lm1-2b8-55b-c4-perplexity
/
perplexity25
/evaluation
/generation
/slim.perplexity25_GEM-web_nlg_en_PALM_prompt_1.json
{ | |
"results": [ | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"bleu": 0.3392027660661122, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"bleu_stderr": 0.021974513144963268 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_precision": 0.0649237002813453, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_precision_stderr": 0.0013836117996389116 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_recall": 0.3252507747834447, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_recall_stderr": 0.004830093238320235 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge1_fmeasure": 0.10098636709306053, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge1_fmeasure_stderr": 0.0018586237655962318 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_precision": 0.029412803554314693, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_precision_stderr": 0.000840010407149539 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_recall": 0.1489985792608451, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_recall_stderr": 0.0033169694716508446 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rouge2_fmeasure": 0.0456950711509664, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rouge2_fmeasure_stderr": 0.0011687980868032475 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_precision": 0.06161199072970022, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_precision_stderr": 0.001272442500492367 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_recall": 0.30851984711534297, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_recall_stderr": 0.004479724678989707 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeL_fmeasure": 0.09597600266569703, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeL_fmeasure_stderr": 0.0017278112461683963 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_precision": 0.062043590899161734, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_precision_stderr": 0.0013176307073547546 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_recall": 0.3089691495669342, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_recall_stderr": 0.004492168762956442 | |
}, | |
{ | |
"task_name": "GEM/web_nlg_en", | |
"prompt_name": "PALM_prompt", | |
"rougeLsum_fmeasure": 0.09646257885608286, | |
"dataset_path": "GEM/web_nlg", | |
"dataset_name": "en", | |
"subset": null, | |
"rougeLsum_fmeasure_stderr": 0.0017756689280714777 | |
} | |
], | |
"config": { | |
"model": "hf-causal", | |
"model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-c4-repetitions/perplexity25/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16", | |
"task_args": "", | |
"num_fewshot": 1, | |
"batch_size": 4, | |
"device": "cuda", | |
"use_cache": false, | |
"limit": 3000, | |
"bootstrap_iters": 10, | |
"seed": 1234 | |
} | |
} |