lm1-8b7-178b-oscar-repetitions / 8b7178b44b /evaluation /generation /slim.8b7178b44b_GEM-web_nlg_en_PALM_prompt_3.json
Muennighoff's picture
Add files
e989436
raw
history blame
4.03 kB
{
"results": [
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"bleu": 1.0349566025343735,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"bleu_stderr": 0.03887674612290353
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_precision": 0.20438835211537978,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_precision_stderr": 0.005665805208271208
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_recall": 0.39198992418866246,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_recall_stderr": 0.004947498335222262
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge1_fmeasure": 0.2237082219543607,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge1_fmeasure_stderr": 0.004572034500678942
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_precision": 0.11365499071190853,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_precision_stderr": 0.00397252439813096
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_recall": 0.2111302605559281,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_recall_stderr": 0.003926393767113415
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rouge2_fmeasure": 0.1207306381621409,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rouge2_fmeasure_stderr": 0.003262380279344123
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_precision": 0.17756713475764335,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_precision_stderr": 0.004833439612735641
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_recall": 0.3583014136638203,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_recall_stderr": 0.004476979554028604
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeL_fmeasure": 0.196789710556852,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeL_fmeasure_stderr": 0.0038281461566836554
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_precision": 0.18556820599751273,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_precision_stderr": 0.00508629808674153
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_recall": 0.3664900622762122,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_recall_stderr": 0.004561176586485258
},
{
"task_name": "GEM/web_nlg_en",
"prompt_name": "PALM_prompt",
"rougeLsum_fmeasure": 0.20416935580581227,
"dataset_path": "GEM/web_nlg",
"dataset_name": "en",
"subset": null,
"rougeLsum_fmeasure_stderr": 0.004021347054930177
}
],
"config": {
"model": "hf-causal",
"model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-8b7-178b-oscar-repetitions/8b7178b44b/transformers,use_accelerate=True,tokenizer=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/gpt2,dtype=bfloat16",
"task_args": "",
"num_fewshot": 3,
"batch_size": 16,
"device": "cuda",
"use_cache": false,
"limit": 3000,
"bootstrap_iters": 10,
"seed": 1234
}
}