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/xnli/ar/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ar",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.5040160642570282
},
"arguments": "Namespace(config_name=None, dataset_config_name='ar', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ar/MNLI_crowdsource/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ar",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.39879518072289155
},
"arguments": "Namespace(config_name=None, dataset_config_name='ar', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ar/can_we_infer/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ar",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.506425702811245
},
"arguments": "Namespace(config_name=None, dataset_config_name='ar', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ar/guaranteed_possible_impossible/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ar",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.4799196787148594
},
"arguments": "Namespace(config_name=None, dataset_config_name='ar', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ar/justified_in_saying/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ar",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.41526104417670684
},
"arguments": "Namespace(config_name=None, dataset_config_name='ar', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/en/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "en",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.5590361445783133
},
"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/en/MNLI_crowdsource/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "en",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.342570281124498
},
"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/en/can_we_infer/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "en",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.5449799196787148
},
"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/en/guaranteed_possible_impossible/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "en",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.41164658634538154
},
"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/en/justified_in_saying/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "en",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.4634538152610442
},
"arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/es/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "es",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.5373493975903615
},
"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/es/MNLI_crowdsource/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "es",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.40441767068273093
},
"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/es/can_we_infer/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "es",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.5277108433734939
},
"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/es/guaranteed_possible_impossible/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "es",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.44216867469879517
},
"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/es/justified_in_saying/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "es",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.4534136546184739
},
"arguments": "Namespace(config_name=None, dataset_config_name='es', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/fr/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "fr",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.5248995983935743
},
"arguments": "Namespace(config_name=None, dataset_config_name='fr', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/fr/MNLI_crowdsource/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "fr",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.3895582329317269
},
"arguments": "Namespace(config_name=None, dataset_config_name='fr', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/fr/can_we_infer/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "fr",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.5337349397590362
},
"arguments": "Namespace(config_name=None, dataset_config_name='fr', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/fr/guaranteed_possible_impossible/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "fr",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.42971887550200805
},
"arguments": "Namespace(config_name=None, dataset_config_name='fr', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/fr/justified_in_saying/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "fr",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.4738955823293173
},
"arguments": "Namespace(config_name=None, dataset_config_name='fr', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/hi/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "hi",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.4983935742971888
},
"arguments": "Namespace(config_name=None, dataset_config_name='hi', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/hi/MNLI_crowdsource/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "hi",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.38714859437751004
},
"arguments": "Namespace(config_name=None, dataset_config_name='hi', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/hi/can_we_infer/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "hi",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.45542168674698796
},
"arguments": "Namespace(config_name=None, dataset_config_name='hi', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/hi/guaranteed_possible_impossible/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "hi",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.41405622489959837
},
"arguments": "Namespace(config_name=None, dataset_config_name='hi', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/hi/justified_in_saying/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "hi",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.38795180722891565
},
"arguments": "Namespace(config_name=None, dataset_config_name='hi', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/sw/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "sw",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.43493975903614457
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "sw",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.363855421686747
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "sw",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.42891566265060244
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "sw",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.3457831325301205
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "sw",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.3650602409638554
},
"arguments": "Namespace(config_name=None, dataset_config_name='sw', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ur/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ur",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.43493975903614457
},
"arguments": "Namespace(config_name=None, dataset_config_name='ur', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ur/MNLI_crowdsource/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ur",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.3895582329317269
},
"arguments": "Namespace(config_name=None, dataset_config_name='ur', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ur/can_we_infer/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ur",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.45180722891566266
},
"arguments": "Namespace(config_name=None, dataset_config_name='ur', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ur/guaranteed_possible_impossible/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ur",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.40120481927710844
},
"arguments": "Namespace(config_name=None, dataset_config_name='ur', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/ur/justified_in_saying/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "ur",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.37630522088353413
},
"arguments": "Namespace(config_name=None, dataset_config_name='ur', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/vi/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "vi",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.5196787148594377
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "vi",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.38112449799196785
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "vi",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.5080321285140562
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "vi",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.38393574297188754
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "vi",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.43614457831325304
},
"arguments": "Namespace(config_name=None, dataset_config_name='vi', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}

9
evaluation_bloomz-7b1-p3/evaluation_l1/xnli/zh/GPT-3_style/results.json Normal file
View File

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "zh",
"template_name": "GPT-3 style",
"evaluation": {
"accuracy": 0.5052208835341365
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xnli', 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='GPT-3 style', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "zh",
"template_name": "MNLI crowdsource",
"evaluation": {
"accuracy": 0.4
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xnli', 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='MNLI crowdsource', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "zh",
"template_name": "can we infer",
"evaluation": {
"accuracy": 0.5228915662650603
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xnli', 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='can we infer', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "zh",
"template_name": "guaranteed/possible/impossible",
"evaluation": {
"accuracy": 0.4738955823293173
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xnli', 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='guaranteed/possible/impossible', tokenizer_name=None, use_slow_tokenizer=False)"
}

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

@@ -0,0 +1,9 @@
{
"dataset_name": "xnli",
"dataset_config_name": "zh",
"template_name": "justified in saying",
"evaluation": {
"accuracy": 0.45863453815261046
},
"arguments": "Namespace(config_name=None, dataset_config_name='zh', dataset_name='xnli', 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='justified in saying', tokenizer_name=None, use_slow_tokenizer=False)"
}