lm1-4b2-84b-c4seeds
/
4b284b42bc4seed3
/evaluation
/generation
/slim.4b284b42bc4seed3_GEM-web_nlg_en_PALM_prompt_1.json
{ | |
"results": [ | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"bleu": 0.5379028926508614, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"bleu_stderr": 0.03687531363084996 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_precision": 0.07309631370592509, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_precision_stderr": 0.0013732592923049792 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_recall": 0.37028400918910054, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_recall_stderr": 0.005457409473033004 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_fmeasure": 0.11471205984296223, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_fmeasure_stderr": 0.0018805014434852309 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_precision": 0.03377367444978376, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_precision_stderr": 0.0008548764812808871 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_recall": 0.17908798528025902, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_recall_stderr": 0.0037028704616034083 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_fmeasure": 0.0530528813586448, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_fmeasure_stderr": 0.001199555498269871 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_precision": 0.0688117484703189, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_precision_stderr": 0.001222213532774839 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_recall": 0.3480091072970109, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_recall_stderr": 0.004950245404519997 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_fmeasure": 0.10809733095164265, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_fmeasure_stderr": 0.0016925101881231047 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_precision": 0.0692047577374815, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_precision_stderr": 0.0012923071491164684 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_recall": 0.34853415419811595, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_recall_stderr": 0.004966041933763933 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_fmeasure": 0.10846385734949525, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_fmeasure_stderr": 0.0017609033136198704 | |
} | |
], | |
"config": { | |
"model": "hf-causal", | |
"model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4seeds/4b284b42bc4seed3/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", | |
"task_args": "", | |
"num_fewshot": 1, | |
"batch_size": 16, | |
"device": "cuda", | |
"use_cache": false, | |
"limit": 3000, | |
"bootstrap_iters": 10, | |
"seed": 1234 | |
} | |
} |