463623, "max": 0.7908093278463623, "mean": 0.7908093278463623, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=mistralai_mixtral-8x7b-instruct-v0.1", "run_spec": { "name": "helpdesk_call_summarization:model=mistralai_mixtral-8x7b-instruct-v0.1", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/mixtral-8x7b-instruct-v0.1", "model": "mistralai/mixtral-8x7b-instruct-v0.1", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 1.2561425000061224, "sum_squared": 1.5778939803216312, "min": 1.2561425000061224, "max": 1.2561425000061224, "mean": 1.2561425000061224, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": -10.782968182736136, "sum_squared": 116.27240282989985, "min": -10.782968182736136, "max": -10.782968182736136, "mean": -10.782968182736136, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 513.7654320987655, "sum_squared": 263954.9192196312, "min": 513.7654320987655, "max": 513.7654320987655, "mean": 513.7654320987655, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.1025214446278306, "sum_squared": 1.2155535358642386, "min": 1.1025214446278306, "max": 1.1025214446278306, "mean": 1.1025214446278306, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.030279450020114118, "sum_squared": 0.0009168450935205889, "min": 0.030279450020114118, "max": 0.030279450020114118, "mean": 0.030279450020114118, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": -0.02098811541034788, "sum_squared": 0.00044050098847808213, "min": -0.02098811541034788, "max": -0.02098811541034788, "mean": -0.02098811541034788, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1491.9197530864199, "sum_squared": 2225824.549649444, "min": 1491.9197530864199, "max": 1491.9197530864199, "mean": 1491.9197530864199, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.2561425000061224, "sum_squared": 1.5778939803216312, "min": 1.2561425000061224, "max": 1.2561425000061224, "mean": 1.2561425000061224, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.2561425000061224, "sum_squared": 1.5778939803216312, "min": 1.2561425000061224, "max": 1.2561425000061224, "mean": 1.2561425000061224, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": -10.782968182736136, "sum_squared": 116.27240282989985, "min": -10.782968182736136, "max": -10.782968182736136, "mean": -10.782968182736136, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": -10.782968182736136, "sum_squared": 116.27240282989985, "min": -10.782968182736136, "max": -10.782968182736136, "mean": -10.782968182736136, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 110.48148148148148, "sum_squared": 12206.157750342936, "min": 110.48148148148148, "max": 110.48148148148148, "mean": 110.48148148148148, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 513.7654320987655, "sum_squared": 263954.9192196312, "min": 513.7654320987655, "max": 513.7654320987655, "mean": 513.7654320987655, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 513.7654320987655, "sum_squared": 263954.9192196312, "min": 513.7654320987655, "max": 513.7654320987655, "mean": 513.7654320987655, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.758573388203016, "sum_squared": 0.5754335852898036, "min": 0.758573388203016, "max": 0.758573388203016, "mean": 0.758573388203016, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.758573388203016, "sum_squared": 0.5754335852898036, "min": 0.758573388203016, "max": 0.758573388203016, "mean": 0.758573388203016, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.758573388203016, "sum_squared": 0.5754335852898036, "min": 0.758573388203016, "max": 0.758573388203016, "mean": 0.758573388203016, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=openai_gpt-4o-2024-08-06", "run_spec": { "name": "helpdesk_call_summarization:model=openai_gpt-4o-2024-08-06", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "openai/gpt-4o-2024-08-06", "model": "openai/gpt-4o-2024-08-06", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1298.9444444444443, "sum_squared": 1687256.669753086, "min": 1298.9444444444443, "max": 1298.9444444444443, "mean": 1298.9444444444443, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 6.5025614706086525, "sum_squared": 42.28330567904416, "min": 6.5025614706086525, "max": 6.5025614706086525, "mean": 6.5025614706086525, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 550.4197530864197, "sum_squared": 302961.9045877153, "min": 550.4197530864197, "max": 550.4197530864197, "mean": 550.4197530864197, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1298.9444444444443, "sum_squared": 1687256.669753086, "min": 1298.9444444444443, "max": 1298.9444444444443, "mean": 1298.9444444444443, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1298.9444444444443, "sum_squared": 1687256.669753086, "min": 1298.9444444444443, "max": 1298.9444444444443, "mean": 1298.9444444444443, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 6.5025614706086525, "sum_squared": 42.28330567904416, "min": 6.5025614706086525, "max": 6.5025614706086525, "mean": 6.5025614706086525, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 6.5025614706086525, "sum_squared": 42.28330567904416, "min": 6.5025614706086525, "max": 6.5025614706086525, "mean": 6.5025614706086525, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 99.65432098765432, "sum_squared": 9930.983691510439, "min": 99.65432098765432, "max": 99.65432098765432, "mean": 99.65432098765432, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 550.4197530864197, "sum_squared": 302961.9045877153, "min": 550.4197530864197, "max": 550.4197530864197, "mean": 550.4197530864197, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 550.4197530864197, "sum_squared": 302961.9045877153, "min": 550.4197530864197, "max": 550.4197530864197, "mean": 550.4197530864197, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.8134430727023293, "sum_squared": 0.661689632527407, "min": 0.8134430727023293, "max": 0.8134430727023293, "mean": 0.8134430727023293, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.8134430727023293, "sum_squared": 0.661689632527407, "min": 0.8134430727023293, "max": 0.8134430727023293, "mean": 0.8134430727023293, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.8134430727023293, "sum_squared": 0.661689632527407, "min": 0.8134430727023293, "max": 0.8134430727023293, "mean": 0.8134430727023293, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=openai_gpt-4o-mini-2024-07-18", "run_spec": { "name": "helpdesk_call_summarization:model=openai_gpt-4o-mini-2024-07-18", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "openai/gpt-4o-mini-2024-07-18", "model": "openai/gpt-4o-mini-2024-07-18", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1298.9444444444443, "sum_squared": 1687256.669753086, "min": 1298.9444444444443, "max": 1298.9444444444443, "mean": 1298.9444444444443, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 2.032809728457604, "sum_squared": 4.132315392111877, "min": 2.032809728457604, "max": 2.032809728457604, "mean": 2.032809728457604, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 539.2037037037037, "sum_squared": 290740.6340877915, "min": 539.2037037037037, "max": 539.2037037037037, "mean": 539.2037037037037, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1298.9444444444443, "sum_squared": 1687256.669753086, "min": 1298.9444444444443, "max": 1298.9444444444443, "mean": 1298.9444444444443, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1298.9444444444443, "sum_squared": 1687256.669753086, "min": 1298.9444444444443, "max": 1298.9444444444443, "mean": 1298.9444444444443, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 2.032809728457604, "sum_squared": 4.132315392111877, "min": 2.032809728457604, "max": 2.032809728457604, "mean": 2.032809728457604, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 2.032809728457604, "sum_squared": 4.132315392111877, "min": 2.032809728457604, "max": 2.032809728457604, "mean": 2.032809728457604, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 95.28395061728395, "sum_squared": 9079.031245237007, "min": 95.28395061728395, "max": 95.28395061728395, "mean": 95.28395061728395, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 539.2037037037037, "sum_squared": 290740.6340877915, "min": 539.2037037037037, "max": 539.2037037037037, "mean": 539.2037037037037, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 539.2037037037037, "sum_squared": 290740.6340877915, "min": 539.2037037037037, "max": 539.2037037037037, "mean": 539.2037037037037, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7921810699588453, "sum_squared": 0.627550847601141, "min": 0.7921810699588453, "max": 0.7921810699588453, "mean": 0.7921810699588453, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7921810699588453, "sum_squared": 0.627550847601141, "min": 0.7921810699588453, "max": 0.7921810699588453, "mean": 0.7921810699588453, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7921810699588453, "sum_squared": 0.627550847601141, "min": 0.7921810699588453, "max": 0.7921810699588453, "mean": 0.7921810699588453, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=qwen_qwen2.5-72b-instruct-turbo", "run_spec": { "name": "helpdesk_call_summarization:model=qwen_qwen2.5-72b-instruct-turbo", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/qwen2.5-72b-instruct-turbo", "model": "qwen/qwen2.5-72b-instruct-turbo", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1332.7777777777778, "sum_squared": 1776296.6049382717, "min": 1332.7777777777778, "max": 1332.7777777777778, "mean": 1332.7777777777778, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 1.812830861703849, "sum_squared": 3.28635573314592, "min": 1.812830861703849, "max": 1.812830861703849, "mean": 1.812830861703849, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": -14.090352056650909, "sum_squared": 198.5380210803665, "min": -14.090352056650909, "max": -14.090352056650909, "mean": -14.090352056650909, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 467.11728395061726, "sum_squared": 218198.5569654016, "min": 467.11728395061726, "max": 467.11728395061726, "mean": 467.11728395061726, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.1723796063944947, "sum_squared": 1.3744739414897102, "min": 1.1723796063944947, "max": 1.1723796063944947, "mean": 1.1723796063944947, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.043518152153536924, "sum_squared": 0.0018938295668583904, "min": 0.043518152153536924, "max": 0.043518152153536924, "mean": 0.043518152153536924, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": -0.030164484468402832, "sum_squared": 0.0009098961232445157, "min": -0.030164484468402832, "max": -0.030164484468402832, "mean": -0.030164484468402832, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1332.7777777777778, "sum_squared": 1776296.6049382717, "min": 1332.7777777777778, "max": 1332.7777777777778, "mean": 1332.7777777777778, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1332.7777777777778, "sum_squared": 1776296.6049382717, "min": 1332.7777777777778, "max": 1332.7777777777778, "mean": 1332.7777777777778, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.812830861703849, "sum_squared": 3.28635573314592, "min": 1.812830861703849, "max": 1.812830861703849, "mean": 1.812830861703849, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.812830861703849, "sum_squared": 3.28635573314592, "min": 1.812830861703849, "max": 1.812830861703849, "mean": 1.812830861703849, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": -14.090352056650909, "sum_squared": 198.5380210803665, "min": -14.090352056650909, "max": -14.090352056650909, "mean": -14.090352056650909, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": -14.090352056650909, "sum_squared": 198.5380210803665, "min": -14.090352056650909, "max": -14.090352056650909, "mean": -14.090352056650909, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 88.59876543209876, "sum_squared": 7849.741236092058, "min": 88.59876543209876, "max": 88.59876543209876, "mean": 88.59876543209876, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 467.11728395061726, "sum_squared": 218198.5569654016, "min": 467.11728395061726, "max": 467.11728395061726, "mean": 467.11728395061726, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 467.11728395061726, "sum_squared": 218198.5569654016, "min": 467.11728395061726, "max": 467.11728395061726, "mean": 467.11728395061726, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7969821673525352, "sum_squared": 0.6351805750779445, "min": 0.7969821673525352, "max": 0.7969821673525352, "mean": 0.7969821673525352, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7969821673525352, "sum_squared": 0.6351805750779445, "min": 0.7969821673525352, "max": 0.7969821673525352, "mean": 0.7969821673525352, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7969821673525352, "sum_squared": 0.6351805750779445, "min": 0.7969821673525352, "max": 0.7969821673525352, "mean": 0.7969821673525352, "variance": 0.0, "stddev": 0.0 } ] }, { "run_path": "benchmark_output/runs/v0.1.0/helpdesk_call_summarization:model=qwen_qwen2.5-7b-instruct-turbo", "run_spec": { "name": "helpdesk_call_summarization:model=qwen_qwen2.5-7b-instruct-turbo", "scenario_spec": { "class_name": "helm.benchmark.scenarios.helpdesk_call_summarization_scenario.HelpdeskCallSummarizationScenario", "args": {} }, "adapter_spec": { "method": "generation", "global_prefix": "", "global_suffix": "", "instructions": "The following is a call transcript of a call between a compnay's employee and the company's IT helpdesk. Summarize the call transcript in under 100 words.", "input_prefix": "### Call Transcript\n", "input_suffix": "", "reference_prefix": "A. ", "reference_suffix": "\n", "chain_of_thought_prefix": "", "chain_of_thought_suffix": "\n", "output_prefix": "", "output_suffix": "", "instance_prefix": "\n", "substitutions": [], "max_train_instances": 0, "max_eval_instances": 100000, "num_outputs": 1, "num_train_trials": 1, "num_trials": 1, "sample_train": true, "model_deployment": "together/qwen2.5-7b-instruct-turbo", "model": "qwen/qwen2.5-7b-instruct-turbo", "temperature": 0.0, "max_tokens": 512, "stop_sequences": [], "multi_label": false }, "metric_specs": [ { "class_name": "helm.benchmark.metrics.basic_metrics.BasicGenerationMetric", "args": { "names": [] } }, { "class_name": "helm.benchmark.metrics.basic_metrics.BasicReferenceMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.basic_metrics.InstancesPerSplitMetric", "args": {} }, { "class_name": "helm.benchmark.metrics.annotation_metrics.AnnotationLikertScaleMetric", "args": { "annotator_name": "helpdesk_call_center_summarization", "key": "score", "min_score": 1, "max_score": 10 } } ], "data_augmenter_spec": { "perturbation_specs": [], "should_augment_train_instances": false, "should_include_original_train": false, "should_skip_unchanged_train": false, "should_augment_eval_instances": false, "should_include_original_eval": false, "should_skip_unchanged_eval": false, "seeds_per_instance": 1 }, "groups": [ "helpdesk_call_summarization" ], "annotators": [ { "class_name": "helm.benchmark.annotation.helpdesk_call_summarization_annotator.HelpdeskCallSummarizationAnnotator", "args": {} } ] }, "stats": [ { "name": { "name": "num_references", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test" }, "count": 1, "sum": 1332.7777777777778, "sum_squared": 1776296.6049382717, "min": 1332.7777777777778, "max": 1332.7777777777778, "mean": 1332.7777777777778, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test" }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test" }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "training_co2_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "training_energy_cost", "split": "test" }, "count": 0, "sum": 0.0, "sum_squared": 0.0 }, { "name": { "name": "inference_runtime", "split": "test" }, "count": 1, "sum": 0.772343741522895, "sum_squared": 0.5965148550695845, "min": 0.772343741522895, "max": 0.772343741522895, "mean": 0.772343741522895, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test" }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test" }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test" }, "count": 1, "sum": -19.53333554061788, "sum_squared": 381.55119734236564, "min": -19.53333554061788, "max": -19.53333554061788, "mean": -19.53333554061788, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test" }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test" }, "count": 1, "sum": 360.88271604938274, "sum_squared": 130236.33474317941, "min": 360.88271604938274, "max": 360.88271604938274, "mean": 360.88271604938274, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "perplexity", "split": "test" }, "count": 1, "sum": 1.3298357423655804, "sum_squared": 1.7684631016730143, "min": 1.3298357423655804, "max": 1.3298357423655804, "mean": 1.3298357423655804, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "bits_per_byte", "split": "test" }, "count": 1, "sum": 0.07808810193230022, "sum_squared": 0.006097751663389309, "min": 0.07808810193230022, "max": 0.07808810193230022, "mean": 0.07808810193230022, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob_per_byte", "split": "test" }, "count": 1, "sum": -0.054126547689651514, "sum_squared": 0.0029296831648001195, "min": -0.054126547689651514, "max": -0.054126547689651514, "mean": -0.054126547689651514, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_references", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_trials", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1332.7777777777778, "sum_squared": 1776296.6049382717, "min": 1332.7777777777778, "max": 1332.7777777777778, "mean": 1332.7777777777778, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_prompt_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1332.7777777777778, "sum_squared": 1776296.6049382717, "min": 1332.7777777777778, "max": 1332.7777777777778, "mean": 1332.7777777777778, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_completion_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_output_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.772343741522895, "sum_squared": 0.5965148550695845, "min": 0.772343741522895, "max": 0.772343741522895, "mean": 0.772343741522895, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "inference_runtime", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.772343741522895, "sum_squared": 0.5965148550695845, "min": 0.772343741522895, "max": 0.772343741522895, "mean": 0.772343741522895, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "batch_size", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_length", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_stop", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_endoftext", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "finish_reason_unknown", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 1.0, "sum_squared": 1.0, "min": 1.0, "max": 1.0, "mean": 1.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_train_instances", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "prompt_truncated", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.0, "sum_squared": 0.0, "min": 0.0, "max": 0.0, "mean": 0.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": -19.53333554061788, "sum_squared": 381.55119734236564, "min": -19.53333554061788, "max": -19.53333554061788, "mean": -19.53333554061788, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "logprob", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": -19.53333554061788, "sum_squared": 381.55119734236564, "min": -19.53333554061788, "max": -19.53333554061788, "mean": -19.53333554061788, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_perplexity_tokens", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 68.5246913580247, "sum_squared": 4695.633325712544, "min": 68.5246913580247, "max": 68.5246913580247, "mean": 68.5246913580247, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 360.88271604938274, "sum_squared": 130236.33474317941, "min": 360.88271604938274, "max": 360.88271604938274, "mean": 360.88271604938274, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_bytes", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 360.88271604938274, "sum_squared": 130236.33474317941, "min": 360.88271604938274, "max": 360.88271604938274, "mean": 360.88271604938274, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "num_instances", "split": "test" }, "count": 1, "sum": 162.0, "sum_squared": 26244.0, "min": 162.0, "max": 162.0, "mean": 162.0, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test" }, "count": 1, "sum": 0.7469135802469118, "sum_squared": 0.55787989635726, "min": 0.7469135802469118, "max": 0.7469135802469118, "mean": 0.7469135802469118, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "robustness", "robustness": true, "fairness": false, "computed_on": "worst" } }, "count": 1, "sum": 0.7469135802469118, "sum_squared": 0.55787989635726, "min": 0.7469135802469118, "max": 0.7469135802469118, "mean": 0.7469135802469118, "variance": 0.0, "stddev": 0.0 }, { "name": { "name": "annotation_helpdesk_call_center_summarization_score", "split": "test", "perturbation": { "name": "fairness", "robustness": false, "fairness": true, "computed_on": "worst" } }, "count": 1, "sum": 0.7469135802469118, "sum_squared": 0.55787989635726, "min": 0.7469135802469118, "max": 0.7469135802469118, "mean": 0.7469135802469118, "variance": 0.0, "stddev": 0.0 } ] } ]