lm1-8b7-178b-c4-repetitions / 8b7178b88b /evaluation /generation /slim.8b7178b88b_GEM-web_nlg_en_PALM_prompt_1.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": 0.5634646538427781,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"bleu_stderr": 0.03736988855205685
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_precision": 0.07697654849580207,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_precision_stderr": 0.0012951272613183852
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_recall": 0.3969061719233893,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_recall_stderr": 0.0056851136357343015
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_fmeasure": 0.12131056010238059,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_fmeasure_stderr": 0.0018129711840276857
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_precision": 0.03642271932626906,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_precision_stderr": 0.0007956739937252943
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_recall": 0.2051665201143765,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_recall_stderr": 0.0041594883413927294
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_fmeasure": 0.05785803759738568,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_fmeasure_stderr": 0.0011523078709876162
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_precision": 0.07267273013759175,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_precision_stderr": 0.0011552399753195406
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_recall": 0.3755810527939102,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_recall_stderr": 0.00528545382584031
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_fmeasure": 0.11462071685790733,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_fmeasure_stderr": 0.001624892567642086
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_precision": 0.07333558102707188,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_precision_stderr": 0.0012195728859027712
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_recall": 0.3757551406945481,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_recall_stderr": 0.005219245200343907
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_fmeasure": 0.11540379904462905,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_fmeasure_stderr": 0.0016960835886338564
}
],
"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": 1,
"batch_size": 16,
"device": "cuda",
"use_cache": false,
"limit": 3000,
"bootstrap_iters": 10,
"seed": 1234
}
}