|
{ |
|
"results": { |
|
"b4b": { |
|
"acc,none": 0.7444512946979038, |
|
"acc_stderr,none": 0.04274747119698657, |
|
"acc_norm,none": 0.7444512946979038, |
|
"acc_norm_stderr,none": 0.04274747119698657, |
|
"alias": "b4b" |
|
}, |
|
"b4bqa": { |
|
"acc,none": 0.7516741071428571, |
|
"acc_stderr,none": 0.010208877794084196, |
|
"acc_norm,none": 0.7516741071428571, |
|
"acc_norm_stderr,none": 0.010208877794084196, |
|
"alias": " - b4bqa" |
|
}, |
|
"medmcqa_g2b": { |
|
"acc,none": 0.6408045977011494, |
|
"acc_stderr,none": 0.025755112822545917, |
|
"acc_norm,none": 0.6408045977011494, |
|
"acc_norm_stderr,none": 0.025755112822545917, |
|
"alias": " - medmcqa_g2b" |
|
}, |
|
"medmcqa_orig_filtered": { |
|
"acc,none": 0.7701149425287356, |
|
"acc_stderr,none": 0.022587512669518847, |
|
"acc_norm,none": 0.7701149425287356, |
|
"acc_norm_stderr,none": 0.022587512669518847, |
|
"alias": " - medmcqa_orig_filtered" |
|
}, |
|
"medqa_4options_g2b": { |
|
"acc,none": 0.7063492063492064, |
|
"acc_stderr,none": 0.023456037383982033, |
|
"acc_norm,none": 0.7063492063492064, |
|
"acc_norm_stderr,none": 0.023456037383982033, |
|
"alias": " - medqa_4options_g2b" |
|
}, |
|
"medqa_4options_orig_filtered": { |
|
"acc,none": 0.8201058201058201, |
|
"acc_stderr,none": 0.01978211983276641, |
|
"acc_norm,none": 0.8201058201058201, |
|
"acc_norm_stderr,none": 0.01978211983276641, |
|
"alias": " - medqa_4options_orig_filtered" |
|
} |
|
}, |
|
"groups": { |
|
"b4b": { |
|
"acc,none": 0.7444512946979038, |
|
"acc_stderr,none": 0.04274747119698657, |
|
"acc_norm,none": 0.7444512946979038, |
|
"acc_norm_stderr,none": 0.04274747119698657, |
|
"alias": "b4b" |
|
} |
|
}, |
|
"configs": { |
|
"b4bqa": { |
|
"task": "b4bqa", |
|
"dataset_path": "AIM-Harvard/b4b_drug_qa", |
|
"test_split": "test", |
|
"doc_to_text": "<function process_cd at 0x78639f2e7040>", |
|
"doc_to_target": "correct_choice", |
|
"doc_to_choice": [ |
|
"A", |
|
"B", |
|
"C", |
|
"D" |
|
], |
|
"description": "", |
|
"target_delimiter": " ", |
|
"fewshot_delimiter": "\n\n", |
|
"metric_list": [ |
|
{ |
|
"metric": "acc", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
}, |
|
{ |
|
"metric": "acc_norm", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
} |
|
], |
|
"output_type": "multiple_choice", |
|
"repeats": 1, |
|
"should_decontaminate": false |
|
}, |
|
"medmcqa_g2b": { |
|
"task": "medmcqa_g2b", |
|
"dataset_path": "AIM-Harvard/medmcqa_generic_to_brand", |
|
"training_split": "train", |
|
"validation_split": "validation", |
|
"test_split": "validation", |
|
"doc_to_text": "<function doc_to_text at 0x78639ef36280>", |
|
"doc_to_target": "cop", |
|
"doc_to_choice": [ |
|
"A", |
|
"B", |
|
"C", |
|
"D" |
|
], |
|
"description": "", |
|
"target_delimiter": " ", |
|
"fewshot_delimiter": "\n\n", |
|
"metric_list": [ |
|
{ |
|
"metric": "acc", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
}, |
|
{ |
|
"metric": "acc_norm", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
} |
|
], |
|
"output_type": "multiple_choice", |
|
"repeats": 1, |
|
"should_decontaminate": true, |
|
"doc_to_decontamination_query": "{{question}}" |
|
}, |
|
"medmcqa_orig_filtered": { |
|
"task": "medmcqa_orig_filtered", |
|
"dataset_path": "AIM-Harvard/medmcqa_original", |
|
"training_split": "train", |
|
"validation_split": "validation", |
|
"test_split": "validation", |
|
"doc_to_text": "<function doc_to_text at 0x78639f2d9e50>", |
|
"doc_to_target": "cop", |
|
"doc_to_choice": [ |
|
"A", |
|
"B", |
|
"C", |
|
"D" |
|
], |
|
"description": "", |
|
"target_delimiter": " ", |
|
"fewshot_delimiter": "\n\n", |
|
"metric_list": [ |
|
{ |
|
"metric": "acc", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
}, |
|
{ |
|
"metric": "acc_norm", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
} |
|
], |
|
"output_type": "multiple_choice", |
|
"repeats": 1, |
|
"should_decontaminate": true, |
|
"doc_to_decontamination_query": "{{question}}" |
|
}, |
|
"medqa_4options_g2b": { |
|
"task": "medqa_4options_g2b", |
|
"dataset_path": "AIM-Harvard/gbaker_medqa_usmle_4_options_hf_generic_to_brand", |
|
"training_split": "train", |
|
"validation_split": "validation", |
|
"test_split": "test", |
|
"doc_to_text": "<function doc_to_text at 0x78639f24a0d0>", |
|
"doc_to_target": "<function doc_to_target at 0x78639f24a550>", |
|
"doc_to_choice": [ |
|
"A", |
|
"B", |
|
"C", |
|
"D" |
|
], |
|
"description": "", |
|
"target_delimiter": " ", |
|
"fewshot_delimiter": "\n\n", |
|
"metric_list": [ |
|
{ |
|
"metric": "acc", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
}, |
|
{ |
|
"metric": "acc_norm", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
} |
|
], |
|
"output_type": "multiple_choice", |
|
"repeats": 1, |
|
"should_decontaminate": false |
|
}, |
|
"medqa_4options_orig_filtered": { |
|
"task": "medqa_4options_orig_filtered", |
|
"dataset_path": "AIM-Harvard/gbaker_medqa_usmle_4_options_hf_original", |
|
"training_split": "train", |
|
"validation_split": "validation", |
|
"test_split": "test", |
|
"doc_to_text": "<function doc_to_text at 0x78639f24a670>", |
|
"doc_to_target": "<function doc_to_target at 0x78639f24a8b0>", |
|
"doc_to_choice": [ |
|
"A", |
|
"B", |
|
"C", |
|
"D" |
|
], |
|
"description": "", |
|
"target_delimiter": " ", |
|
"fewshot_delimiter": "\n\n", |
|
"metric_list": [ |
|
{ |
|
"metric": "acc", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
}, |
|
{ |
|
"metric": "acc_norm", |
|
"aggregation": "mean", |
|
"higher_is_better": true |
|
} |
|
], |
|
"output_type": "multiple_choice", |
|
"repeats": 1, |
|
"should_decontaminate": false |
|
} |
|
}, |
|
"versions": { |
|
"b4b": "N/A", |
|
"b4bqa": "Yaml", |
|
"medmcqa_g2b": "Yaml", |
|
"medmcqa_orig_filtered": "Yaml", |
|
"medqa_4options_g2b": "Yaml", |
|
"medqa_4options_orig_filtered": "Yaml" |
|
}, |
|
"n-shot": { |
|
"b4b": 0, |
|
"b4bqa": 0, |
|
"medmcqa_g2b": 0, |
|
"medmcqa_orig_filtered": 0, |
|
"medqa_4options_g2b": 0, |
|
"medqa_4options_orig_filtered": 0 |
|
}, |
|
"config": { |
|
"model": "hf", |
|
"model_args": "pretrained=johnsnowlabs/JSL-MedLlama-3-8B-v9,parallelize=True,load_in_4bit=True", |
|
"batch_size": "auto", |
|
"batch_sizes": [ |
|
8 |
|
], |
|
"device": null, |
|
"use_cache": null, |
|
"limit": null, |
|
"bootstrap_iters": 100000, |
|
"gen_kwargs": null |
|
}, |
|
"git_hash": "a6ca0b90" |
|
} |