lm1-2b8-55b-c4-dedup
/
evaluation
/slim.limited=3000.model=lm1-2b8-55b-nodups.task=e2e_nlg_cleaned.templates=generate_text_restaurant.fewshot=1.batchsize=16.seed=1234.timestamp=2023-01-20T22:42:14.json
{ | |
"results": [ | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"bleu": 6.208120134028529, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"bleu_stderr": 0.1090255908153869 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rouge1_precision": 0.31731355835055003, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rouge1_precision_stderr": 0.004099274504817032 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rouge1_recall": 0.3324854292034225, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rouge1_recall_stderr": 0.0036680738174009278 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rouge1_fmeasure": 0.2919082916343828, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rouge1_fmeasure_stderr": 0.003144827416065499 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rouge2_precision": 0.1359171818938097, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rouge2_precision_stderr": 0.0024425949074616932 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rouge2_recall": 0.13681320433598393, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rouge2_recall_stderr": 0.0021042970297826253 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rouge2_fmeasure": 0.12033860836317081, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rouge2_fmeasure_stderr": 0.0018255484904436405 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rougeL_precision": 0.24305871613826696, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rougeL_precision_stderr": 0.00324135869732354 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rougeL_recall": 0.2577033439038558, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rougeL_recall_stderr": 0.0029570712013500815 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rougeL_fmeasure": 0.22283049594155988, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rougeL_fmeasure_stderr": 0.0023764897834232346 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rougeLsum_precision": 0.2640833510596156, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rougeLsum_precision_stderr": 0.00363828067497133 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rougeLsum_recall": 0.273011704389814, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rougeLsum_recall_stderr": 0.0031687493131647084 | |
}, | |
{ | |
"task_name": "e2e_nlg_cleaned", | |
"prompt_name": "generate_text_restaurant", | |
"rougeLsum_fmeasure": 0.2404191146065513, | |
"dataset_path": "e2e_nlg_cleaned", | |
"dataset_name": null, | |
"subset": null, | |
"rougeLsum_fmeasure_stderr": 0.002734721442980669 | |
} | |
], | |
"config": { | |
"model": "hf-causal", | |
"model_args": "pretrained=/pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/lm1-2b8-55b-nodups/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 | |
} | |
} |