Add eval files

evaluation_l1/Muennighoff_xstory_cloze/eu/Choose_Story_Ending/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "eu",
+  "template_name": "Choose Story Ending",
+  "evaluation": {
+    "accuracy": 0.49636002647253474
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='eu', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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='Choose Story Ending', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xstory_cloze/hi/Answer_Given_options/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "hi",
+  "template_name": "Answer Given options",
+  "evaluation": {
+    "accuracy": 0.4990072799470549
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='hi', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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='Answer Given options', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xstory_cloze/hi/Story_Continuation_and_Options/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "hi",
+  "template_name": "Story Continuation and Options",
+  "evaluation": {
+    "accuracy": 0.5347452018530774
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='hi', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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='Story Continuation and Options', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xstory_cloze/id/Answer_Given_options/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "id",
+  "template_name": "Answer Given options",
+  "evaluation": {
+    "accuracy": 0.5115817339510258
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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='Answer Given options', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xstory_cloze/id/Choose_Story_Ending/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "id",
+  "template_name": "Choose Story Ending",
+  "evaluation": {
+    "accuracy": 0.5724685638649901
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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='Choose Story Ending', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xstory_cloze/id/Generate_Ending/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "id",
+  "template_name": "Generate Ending",
+  "evaluation": {
+    "accuracy": 0.5440105890138981
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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='Generate Ending', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xstory_cloze/id/Novel_Correct_Ending/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "id",
+  "template_name": "Novel Correct Ending",
+  "evaluation": {
+    "accuracy": 0.5579086697551291
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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='Novel Correct Ending', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xstory_cloze/id/Story_Continuation_and_Options/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "id",
+  "template_name": "Story Continuation and Options",
+  "evaluation": {
+    "accuracy": 0.57114493712773
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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='Story Continuation and Options', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xwinograd/en/Replace/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xwinograd",
+  "dataset_config_name": "en",
+  "template_name": "Replace",
+  "evaluation": {
+    "accuracy": 0.5066666666666667
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='Muennighoff/xwinograd', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='test', target_max_length=256, template_config_name='en', template_name='Replace', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xwinograd/en/stand_for/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xwinograd",
+  "dataset_config_name": "en",
+  "template_name": "stand for",
+  "evaluation": {
+    "accuracy": 0.4890322580645161
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='Muennighoff/xwinograd', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='test', target_max_length=256, template_config_name='en', template_name='stand for', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/Muennighoff_xwinograd/en/underscore_refer_to/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xwinograd",
+  "dataset_config_name": "en",
+  "template_name": "underscore refer to",
+  "evaluation": {
+    "accuracy": 0.5101075268817205
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='en', dataset_name='Muennighoff/xwinograd', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='test', target_max_length=256, template_config_name='en', template_name='underscore refer to', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_l1/xcopa/zh/C1_or_C2?_premise/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "zh",
+  "template_name": "C1 or C2? premise, so/because\u2026",
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l1/xcopa/zh/cause_effect/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "zh",
+  "template_name": "cause_effect",
+  "evaluation": {
+    "accuracy": 0.54
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l1/xcopa/zh/i_am_hesitating/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "zh",
+  "template_name": "i_am_hesitating",
+  "evaluation": {
+    "accuracy": 0.5
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l1/xcopa/zh/plausible_alternatives/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "zh",
+  "template_name": "plausible_alternatives",
+  "evaluation": {
+    "accuracy": 0.55
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l1/xnli/fr/MNLI_crowdsource/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "fr",
+  "template_name": "MNLI crowdsource",
+  "evaluation": {
+    "accuracy": 0.3333333333333333
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l1/xnli/fr/guaranteed_possible_impossible/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "fr",
+  "template_name": "guaranteed/possible/impossible",
+  "evaluation": {
+    "accuracy": 0.3337349397590361
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xcopa/tr/cause_effect/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "tr",
+  "template_name": "cause_effect",
+  "evaluation": {
+    "accuracy": 0.52
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='tr', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xcopa/tr/plausible_alternatives/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "tr",
+  "template_name": "plausible_alternatives",
+  "evaluation": {
+    "accuracy": 0.57
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='tr', dataset_name='xcopa', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xnli/ru/GPT-3_style/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "ru",
+  "template_name": "GPT-3 style",
+  "evaluation": {
+    "accuracy": 0.36987951807228914
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='ru', dataset_name='xnli', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xnli/ru/can_we_infer/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "ru",
+  "template_name": "can we infer",
+  "evaluation": {
+    "accuracy": 0.3963855421686747
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='ru', dataset_name='xnli', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xnli/ru/justified_in_saying/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "ru",
+  "template_name": "justified in saying",
+  "evaluation": {
+    "accuracy": 0.39076305220883534
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='ru', dataset_name='xnli', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xnli/th/GPT-3_style/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "th",
+  "template_name": "GPT-3 style",
+  "evaluation": {
+    "accuracy": 0.3413654618473896
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='th', dataset_name='xnli', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xnli/th/MNLI_crowdsource/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "th",
+  "template_name": "MNLI crowdsource",
+  "evaluation": {
+    "accuracy": 0.3333333333333333
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='th', dataset_name='xnli', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xnli/th/can_we_infer/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "th",
+  "template_name": "can we infer",
+  "evaluation": {
+    "accuracy": 0.3437751004016064
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='th', dataset_name='xnli', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xnli/th/guaranteed_possible_impossible/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "th",
+  "template_name": "guaranteed/possible/impossible",
+  "evaluation": {
+    "accuracy": 0.3333333333333333
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='th', dataset_name='xnli', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_l2/xnli/th/justified_in_saying/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "th",
+  "template_name": "justified in saying",
+  "evaluation": {
+    "accuracy": 0.3550200803212851
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='th', dataset_name='xnli', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/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)"
+}

evaluation_xnlimtht/xnli/es/GPT-3_style_esmt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "es",
+  "template_name": "GPT-3 style_esmt",
+  "evaluation": {
+    "accuracy": 0.4321285140562249
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='es', template_name='GPT-3 style_esmt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/es/MNLI_crowdsource_esmt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "es",
+  "template_name": "MNLI crowdsource_esmt",
+  "evaluation": {
+    "accuracy": 0.3329317269076305
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='es', template_name='MNLI crowdsource_esmt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/es/can_we_infer_esmt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "es",
+  "template_name": "can we infer_esmt",
+  "evaluation": {
+    "accuracy": 0.3333333333333333
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='es', template_name='can we infer_esmt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/es/justified_in_saying_esmt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "es",
+  "template_name": "justified in saying_esmt",
+  "evaluation": {
+    "accuracy": 0.3333333333333333
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='es', template_name='justified in saying_esmt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/fr/GPT-3_style_frht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "fr",
+  "template_name": "GPT-3 style_frht",
+  "evaluation": {
+    "accuracy": 0.3261044176706827
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='fr', template_name='GPT-3 style_frht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/fr/MNLI_crowdsource_frht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "fr",
+  "template_name": "MNLI crowdsource_frht",
+  "evaluation": {
+    "accuracy": 0.3333333333333333
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='fr', template_name='MNLI crowdsource_frht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/fr/guaranteed_possible_impossible_frht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "fr",
+  "template_name": "guaranteed/possible/impossible_frht",
+  "evaluation": {
+    "accuracy": 0.3333333333333333
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='fr', template_name='guaranteed/possible/impossible_frht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/fr/justified_in_saying_frht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "fr",
+  "template_name": "justified in saying_frht",
+  "evaluation": {
+    "accuracy": 0.4353413654618474
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='fr', template_name='justified in saying_frht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/vi/GPT-3_style_viht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "vi",
+  "template_name": "GPT-3 style_viht",
+  "evaluation": {
+    "accuracy": 0.39196787148594375
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='vi', template_name='GPT-3 style_viht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/vi/MNLI_crowdsource_viht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "vi",
+  "template_name": "MNLI crowdsource_viht",
+  "evaluation": {
+    "accuracy": 0.3534136546184739
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='vi', template_name='MNLI crowdsource_viht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/vi/can_we_infer_viht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "vi",
+  "template_name": "can we infer_viht",
+  "evaluation": {
+    "accuracy": 0.372289156626506
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='vi', template_name='can we infer_viht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/vi/guaranteed_possible_impossible_viht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "vi",
+  "template_name": "guaranteed/possible/impossible_viht",
+  "evaluation": {
+    "accuracy": 0.37309236947791163
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='vi', template_name='guaranteed/possible/impossible_viht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/vi/guaranteed_possible_impossible_vimt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "vi",
+  "template_name": "guaranteed/possible/impossible_vimt",
+  "evaluation": {
+    "accuracy": 0.3309236947791165
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='vi', template_name='guaranteed/possible/impossible_vimt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xnlimtht/xnli/vi/justified_in_saying_viht/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "vi",
+  "template_name": "justified in saying_viht",
+  "evaluation": {
+    "accuracy": 0.3650602409638554
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='vi', template_name='justified in saying_viht', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xwinostorycopamt/Muennighoff_xstory_cloze/eu/Answer_Given_options_eumt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "eu",
+  "template_name": "Answer Given options_eumt",
+  "evaluation": {
+    "accuracy": 0.4440767703507611
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='eu', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='eu', template_name='Answer Given options_eumt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xwinostorycopamt/Muennighoff_xstory_cloze/id/Answer_Given_options_idmt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "id",
+  "template_name": "Answer Given options_idmt",
+  "evaluation": {
+    "accuracy": 0.4831237590999338
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='id', template_name='Answer Given options_idmt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xwinostorycopamt/Muennighoff_xstory_cloze/id/Novel_Correct_Ending_idmt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "id",
+  "template_name": "Novel Correct Ending_idmt",
+  "evaluation": {
+    "accuracy": 0.5274652547981469
+  },
+  "arguments": "Namespace(config_name=None, dataset_config_name='id', dataset_name='Muennighoff/xstory_cloze', debug=False, dtype='float16', max_length=2048, model_name_or_path='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='id', template_name='Novel Correct Ending_idmt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xwinostorycopamt/xcopa/ta/cause_effect_tamt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "ta",
+  "template_name": "cause_effect_tamt",
+  "evaluation": {
+    "accuracy": 0.45
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='ta', template_name='cause_effect_tamt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xwinostorycopamt/xcopa/ta/i_am_hesitating_tamt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "ta",
+  "template_name": "i_am_hesitating_tamt",
+  "evaluation": {
+    "accuracy": 0.42
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='ta', template_name='i_am_hesitating_tamt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xwinostorycopamt/xcopa/ta/plausible_alternatives_tamt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "ta",
+  "template_name": "plausible_alternatives_tamt",
+  "evaluation": {
+    "accuracy": 0.46
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='ta', template_name='plausible_alternatives_tamt', tokenizer_name=None, use_slow_tokenizer=False)"
+}

evaluation_xwinostorycopamt/xcopa/vi/best_option_vimt/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xcopa",
+  "dataset_config_name": "vi",
+  "template_name": "best_option_vimt",
+  "evaluation": {
+    "accuracy": 0.53
+  },
+  "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/350mt0/bloomz-560m', nospace=False, output_dir='/gpfsscratch/rech/six/commun/experiments/muennighoff/bloomckpt/350mt0/bloomz-560m/evaluation', pad_to_max_length=False, per_device_eval_batch_size=8, prefixlm=False, split='validation', target_max_length=256, template_config_name='vi', template_name='best_option_vimt', tokenizer_name=None, use_slow_tokenizer=False)"
+}