lm1-4b2-84b-c4-repetitions / 4b284b12bc4 /evaluation /generation /slim.4b284b12bc4_GEM-web_nlg_en_PALM_prompt_4.json
Muennighoff's picture
Add files
a1243c9
raw
history blame
4.05 kB
{
"results": [
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"bleu": 0.37875018794247045,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"bleu_stderr": 0.024296780304434905
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_precision": 0.07231813177556075,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_precision_stderr": 0.0015118246585830762
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_recall": 0.31870699434574523,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_recall_stderr": 0.0046463072458484975
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_fmeasure": 0.1102139471634063,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_fmeasure_stderr": 0.0019620155943445507
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_precision": 0.033695317164630666,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_precision_stderr": 0.0009163247572691914
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_recall": 0.15554105747469235,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_recall_stderr": 0.003236397171744527
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_fmeasure": 0.0515680827205002,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_fmeasure_stderr": 0.001213141047008391
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_precision": 0.06901574750871842,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_precision_stderr": 0.0013947803448898716
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_recall": 0.30638147232578594,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_recall_stderr": 0.004472401624140904
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_fmeasure": 0.1054595308766645,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_fmeasure_stderr": 0.0018290764447497754
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_precision": 0.06885871876103705,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_precision_stderr": 0.001420315845279487
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_recall": 0.30338100654345734,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_recall_stderr": 0.0043561963135752306
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_fmeasure": 0.10492767242713451,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_fmeasure_stderr": 0.001836832016012516
}
],
"config": {
"model": "hf-causal",
"model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-4b2-84b-c4-repetitions/4b284b12bc4/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
}
}