lm1-8b7-178b-c4-repetitions / 8b7178b88b /evaluation /generation /slim.8b7178b88b_GEM-web_nlg_en_PALM_prompt_4.json
Muennighoff's picture
Add files
ea88016
raw
history blame
4.04 kB
{
"results": [
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"bleu": 0.8048993474300223,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"bleu_stderr": 0.04157871677547976
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_precision": 0.07952578574480215,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_precision_stderr": 0.0012661942787102602
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_recall": 0.4386081219638241,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_recall_stderr": 0.005677266555057123
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_fmeasure": 0.12696507303445326,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_fmeasure_stderr": 0.0017668948839916938
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_precision": 0.0376951838080061,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_precision_stderr": 0.0007761144319241388
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_recall": 0.22894151277692995,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_recall_stderr": 0.004196410528544762
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_fmeasure": 0.06061578220428707,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_fmeasure_stderr": 0.0011175090259571937
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_precision": 0.07134231184259153,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_precision_stderr": 0.0010918605781200322
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_recall": 0.39486889829999067,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_recall_stderr": 0.004923545353622342
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_fmeasure": 0.11378598207760166,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_fmeasure_stderr": 0.0014923445088795922
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_precision": 0.07509698381413346,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_precision_stderr": 0.0012018963700845971
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_recall": 0.41258265582996795,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_recall_stderr": 0.0051847453818666765
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_fmeasure": 0.1197281993743887,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_fmeasure_stderr": 0.0016590997843029978
}
],
"config": {
"model": "hf-causal",
"model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-8b7-178b-c4-repetitions/8b7178b88b/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
"task_args": "",
"num_fewshot": 4,
"batch_size": 16,
"device": "cuda",
"use_cache": false,
"limit": 3000,
"bootstrap_iters": 10,
"seed": 1234
}
}