lm1-4b2-84b-oscar-repetitions
/
4b284b12boscar
/evaluation
/generation
/slim.4b284b12boscar_GEM-web_nlg_en_PALM_prompt_5.json
{ | |
"results": [ | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"bleu": 1.0960931080747824, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"bleu_stderr": 0.06344656593457922 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_precision": 0.21325470731107665, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_precision_stderr": 0.005803174587983953 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_recall": 0.3950162351216541, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_recall_stderr": 0.004977496708782323 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_fmeasure": 0.22442306072164447, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_fmeasure_stderr": 0.004490024810310396 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_precision": 0.11715823165872309, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_precision_stderr": 0.004000375023616823 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_recall": 0.2112966644324947, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_recall_stderr": 0.003874637116812206 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_fmeasure": 0.11914106590715341, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_fmeasure_stderr": 0.003087212545803013 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_precision": 0.1866801247482292, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_precision_stderr": 0.004986038924327078 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_recall": 0.36529275376302506, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_recall_stderr": 0.0045911093135183 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_fmeasure": 0.19921393354497344, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_fmeasure_stderr": 0.0037728930475513494 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_precision": 0.1931918251374917, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_precision_stderr": 0.00519492169870331 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_recall": 0.3699764207220656, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_recall_stderr": 0.004618932928096925 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_fmeasure": 0.20449901465956502, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_fmeasure_stderr": 0.003922872033805495 | |
} | |
], | |
"config": { | |
"model": "hf-causal", | |
"model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-oscar-repetitions/4b284b12boscar/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 | |
} | |
} |