Add

.gitattributes

@@ -32,3 +32,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,31 @@
+{
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "BloomModel"
+  ],
+  "attention_dropout": 0.0,
+  "attention_softmax_in_fp32": true,
+  "bias_dropout_fusion": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_dropout": 0.0,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "masked_softmax_fusion": true,
+  "model_type": "bloom",
+  "n_embed": 4096,
+  "n_inner": null,
+  "n_layer": 30,
+  "num_attention_heads": 32,
+  "offset_alibi": 100,
+  "pad_token_id": 3,
+  "pretraining_tp": 4,
+  "seq_length": 2048,
+  "skip_bias_add": true,
+  "skip_bias_add_qkv": false,
+  "slow_but_exact": false,
+  "transformers_version": "4.21.0.dev0",
+  "unk_token_id": 0,
+  "use_cache": true,
+  "vocab_size": 250880
+}

evaluation/Muennighoff_xstory_cloze/ru/Answer_Given_options/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "ru",
+  "template_name": "Answer Given options",
+  "evaluation": {
+    "accuracy": 0.6340172071475844
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xstory_cloze', dataset_config_name='ru', template_config_name='en', template_name='Answer Given options', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xstory_cloze/ru/Choose_Story_Ending/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "ru",
+  "template_name": "Choose Story Ending",
+  "evaluation": {
+    "accuracy": 0.7908669755129054
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xstory_cloze', dataset_config_name='ru', template_config_name='en', template_name='Choose Story Ending', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xstory_cloze/ru/Generate_Ending/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "ru",
+  "template_name": "Generate Ending",
+  "evaluation": {
+    "accuracy": 0.5532759761747187
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xstory_cloze', dataset_config_name='ru', template_config_name='en', template_name='Generate Ending', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xstory_cloze/ru/Novel_Correct_Ending/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "ru",
+  "template_name": "Novel Correct Ending",
+  "evaluation": {
+    "accuracy": 0.7379219060225016
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xstory_cloze', dataset_config_name='ru', template_config_name='en', template_name='Novel Correct Ending', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xstory_cloze/ru/Story_Continuation_and_Options/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xstory_cloze",
+  "dataset_config_name": "ru",
+  "template_name": "Story Continuation and Options",
+  "evaluation": {
+    "accuracy": 0.7531436135009927
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xstory_cloze', dataset_config_name='ru', template_config_name='en', template_name='Story Continuation and Options', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xwinograd/ru/Replace/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xwinograd",
+  "dataset_config_name": "ru",
+  "template_name": "Replace",
+  "evaluation": {
+    "accuracy": 0.5396825396825397
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xwinograd', dataset_config_name='ru', template_config_name='en', template_name='Replace', split='test', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xwinograd/ru/True_or_False/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xwinograd",
+  "dataset_config_name": "ru",
+  "template_name": "True or False",
+  "evaluation": {
+    "accuracy": 0.46984126984126984
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xwinograd', dataset_config_name='ru', template_config_name='en', template_name='True or False', split='test', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xwinograd/ru/does_underscore_refer_to/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xwinograd",
+  "dataset_config_name": "ru",
+  "template_name": "does underscore refer to",
+  "evaluation": {
+    "accuracy": 0.5079365079365079
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xwinograd', dataset_config_name='ru', template_config_name='en', template_name='does underscore refer to', split='test', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xwinograd/ru/stand_for/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xwinograd",
+  "dataset_config_name": "ru",
+  "template_name": "stand for",
+  "evaluation": {
+    "accuracy": 0.4888888888888889
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xwinograd', dataset_config_name='ru', template_config_name='en', template_name='stand for', split='test', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/Muennighoff_xwinograd/ru/underscore_refer_to/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "Muennighoff/xwinograd",
+  "dataset_config_name": "ru",
+  "template_name": "underscore refer to",
+  "evaluation": {
+    "accuracy": 0.5079365079365079
+  },
+  "arguments": "Namespace(dataset_name='Muennighoff/xwinograd', dataset_config_name='ru', template_config_name='en', template_name='underscore refer to', split='test', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/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": 0.5
+  },
+  "arguments": "Namespace(dataset_name='xnli', dataset_config_name='ru', template_config_name='en', template_name='GPT-3 style', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/xnli/ru/MNLI_crowdsource/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "ru",
+  "template_name": "MNLI crowdsource",
+  "evaluation": {
+    "accuracy": 0.38473895582329315
+  },
+  "arguments": "Namespace(dataset_name='xnli', dataset_config_name='ru', template_config_name='en', template_name='MNLI crowdsource', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/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": 0.4971887550200803
+  },
+  "arguments": "Namespace(dataset_name='xnli', dataset_config_name='ru', template_config_name='en', template_name='can we infer', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/xnli/ru/guaranteed_possible_impossible/results.json

@@ -0,0 +1,9 @@
+{
+  "dataset_name": "xnli",
+  "dataset_config_name": "ru",
+  "template_name": "guaranteed/possible/impossible",
+  "evaluation": {
+    "accuracy": 0.36626506024096384
+  },
+  "arguments": "Namespace(dataset_name='xnli', dataset_config_name='ru', template_config_name='en', template_name='guaranteed/possible/impossible', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

evaluation/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": 0.46265060240963857
+  },
+  "arguments": "Namespace(dataset_name='xnli', dataset_config_name='ru', template_config_name='en', template_name='justified in saying', split='validation', max_length=2048, target_max_length=256, pad_to_max_length=False, model_name_or_path='bloomz-7b1-xp3ru', config_name=None, tokenizer_name=None, use_slow_tokenizer=False, per_device_eval_batch_size=4, output_dir='bloomz-7b1-xp3ru/evaluation', debug=False, prefixlm=False, dtype='float16', nospace=False, scratchpad=False)"
+}

pytorch_model.bin.index.json

@@ -0,0 +1,372 @@
+{
+  "metadata": {
+    "total_size": 14138032128
+  },
+  "weight_map": {
+    "h.0.input_layernorm.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.input_layernorm.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.mlp.dense_4h_to_h.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.mlp.dense_4h_to_h.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.mlp.dense_h_to_4h.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.mlp.dense_h_to_4h.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.post_attention_layernorm.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.post_attention_layernorm.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.self_attention.dense.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.self_attention.dense.weight": "pytorch_model_00002-of-00032.bin",
+    "h.0.self_attention.query_key_value.bias": "pytorch_model_00002-of-00032.bin",
+    "h.0.self_attention.query_key_value.weight": "pytorch_model_00002-of-00032.bin",
+    "h.1.input_layernorm.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.input_layernorm.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.mlp.dense_4h_to_h.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.mlp.dense_4h_to_h.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.mlp.dense_h_to_4h.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.mlp.dense_h_to_4h.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.post_attention_layernorm.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.post_attention_layernorm.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.self_attention.dense.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.self_attention.dense.weight": "pytorch_model_00003-of-00032.bin",
+    "h.1.self_attention.query_key_value.bias": "pytorch_model_00003-of-00032.bin",
+    "h.1.self_attention.query_key_value.weight": "pytorch_model_00003-of-00032.bin",
+    "h.10.input_layernorm.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.input_layernorm.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.mlp.dense_4h_to_h.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.mlp.dense_4h_to_h.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.mlp.dense_h_to_4h.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.mlp.dense_h_to_4h.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.post_attention_layernorm.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.post_attention_layernorm.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.self_attention.dense.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.self_attention.dense.weight": "pytorch_model_00012-of-00032.bin",
+    "h.10.self_attention.query_key_value.bias": "pytorch_model_00012-of-00032.bin",
+    "h.10.self_attention.query_key_value.weight": "pytorch_model_00012-of-00032.bin",
+    "h.11.input_layernorm.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.input_layernorm.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.mlp.dense_4h_to_h.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.mlp.dense_4h_to_h.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.mlp.dense_h_to_4h.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.mlp.dense_h_to_4h.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.post_attention_layernorm.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.post_attention_layernorm.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.self_attention.dense.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.self_attention.dense.weight": "pytorch_model_00013-of-00032.bin",
+    "h.11.self_attention.query_key_value.bias": "pytorch_model_00013-of-00032.bin",
+    "h.11.self_attention.query_key_value.weight": "pytorch_model_00013-of-00032.bin",
+    "h.12.input_layernorm.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.input_layernorm.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.mlp.dense_4h_to_h.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.mlp.dense_4h_to_h.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.mlp.dense_h_to_4h.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.mlp.dense_h_to_4h.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.post_attention_layernorm.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.post_attention_layernorm.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.self_attention.dense.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.self_attention.dense.weight": "pytorch_model_00014-of-00032.bin",
+    "h.12.self_attention.query_key_value.bias": "pytorch_model_00014-of-00032.bin",
+    "h.12.self_attention.query_key_value.weight": "pytorch_model_00014-of-00032.bin",
+    "h.13.input_layernorm.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.input_layernorm.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.mlp.dense_4h_to_h.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.mlp.dense_4h_to_h.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.mlp.dense_h_to_4h.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.mlp.dense_h_to_4h.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.post_attention_layernorm.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.post_attention_layernorm.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.self_attention.dense.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.self_attention.dense.weight": "pytorch_model_00015-of-00032.bin",
+    "h.13.self_attention.query_key_value.bias": "pytorch_model_00015-of-00032.bin",
+    "h.13.self_attention.query_key_value.weight": "pytorch_model_00015-of-00032.bin",
+    "h.14.input_layernorm.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.input_layernorm.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.mlp.dense_4h_to_h.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.mlp.dense_4h_to_h.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.mlp.dense_h_to_4h.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.mlp.dense_h_to_4h.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.post_attention_layernorm.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.post_attention_layernorm.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.self_attention.dense.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.self_attention.dense.weight": "pytorch_model_00016-of-00032.bin",
+    "h.14.self_attention.query_key_value.bias": "pytorch_model_00016-of-00032.bin",
+    "h.14.self_attention.query_key_value.weight": "pytorch_model_00016-of-00032.bin",
+    "h.15.input_layernorm.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.input_layernorm.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.mlp.dense_4h_to_h.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.mlp.dense_4h_to_h.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.mlp.dense_h_to_4h.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.mlp.dense_h_to_4h.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.post_attention_layernorm.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.post_attention_layernorm.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.self_attention.dense.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.self_attention.dense.weight": "pytorch_model_00017-of-00032.bin",
+    "h.15.self_attention.query_key_value.bias": "pytorch_model_00017-of-00032.bin",
+    "h.15.self_attention.query_key_value.weight": "pytorch_model_00017-of-00032.bin",
+    "h.16.input_layernorm.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.input_layernorm.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.mlp.dense_4h_to_h.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.mlp.dense_4h_to_h.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.mlp.dense_h_to_4h.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.mlp.dense_h_to_4h.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.post_attention_layernorm.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.post_attention_layernorm.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.self_attention.dense.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.self_attention.dense.weight": "pytorch_model_00018-of-00032.bin",
+    "h.16.self_attention.query_key_value.bias": "pytorch_model_00018-of-00032.bin",
+    "h.16.self_attention.query_key_value.weight": "pytorch_model_00018-of-00032.bin",
+    "h.17.input_layernorm.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.input_layernorm.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.mlp.dense_4h_to_h.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.mlp.dense_4h_to_h.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.mlp.dense_h_to_4h.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.mlp.dense_h_to_4h.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.post_attention_layernorm.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.post_attention_layernorm.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.self_attention.dense.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.self_attention.dense.weight": "pytorch_model_00019-of-00032.bin",
+    "h.17.self_attention.query_key_value.bias": "pytorch_model_00019-of-00032.bin",
+    "h.17.self_attention.query_key_value.weight": "pytorch_model_00019-of-00032.bin",
+    "h.18.input_layernorm.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.input_layernorm.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.mlp.dense_4h_to_h.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.mlp.dense_4h_to_h.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.mlp.dense_h_to_4h.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.mlp.dense_h_to_4h.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.post_attention_layernorm.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.post_attention_layernorm.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.self_attention.dense.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.self_attention.dense.weight": "pytorch_model_00020-of-00032.bin",
+    "h.18.self_attention.query_key_value.bias": "pytorch_model_00020-of-00032.bin",
+    "h.18.self_attention.query_key_value.weight": "pytorch_model_00020-of-00032.bin",
+    "h.19.input_layernorm.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.input_layernorm.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.mlp.dense_4h_to_h.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.mlp.dense_4h_to_h.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.mlp.dense_h_to_4h.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.mlp.dense_h_to_4h.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.post_attention_layernorm.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.post_attention_layernorm.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.self_attention.dense.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.self_attention.dense.weight": "pytorch_model_00021-of-00032.bin",
+    "h.19.self_attention.query_key_value.bias": "pytorch_model_00021-of-00032.bin",
+    "h.19.self_attention.query_key_value.weight": "pytorch_model_00021-of-00032.bin",
+    "h.2.input_layernorm.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.input_layernorm.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.mlp.dense_4h_to_h.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.mlp.dense_4h_to_h.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.mlp.dense_h_to_4h.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.mlp.dense_h_to_4h.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.post_attention_layernorm.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.post_attention_layernorm.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.self_attention.dense.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.self_attention.dense.weight": "pytorch_model_00004-of-00032.bin",
+    "h.2.self_attention.query_key_value.bias": "pytorch_model_00004-of-00032.bin",
+    "h.2.self_attention.query_key_value.weight": "pytorch_model_00004-of-00032.bin",
+    "h.20.input_layernorm.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.input_layernorm.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.mlp.dense_4h_to_h.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.mlp.dense_4h_to_h.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.mlp.dense_h_to_4h.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.mlp.dense_h_to_4h.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.post_attention_layernorm.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.post_attention_layernorm.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.self_attention.dense.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.self_attention.dense.weight": "pytorch_model_00022-of-00032.bin",
+    "h.20.self_attention.query_key_value.bias": "pytorch_model_00022-of-00032.bin",
+    "h.20.self_attention.query_key_value.weight": "pytorch_model_00022-of-00032.bin",
+    "h.21.input_layernorm.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.input_layernorm.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.mlp.dense_4h_to_h.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.mlp.dense_4h_to_h.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.mlp.dense_h_to_4h.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.mlp.dense_h_to_4h.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.post_attention_layernorm.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.post_attention_layernorm.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.self_attention.dense.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.self_attention.dense.weight": "pytorch_model_00023-of-00032.bin",
+    "h.21.self_attention.query_key_value.bias": "pytorch_model_00023-of-00032.bin",
+    "h.21.self_attention.query_key_value.weight": "pytorch_model_00023-of-00032.bin",
+    "h.22.input_layernorm.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.input_layernorm.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.mlp.dense_4h_to_h.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.mlp.dense_4h_to_h.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.mlp.dense_h_to_4h.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.mlp.dense_h_to_4h.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.post_attention_layernorm.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.post_attention_layernorm.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.self_attention.dense.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.self_attention.dense.weight": "pytorch_model_00024-of-00032.bin",
+    "h.22.self_attention.query_key_value.bias": "pytorch_model_00024-of-00032.bin",
+    "h.22.self_attention.query_key_value.weight": "pytorch_model_00024-of-00032.bin",
+    "h.23.input_layernorm.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.input_layernorm.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.mlp.dense_4h_to_h.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.mlp.dense_4h_to_h.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.mlp.dense_h_to_4h.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.mlp.dense_h_to_4h.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.post_attention_layernorm.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.post_attention_layernorm.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.self_attention.dense.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.self_attention.dense.weight": "pytorch_model_00025-of-00032.bin",
+    "h.23.self_attention.query_key_value.bias": "pytorch_model_00025-of-00032.bin",
+    "h.23.self_attention.query_key_value.weight": "pytorch_model_00025-of-00032.bin",
+    "h.24.input_layernorm.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.input_layernorm.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.mlp.dense_4h_to_h.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.mlp.dense_4h_to_h.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.mlp.dense_h_to_4h.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.mlp.dense_h_to_4h.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.post_attention_layernorm.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.post_attention_layernorm.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.self_attention.dense.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.self_attention.dense.weight": "pytorch_model_00026-of-00032.bin",
+    "h.24.self_attention.query_key_value.bias": "pytorch_model_00026-of-00032.bin",
+    "h.24.self_attention.query_key_value.weight": "pytorch_model_00026-of-00032.bin",
+    "h.25.input_layernorm.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.input_layernorm.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.mlp.dense_4h_to_h.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.mlp.dense_4h_to_h.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.mlp.dense_h_to_4h.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.mlp.dense_h_to_4h.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.post_attention_layernorm.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.post_attention_layernorm.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.self_attention.dense.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.self_attention.dense.weight": "pytorch_model_00027-of-00032.bin",
+    "h.25.self_attention.query_key_value.bias": "pytorch_model_00027-of-00032.bin",
+    "h.25.self_attention.query_key_value.weight": "pytorch_model_00027-of-00032.bin",
+    "h.26.input_layernorm.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.input_layernorm.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.mlp.dense_4h_to_h.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.mlp.dense_4h_to_h.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.mlp.dense_h_to_4h.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.mlp.dense_h_to_4h.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.post_attention_layernorm.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.post_attention_layernorm.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.self_attention.dense.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.self_attention.dense.weight": "pytorch_model_00028-of-00032.bin",
+    "h.26.self_attention.query_key_value.bias": "pytorch_model_00028-of-00032.bin",
+    "h.26.self_attention.query_key_value.weight": "pytorch_model_00028-of-00032.bin",
+    "h.27.input_layernorm.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.input_layernorm.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.mlp.dense_4h_to_h.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.mlp.dense_4h_to_h.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.mlp.dense_h_to_4h.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.mlp.dense_h_to_4h.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.post_attention_layernorm.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.post_attention_layernorm.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.self_attention.dense.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.self_attention.dense.weight": "pytorch_model_00029-of-00032.bin",
+    "h.27.self_attention.query_key_value.bias": "pytorch_model_00029-of-00032.bin",
+    "h.27.self_attention.query_key_value.weight": "pytorch_model_00029-of-00032.bin",
+    "h.28.input_layernorm.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.input_layernorm.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.mlp.dense_4h_to_h.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.mlp.dense_4h_to_h.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.mlp.dense_h_to_4h.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.mlp.dense_h_to_4h.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.post_attention_layernorm.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.post_attention_layernorm.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.self_attention.dense.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.self_attention.dense.weight": "pytorch_model_00030-of-00032.bin",
+    "h.28.self_attention.query_key_value.bias": "pytorch_model_00030-of-00032.bin",
+    "h.28.self_attention.query_key_value.weight": "pytorch_model_00030-of-00032.bin",
+    "h.29.input_layernorm.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.input_layernorm.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.mlp.dense_4h_to_h.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.mlp.dense_4h_to_h.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.mlp.dense_h_to_4h.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.mlp.dense_h_to_4h.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.post_attention_layernorm.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.post_attention_layernorm.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.self_attention.dense.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.self_attention.dense.weight": "pytorch_model_00031-of-00032.bin",
+    "h.29.self_attention.query_key_value.bias": "pytorch_model_00031-of-00032.bin",
+    "h.29.self_attention.query_key_value.weight": "pytorch_model_00031-of-00032.bin",
+    "h.3.input_layernorm.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.input_layernorm.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.mlp.dense_4h_to_h.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.mlp.dense_4h_to_h.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.mlp.dense_h_to_4h.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.mlp.dense_h_to_4h.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.post_attention_layernorm.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.post_attention_layernorm.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.self_attention.dense.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.self_attention.dense.weight": "pytorch_model_00005-of-00032.bin",
+    "h.3.self_attention.query_key_value.bias": "pytorch_model_00005-of-00032.bin",
+    "h.3.self_attention.query_key_value.weight": "pytorch_model_00005-of-00032.bin",
+    "h.4.input_layernorm.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.input_layernorm.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.mlp.dense_4h_to_h.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.mlp.dense_4h_to_h.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.mlp.dense_h_to_4h.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.mlp.dense_h_to_4h.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.post_attention_layernorm.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.post_attention_layernorm.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.self_attention.dense.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.self_attention.dense.weight": "pytorch_model_00006-of-00032.bin",
+    "h.4.self_attention.query_key_value.bias": "pytorch_model_00006-of-00032.bin",
+    "h.4.self_attention.query_key_value.weight": "pytorch_model_00006-of-00032.bin",
+    "h.5.input_layernorm.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.input_layernorm.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.mlp.dense_4h_to_h.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.mlp.dense_4h_to_h.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.mlp.dense_h_to_4h.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.mlp.dense_h_to_4h.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.post_attention_layernorm.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.post_attention_layernorm.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.self_attention.dense.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.self_attention.dense.weight": "pytorch_model_00007-of-00032.bin",
+    "h.5.self_attention.query_key_value.bias": "pytorch_model_00007-of-00032.bin",
+    "h.5.self_attention.query_key_value.weight": "pytorch_model_00007-of-00032.bin",
+    "h.6.input_layernorm.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.input_layernorm.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.mlp.dense_4h_to_h.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.mlp.dense_4h_to_h.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.mlp.dense_h_to_4h.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.mlp.dense_h_to_4h.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.post_attention_layernorm.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.post_attention_layernorm.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.self_attention.dense.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.self_attention.dense.weight": "pytorch_model_00008-of-00032.bin",
+    "h.6.self_attention.query_key_value.bias": "pytorch_model_00008-of-00032.bin",
+    "h.6.self_attention.query_key_value.weight": "pytorch_model_00008-of-00032.bin",
+    "h.7.input_layernorm.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.input_layernorm.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.mlp.dense_4h_to_h.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.mlp.dense_4h_to_h.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.mlp.dense_h_to_4h.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.mlp.dense_h_to_4h.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.post_attention_layernorm.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.post_attention_layernorm.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.self_attention.dense.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.self_attention.dense.weight": "pytorch_model_00009-of-00032.bin",
+    "h.7.self_attention.query_key_value.bias": "pytorch_model_00009-of-00032.bin",
+    "h.7.self_attention.query_key_value.weight": "pytorch_model_00009-of-00032.bin",
+    "h.8.input_layernorm.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.input_layernorm.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.mlp.dense_4h_to_h.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.mlp.dense_4h_to_h.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.mlp.dense_h_to_4h.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.mlp.dense_h_to_4h.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.post_attention_layernorm.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.post_attention_layernorm.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.self_attention.dense.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.self_attention.dense.weight": "pytorch_model_00010-of-00032.bin",
+    "h.8.self_attention.query_key_value.bias": "pytorch_model_00010-of-00032.bin",
+    "h.8.self_attention.query_key_value.weight": "pytorch_model_00010-of-00032.bin",
+    "h.9.input_layernorm.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.input_layernorm.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.mlp.dense_4h_to_h.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.mlp.dense_4h_to_h.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.mlp.dense_h_to_4h.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.mlp.dense_h_to_4h.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.post_attention_layernorm.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.post_attention_layernorm.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.self_attention.dense.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.self_attention.dense.weight": "pytorch_model_00011-of-00032.bin",
+    "h.9.self_attention.query_key_value.bias": "pytorch_model_00011-of-00032.bin",
+    "h.9.self_attention.query_key_value.weight": "pytorch_model_00011-of-00032.bin",
+    "ln_f.bias": "pytorch_model_00032-of-00032.bin",
+    "ln_f.weight": "pytorch_model_00032-of-00032.bin",
+    "word_embeddings.weight": "pytorch_model_00001-of-00032.bin",
+    "word_embeddings_layernorm.bias": "pytorch_model_00001-of-00032.bin",
+    "word_embeddings_layernorm.weight": "pytorch_model_00001-of-00032.bin"
+  }
+}

pytorch_model_00001-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec9cd558cfdf899889b9c728ea76a9a7bdd165a0bb4bb8afd14c7cbe61541916
+size 2055226892

pytorch_model_00002-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:985f515466ed6be3f1e161cf0fa1f8642d5cd644424eb11ff85317205cb5ce76
+size 402763943

pytorch_model_00003-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:177dd699a09592748d6f790ed41a5853dc9d931564ee23eac9e7c408361464a0
+size 402763943

pytorch_model_00004-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0947d5ef3b0e651d1334ea915bc64a20fb7ee1433ee71fedd41c360c4836df8f
+size 402763943

pytorch_model_00005-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d202b48bf2ce7b34d51b81f8ead81dc6c498088c5a3572f7b3204592fb181ee6
+size 402763943

pytorch_model_00006-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d688567d74390a97d5dcdcce05ffb5119d3a42213eb62b9aa8f3de84740968b8
+size 402763943

pytorch_model_00007-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d6133a21d7623916bfb19379724e7a46820f22d03606b6cd7acee947031d30f
+size 402763943

pytorch_model_00008-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85eda9b92234332f05fa854fd64b583fc86f1d9fa328032fd51c1daf123e76d6
+size 402763943

pytorch_model_00009-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d97fee47d223c6bc5ca2ceb2e3279895e8b0fb5f66ac5da5c33b758e73227411
+size 402763943

pytorch_model_00010-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d9b2915a9b66bec28fffffd8458a54aa416d18132cd0fe5e02ca7859b1a7e3d
+size 402763943

pytorch_model_00011-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27d32ef6fb637dafa5a2dca5686f91ac02b81aa1c83b47d07b3400932d287ac7
+size 402763943

pytorch_model_00012-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68ed6492916eeb4a076d34faddfd89cdef4ac8cfec26ccbcfb0ec30b36769ae1
+size 402763943

pytorch_model_00013-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:922904004e745331255d0e4690f14cd53247ae507c1f471a979a14a012d78455
+size 402763943

pytorch_model_00014-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba8938549b29abc07699cc972596f2340f8360973fb2645fb209016c0b9837c2
+size 402763943

pytorch_model_00015-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:043514372ba5b5b3e8e9ba69d5e37916b1ff72131fc87e732bb061a03b88fad5
+size 402763943

pytorch_model_00016-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62d361be621f84851a9a22c6c10e5ce5a97b01d398f21b81bd80c0884179be3c
+size 402763943

pytorch_model_00017-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ae3e126aace7ff7b7ab103f2378a913fdbfc47c6222bceb32e018691c358c09
+size 402763943

pytorch_model_00018-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef181480d086f98fa9fd809973f22859dbd2e3b19e3b010660acd705dd6f1b0a
+size 402763943

pytorch_model_00019-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e789af96eb4c2b181a72e5e4a8c03e85f02516e41abc9dcd6e88e1769bcecbd
+size 402763943

pytorch_model_00020-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c4df2b4390deceb3c830a72a86c7e7ed373b0a90aee0e4e9332f0e1c36a64e01
+size 402763943

pytorch_model_00021-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c9e8c8ab937dddd8bcd5c0b3aef2d2f5a4271ed56fdec3b4aced5e3a9737c81
+size 402763943

pytorch_model_00022-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c33b29956092113fdc73aa3021cb9558050164e404b34efa88b99075b9dec040
+size 402763943

pytorch_model_00023-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8645bbbb558de56f57e0dc29a083872cddb6d88eabd765e5dccd633b86c68609
+size 402763943

pytorch_model_00024-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb1dbe6b96d8af6919b1e7284134de6580aa1d789d9f83514bdb0fc33221411e
+size 402763943

pytorch_model_00025-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad4c61bcc0649bcaea63a9a081112984df4cafff0ab56896513dca7297687b92
+size 402763943

pytorch_model_00026-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cea4e4aac3246b4e58533f71bc3c0d001d804c7af02de79fcd2821cfc27fc77
+size 402763943

pytorch_model_00027-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9ad539f7bac6ad3301803736069975969d2a10592dc9d96ac3301d3f0185a02
+size 402763943

pytorch_model_00028-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad02db2dbd90e8019845028454fc618bb1a07f9a41cd6939accc97593f2911be
+size 402763943

pytorch_model_00029-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:755e371add07052069ff71fa0903bbc54336d5fda0ecdb6e9d4d66a764667c64
+size 402763943

pytorch_model_00030-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60ef8192a600e0426d358c27af44e24e0ff52d54a06fe191086a6055ceb6535b
+size 402763943

pytorch_model_00031-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:febaa7b6d6b850ccda5a36cb8789369a9e9c4df663a8119bafbadcc9daab6c10
+size 402763943

pytorch_model_00032-of-00032.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:104b1814a9bc1315c31f71dec692f5578d16b77c8613fd8d4e76a840ff749bcd
+size 17595