bigscience/bloomz-7b1-p3
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: bigscience/bloomz-7b1-p3
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-06-15 07:40:14 +08:00
commit 78a6661ff1
634 changed files with 7477 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/id/C1_or_C2?_premise/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "id",
"template_name": "C1 or C2? premise, so/because\u2026",
"evaluation": {
"accuracy": 0.47
},
"arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='so/because\u2026,validation', target_max_length=256, template_config_name='en', template_name='C1 or C2? premise', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/id/best_option/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "id",
"template_name": "best_option",
"evaluation": {
"accuracy": 0.51
},
"arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='best_option', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/id/cause_effect/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "id",
"template_name": "cause_effect",
"evaluation": {
"accuracy": 0.65
},
"arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='cause_effect', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/id/i_am_hesitating/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "id",
"template_name": "i_am_hesitating",
"evaluation": {
"accuracy": 0.66
},
"arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='i_am_hesitating', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/id/plausible_alternatives/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "id",
"template_name": "plausible_alternatives",
"evaluation": {
"accuracy": 0.67
},
"arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='plausible_alternatives', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/sw/C1_or_C2?_premise/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "sw",
"template_name": "C1 or C2? premise, so/because\u2026",
"evaluation": {
"accuracy": 0.58
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='so/because\u2026,validation', target_max_length=256, template_config_name='en', template_name='C1 or C2? premise', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/sw/best_option/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "sw",
"template_name": "best_option",
"evaluation": {
"accuracy": 0.57
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='best_option', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/sw/cause_effect/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "sw",
"template_name": "cause_effect",
"evaluation": {
"accuracy": 0.46
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='cause_effect', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/sw/i_am_hesitating/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "sw",
"template_name": "i_am_hesitating",
"evaluation": {
"accuracy": 0.48
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='i_am_hesitating', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/sw/plausible_alternatives/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "sw",
"template_name": "plausible_alternatives",
"evaluation": {
"accuracy": 0.45
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='plausible_alternatives', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/ta/C1_or_C2?_premise/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "ta",
"template_name": "C1 or C2? premise, so/because\u2026",
"evaluation": {
"accuracy": 0.57
},
"arguments": "Namespace(config_name=None, dataset_config_name='ta', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='so/because\u2026,validation', target_max_length=256, template_config_name='en', template_name='C1 or C2? premise', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/ta/best_option/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "ta",
"template_name": "best_option",
"evaluation": {
"accuracy": 0.67
},
"arguments": "Namespace(config_name=None, dataset_config_name='ta', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='best_option', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/ta/cause_effect/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "ta",
"template_name": "cause_effect",
"evaluation": {
"accuracy": 0.71
},
"arguments": "Namespace(config_name=None, dataset_config_name='ta', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='cause_effect', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/ta/i_am_hesitating/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "ta",
"template_name": "i_am_hesitating",
"evaluation": {
"accuracy": 0.71
},
"arguments": "Namespace(config_name=None, dataset_config_name='ta', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='i_am_hesitating', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/ta/plausible_alternatives/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "ta",
"template_name": "plausible_alternatives",
"evaluation": {
"accuracy": 0.69
},
"arguments": "Namespace(config_name=None, dataset_config_name='ta', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='plausible_alternatives', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/vi/C1_or_C2?_premise/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "vi",
"template_name": "C1 or C2? premise, so/because\u2026",
"evaluation": {
"accuracy": 0.55
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='so/because\u2026,validation', target_max_length=256, template_config_name='en', template_name='C1 or C2? premise', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/vi/best_option/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "vi",
"template_name": "best_option",
"evaluation": {
"accuracy": 0.61
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='best_option', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/vi/cause_effect/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "vi",
"template_name": "cause_effect",
"evaluation": {
"accuracy": 0.67
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='cause_effect', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/vi/i_am_hesitating/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "vi",
"template_name": "i_am_hesitating",
"evaluation": {
"accuracy": 0.66
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='i_am_hesitating', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/vi/plausible_alternatives/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "vi",
"template_name": "plausible_alternatives",
"evaluation": {
"accuracy": 0.65
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='plausible_alternatives', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/zh/C1_or_C2?_premise,_so_because…/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "zh",
"template_name": "C1 or C2? premise, so/because\u2026",
"evaluation": {
"accuracy": 0.62
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='C1 or C2? premise, so/because\u2026', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/zh/C1_or_C2?_premise/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "zh",
"template_name": "C1 or C2? premise, so/because\u2026",
"evaluation": {
"accuracy": 0.62
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='so/because\u2026,validation', target_max_length=256, template_config_name='en', template_name='C1 or C2? premise', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/zh/best_option/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "zh",
"template_name": "best_option",
"evaluation": {
"accuracy": 0.61
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='best_option', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/zh/cause_effect/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "zh",
"template_name": "cause_effect",
"evaluation": {
"accuracy": 0.77
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='cause_effect', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/zh/i_am_hesitating/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "zh",
"template_name": "i_am_hesitating",
"evaluation": {
"accuracy": 0.72
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='i_am_hesitating', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xcopa/zh/plausible_alternatives/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xcopa",
"dataset_config_name": "zh",
"template_name": "plausible_alternatives",
"evaluation": {
"accuracy": 0.74
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000', output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/6b3t0/p31lossseqglobal_step1000/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='en', template_name='plausible_alternatives', tokenizer_name=None, use_slow_tokenizer=False)"
}