lm1-8b7-178b-oscar-repetitions
/
8b7178b44b
/evaluation
/generation
/slim.8b7178b44b_GEM-web_nlg_en_PALM_prompt_5.json
{ | |
"results": [ | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"bleu": 1.194760741144373, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"bleu_stderr": 0.09038400286323779 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_precision": 0.22388107728965712, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_precision_stderr": 0.0058375878448932725 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_recall": 0.41258053843820386, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_recall_stderr": 0.004808891702805189 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_fmeasure": 0.24067238350377898, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_fmeasure_stderr": 0.004627635578763758 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_precision": 0.12470779957706617, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_precision_stderr": 0.004011608962377894 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_recall": 0.22396081182592556, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_recall_stderr": 0.003932938864761309 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_fmeasure": 0.12988590207293185, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_fmeasure_stderr": 0.003239297680717399 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_precision": 0.19195297464964095, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_precision_stderr": 0.004914089701123312 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_recall": 0.37374160359881364, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_recall_stderr": 0.00433847098671097 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_fmeasure": 0.20930811896578502, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_fmeasure_stderr": 0.0038259333936250185 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_precision": 0.20275170487892208, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_precision_stderr": 0.005253109590732199 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_recall": 0.3844755575959121, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_recall_stderr": 0.004448663215506157 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_fmeasure": 0.21923266732565286, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_fmeasure_stderr": 0.004087600756703223 | |
} | |
], | |
"config": { | |
"model": "hf-causal", | |
"model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-8b7-178b-oscar-repetitions/8b7178b44b/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", | |
"task_args": "", | |
"num_fewshot": 5, | |
"batch_size": 16, | |
"device": "cuda", | |
"use_cache": false, | |
"limit": 3000, | |
"bootstrap_iters": 10, | |
"seed": 1234 | |
} | |
} |