Hugging Face's logo

yuersan
/
test

Model card Files Community
yuersan commited on
Commit
f1c87d0
•
1 Parent(s): 87a6ac7

Upload 13 files

Browse files
Files changed (13) hide show
  1. config.json +28 -0
  2. merges.txt +0 -0
  3. model.safetensors.index.json +671 -0
  4. optimizer.pt +3 -0
  5. pytorch_model.bin.index.json +671 -0
  6. rng_state.pth +3 -0
  7. scheduler.pt +3 -0
  8. special_tokens_map.json +1 -0
  9. tokenizer.json +0 -0
  10. tokenizer_config.json +1 -0
  11. trainer_state.json +3640 -0
  12. training_args.bin +3 -0
  13. vocab.json +0 -0
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "EleutherAI/gpt-neox-20b",
3
+ "architectures": [
4
+ "GPTNeoXForCausalLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": 0.1,
9
+ "eos_token_id": 0,
10
+ "hidden_act": "gelu_fast",
11
+ "hidden_dropout_prob": 0,
12
+ "hidden_size": 6144,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 24576,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 2048,
17
+ "model_type": "gpt_neox",
18
+ "num_attention_heads": 64,
19
+ "num_hidden_layers": 44,
20
+ "rotary_emb_base": 10000,
21
+ "rotary_pct": 0.25,
22
+ "tie_word_embeddings": false,
23
+ "torch_dtype": "bfloat16",
24
+ "transformers_version": "4.30.0.dev0",
25
+ "use_cache": true,
26
+ "use_parallel_residual": true,
27
+ "vocab_size": 50432
28
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors.index.json ADDED
@@ -0,0 +1,671 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 41293685880
4
+ },
5
+ "weight_map": {
6
+ "embed_out.weight": "model-00046-of-00046.safetensors",
7
+ "gpt_neox.embed_in.weight": "model-00001-of-00046.safetensors",
8
+ "gpt_neox.final_layer_norm.bias": "model-00045-of-00046.safetensors",
9
+ "gpt_neox.final_layer_norm.weight": "model-00045-of-00046.safetensors",
10
+ "gpt_neox.layers.0.attention.bias": "model-00001-of-00046.safetensors",
11
+ "gpt_neox.layers.0.attention.dense.bias": "model-00001-of-00046.safetensors",
12
+ "gpt_neox.layers.0.attention.dense.weight": "model-00001-of-00046.safetensors",
13
+ "gpt_neox.layers.0.attention.masked_bias": "model-00001-of-00046.safetensors",
14
+ "gpt_neox.layers.0.attention.query_key_value.bias": "model-00001-of-00046.safetensors",
15
+ "gpt_neox.layers.0.attention.query_key_value.weight": "model-00001-of-00046.safetensors",
16
+ "gpt_neox.layers.0.attention.rotary_emb.inv_freq": "model-00001-of-00046.safetensors",
17
+ "gpt_neox.layers.0.input_layernorm.bias": "model-00001-of-00046.safetensors",
18
+ "gpt_neox.layers.0.input_layernorm.weight": "model-00001-of-00046.safetensors",
19
+ "gpt_neox.layers.0.mlp.dense_4h_to_h.bias": "model-00002-of-00046.safetensors",
20
+ "gpt_neox.layers.0.mlp.dense_4h_to_h.weight": "model-00002-of-00046.safetensors",
21
+ "gpt_neox.layers.0.mlp.dense_h_to_4h.bias": "model-00002-of-00046.safetensors",
22
+ "gpt_neox.layers.0.mlp.dense_h_to_4h.weight": "model-00002-of-00046.safetensors",
23
+ "gpt_neox.layers.0.post_attention_layernorm.bias": "model-00001-of-00046.safetensors",
24
+ "gpt_neox.layers.0.post_attention_layernorm.weight": "model-00001-of-00046.safetensors",
25
+ "gpt_neox.layers.1.attention.bias": "model-00002-of-00046.safetensors",
26
+ "gpt_neox.layers.1.attention.dense.bias": "model-00002-of-00046.safetensors",
27
+ "gpt_neox.layers.1.attention.dense.weight": "model-00002-of-00046.safetensors",
28
+ "gpt_neox.layers.1.attention.masked_bias": "model-00002-of-00046.safetensors",
29
+ "gpt_neox.layers.1.attention.query_key_value.bias": "model-00002-of-00046.safetensors",
30
+ "gpt_neox.layers.1.attention.query_key_value.weight": "model-00002-of-00046.safetensors",
31
+ "gpt_neox.layers.1.attention.rotary_emb.inv_freq": "model-00002-of-00046.safetensors",
32
+ "gpt_neox.layers.1.input_layernorm.bias": "model-00002-of-00046.safetensors",
33
+ "gpt_neox.layers.1.input_layernorm.weight": "model-00002-of-00046.safetensors",
34
+ "gpt_neox.layers.1.mlp.dense_4h_to_h.bias": "model-00003-of-00046.safetensors",
35
+ "gpt_neox.layers.1.mlp.dense_4h_to_h.weight": "model-00003-of-00046.safetensors",
36
+ "gpt_neox.layers.1.mlp.dense_h_to_4h.bias": "model-00003-of-00046.safetensors",
37
+ "gpt_neox.layers.1.mlp.dense_h_to_4h.weight": "model-00003-of-00046.safetensors",
38
+ "gpt_neox.layers.1.post_attention_layernorm.bias": "model-00002-of-00046.safetensors",
39
+ "gpt_neox.layers.1.post_attention_layernorm.weight": "model-00002-of-00046.safetensors",
40
+ "gpt_neox.layers.10.attention.bias": "model-00011-of-00046.safetensors",
41
+ "gpt_neox.layers.10.attention.dense.bias": "model-00011-of-00046.safetensors",
42
+ "gpt_neox.layers.10.attention.dense.weight": "model-00011-of-00046.safetensors",
43
+ "gpt_neox.layers.10.attention.masked_bias": "model-00011-of-00046.safetensors",
44
+ "gpt_neox.layers.10.attention.query_key_value.bias": "model-00011-of-00046.safetensors",
45
+ "gpt_neox.layers.10.attention.query_key_value.weight": "model-00011-of-00046.safetensors",
46
+ "gpt_neox.layers.10.attention.rotary_emb.inv_freq": "model-00011-of-00046.safetensors",
47
+ "gpt_neox.layers.10.input_layernorm.bias": "model-00011-of-00046.safetensors",
48
+ "gpt_neox.layers.10.input_layernorm.weight": "model-00011-of-00046.safetensors",
49
+ "gpt_neox.layers.10.mlp.dense_4h_to_h.bias": "model-00012-of-00046.safetensors",
50
+ "gpt_neox.layers.10.mlp.dense_4h_to_h.weight": "model-00012-of-00046.safetensors",
51
+ "gpt_neox.layers.10.mlp.dense_h_to_4h.bias": "model-00012-of-00046.safetensors",
52
+ "gpt_neox.layers.10.mlp.dense_h_to_4h.weight": "model-00012-of-00046.safetensors",
53
+ "gpt_neox.layers.10.post_attention_layernorm.bias": "model-00011-of-00046.safetensors",
54
+ "gpt_neox.layers.10.post_attention_layernorm.weight": "model-00011-of-00046.safetensors",
55
+ "gpt_neox.layers.11.attention.bias": "model-00012-of-00046.safetensors",
56
+ "gpt_neox.layers.11.attention.dense.bias": "model-00012-of-00046.safetensors",
57
+ "gpt_neox.layers.11.attention.dense.weight": "model-00012-of-00046.safetensors",
58
+ "gpt_neox.layers.11.attention.masked_bias": "model-00012-of-00046.safetensors",
59
+ "gpt_neox.layers.11.attention.query_key_value.bias": "model-00012-of-00046.safetensors",
60
+ "gpt_neox.layers.11.attention.query_key_value.weight": "model-00012-of-00046.safetensors",
61
+ "gpt_neox.layers.11.attention.rotary_emb.inv_freq": "model-00012-of-00046.safetensors",
62
+ "gpt_neox.layers.11.input_layernorm.bias": "model-00012-of-00046.safetensors",
63
+ "gpt_neox.layers.11.input_layernorm.weight": "model-00012-of-00046.safetensors",
64
+ "gpt_neox.layers.11.mlp.dense_4h_to_h.bias": "model-00013-of-00046.safetensors",
65
+ "gpt_neox.layers.11.mlp.dense_4h_to_h.weight": "model-00013-of-00046.safetensors",
66
+ "gpt_neox.layers.11.mlp.dense_h_to_4h.bias": "model-00013-of-00046.safetensors",
67
+ "gpt_neox.layers.11.mlp.dense_h_to_4h.weight": "model-00013-of-00046.safetensors",
68
+ "gpt_neox.layers.11.post_attention_layernorm.bias": "model-00012-of-00046.safetensors",
69
+ "gpt_neox.layers.11.post_attention_layernorm.weight": "model-00012-of-00046.safetensors",
70
+ "gpt_neox.layers.12.attention.bias": "model-00013-of-00046.safetensors",
71
+ "gpt_neox.layers.12.attention.dense.bias": "model-00013-of-00046.safetensors",
72
+ "gpt_neox.layers.12.attention.dense.weight": "model-00013-of-00046.safetensors",
73
+ "gpt_neox.layers.12.attention.masked_bias": "model-00013-of-00046.safetensors",
74
+ "gpt_neox.layers.12.attention.query_key_value.bias": "model-00013-of-00046.safetensors",
75
+ "gpt_neox.layers.12.attention.query_key_value.weight": "model-00013-of-00046.safetensors",
76
+ "gpt_neox.layers.12.attention.rotary_emb.inv_freq": "model-00013-of-00046.safetensors",
77
+ "gpt_neox.layers.12.input_layernorm.bias": "model-00013-of-00046.safetensors",
78
+ "gpt_neox.layers.12.input_layernorm.weight": "model-00013-of-00046.safetensors",
79
+ "gpt_neox.layers.12.mlp.dense_4h_to_h.bias": "model-00014-of-00046.safetensors",
80
+ "gpt_neox.layers.12.mlp.dense_4h_to_h.weight": "model-00014-of-00046.safetensors",
81
+ "gpt_neox.layers.12.mlp.dense_h_to_4h.bias": "model-00014-of-00046.safetensors",
82
+ "gpt_neox.layers.12.mlp.dense_h_to_4h.weight": "model-00014-of-00046.safetensors",
83
+ "gpt_neox.layers.12.post_attention_layernorm.bias": "model-00013-of-00046.safetensors",
84
+ "gpt_neox.layers.12.post_attention_layernorm.weight": "model-00013-of-00046.safetensors",
85
+ "gpt_neox.layers.13.attention.bias": "model-00014-of-00046.safetensors",
86
+ "gpt_neox.layers.13.attention.dense.bias": "model-00014-of-00046.safetensors",
87
+ "gpt_neox.layers.13.attention.dense.weight": "model-00014-of-00046.safetensors",
88
+ "gpt_neox.layers.13.attention.masked_bias": "model-00014-of-00046.safetensors",
89
+ "gpt_neox.layers.13.attention.query_key_value.bias": "model-00014-of-00046.safetensors",
90
+ "gpt_neox.layers.13.attention.query_key_value.weight": "model-00014-of-00046.safetensors",
91
+ "gpt_neox.layers.13.attention.rotary_emb.inv_freq": "model-00014-of-00046.safetensors",
92
+ "gpt_neox.layers.13.input_layernorm.bias": "model-00014-of-00046.safetensors",
93
+ "gpt_neox.layers.13.input_layernorm.weight": "model-00014-of-00046.safetensors",
94
+ "gpt_neox.layers.13.mlp.dense_4h_to_h.bias": "model-00015-of-00046.safetensors",
95
+ "gpt_neox.layers.13.mlp.dense_4h_to_h.weight": "model-00015-of-00046.safetensors",
96
+ "gpt_neox.layers.13.mlp.dense_h_to_4h.bias": "model-00015-of-00046.safetensors",
97
+ "gpt_neox.layers.13.mlp.dense_h_to_4h.weight": "model-00015-of-00046.safetensors",
98
+ "gpt_neox.layers.13.post_attention_layernorm.bias": "model-00014-of-00046.safetensors",
99
+ "gpt_neox.layers.13.post_attention_layernorm.weight": "model-00014-of-00046.safetensors",
100
+ "gpt_neox.layers.14.attention.bias": "model-00015-of-00046.safetensors",
101
+ "gpt_neox.layers.14.attention.dense.bias": "model-00015-of-00046.safetensors",
102
+ "gpt_neox.layers.14.attention.dense.weight": "model-00015-of-00046.safetensors",
103
+ "gpt_neox.layers.14.attention.masked_bias": "model-00015-of-00046.safetensors",
104
+ "gpt_neox.layers.14.attention.query_key_value.bias": "model-00015-of-00046.safetensors",
105
+ "gpt_neox.layers.14.attention.query_key_value.weight": "model-00015-of-00046.safetensors",
106
+ "gpt_neox.layers.14.attention.rotary_emb.inv_freq": "model-00015-of-00046.safetensors",
107
+ "gpt_neox.layers.14.input_layernorm.bias": "model-00015-of-00046.safetensors",
108
+ "gpt_neox.layers.14.input_layernorm.weight": "model-00015-of-00046.safetensors",
109
+ "gpt_neox.layers.14.mlp.dense_4h_to_h.bias": "model-00016-of-00046.safetensors",
110
+ "gpt_neox.layers.14.mlp.dense_4h_to_h.weight": "model-00016-of-00046.safetensors",
111
+ "gpt_neox.layers.14.mlp.dense_h_to_4h.bias": "model-00016-of-00046.safetensors",
112
+ "gpt_neox.layers.14.mlp.dense_h_to_4h.weight": "model-00016-of-00046.safetensors",
113
+ "gpt_neox.layers.14.post_attention_layernorm.bias": "model-00015-of-00046.safetensors",
114
+ "gpt_neox.layers.14.post_attention_layernorm.weight": "model-00015-of-00046.safetensors",
115
+ "gpt_neox.layers.15.attention.bias": "model-00016-of-00046.safetensors",
116
+ "gpt_neox.layers.15.attention.dense.bias": "model-00016-of-00046.safetensors",
117
+ "gpt_neox.layers.15.attention.dense.weight": "model-00016-of-00046.safetensors",
118
+ "gpt_neox.layers.15.attention.masked_bias": "model-00016-of-00046.safetensors",
119
+ "gpt_neox.layers.15.attention.query_key_value.bias": "model-00016-of-00046.safetensors",
120
+ "gpt_neox.layers.15.attention.query_key_value.weight": "model-00016-of-00046.safetensors",
121
+ "gpt_neox.layers.15.attention.rotary_emb.inv_freq": "model-00016-of-00046.safetensors",
122
+ "gpt_neox.layers.15.input_layernorm.bias": "model-00016-of-00046.safetensors",
123
+ "gpt_neox.layers.15.input_layernorm.weight": "model-00016-of-00046.safetensors",
124
+ "gpt_neox.layers.15.mlp.dense_4h_to_h.bias": "model-00017-of-00046.safetensors",
125
+ "gpt_neox.layers.15.mlp.dense_4h_to_h.weight": "model-00017-of-00046.safetensors",
126
+ "gpt_neox.layers.15.mlp.dense_h_to_4h.bias": "model-00017-of-00046.safetensors",
127
+ "gpt_neox.layers.15.mlp.dense_h_to_4h.weight": "model-00017-of-00046.safetensors",
128
+ "gpt_neox.layers.15.post_attention_layernorm.bias": "model-00016-of-00046.safetensors",
129
+ "gpt_neox.layers.15.post_attention_layernorm.weight": "model-00016-of-00046.safetensors",
130
+ "gpt_neox.layers.16.attention.bias": "model-00017-of-00046.safetensors",
131
+ "gpt_neox.layers.16.attention.dense.bias": "model-00017-of-00046.safetensors",
132
+ "gpt_neox.layers.16.attention.dense.weight": "model-00017-of-00046.safetensors",
133
+ "gpt_neox.layers.16.attention.masked_bias": "model-00017-of-00046.safetensors",
134
+ "gpt_neox.layers.16.attention.query_key_value.bias": "model-00017-of-00046.safetensors",
135
+ "gpt_neox.layers.16.attention.query_key_value.weight": "model-00017-of-00046.safetensors",
136
+ "gpt_neox.layers.16.attention.rotary_emb.inv_freq": "model-00017-of-00046.safetensors",
137
+ "gpt_neox.layers.16.input_layernorm.bias": "model-00017-of-00046.safetensors",
138
+ "gpt_neox.layers.16.input_layernorm.weight": "model-00017-of-00046.safetensors",
139
+ "gpt_neox.layers.16.mlp.dense_4h_to_h.bias": "model-00018-of-00046.safetensors",
140
+ "gpt_neox.layers.16.mlp.dense_4h_to_h.weight": "model-00018-of-00046.safetensors",
141
+ "gpt_neox.layers.16.mlp.dense_h_to_4h.bias": "model-00018-of-00046.safetensors",
142
+ "gpt_neox.layers.16.mlp.dense_h_to_4h.weight": "model-00018-of-00046.safetensors",
143
+ "gpt_neox.layers.16.post_attention_layernorm.bias": "model-00017-of-00046.safetensors",
144
+ "gpt_neox.layers.16.post_attention_layernorm.weight": "model-00017-of-00046.safetensors",
145
+ "gpt_neox.layers.17.attention.bias": "model-00018-of-00046.safetensors",
146
+ "gpt_neox.layers.17.attention.dense.bias": "model-00018-of-00046.safetensors",
147
+ "gpt_neox.layers.17.attention.dense.weight": "model-00018-of-00046.safetensors",
148
+ "gpt_neox.layers.17.attention.masked_bias": "model-00018-of-00046.safetensors",
149
+ "gpt_neox.layers.17.attention.query_key_value.bias": "model-00018-of-00046.safetensors",
150
+ "gpt_neox.layers.17.attention.query_key_value.weight": "model-00018-of-00046.safetensors",
151
+ "gpt_neox.layers.17.attention.rotary_emb.inv_freq": "model-00018-of-00046.safetensors",
152
+ "gpt_neox.layers.17.input_layernorm.bias": "model-00018-of-00046.safetensors",
153
+ "gpt_neox.layers.17.input_layernorm.weight": "model-00018-of-00046.safetensors",
154
+ "gpt_neox.layers.17.mlp.dense_4h_to_h.bias": "model-00019-of-00046.safetensors",
155
+ "gpt_neox.layers.17.mlp.dense_4h_to_h.weight": "model-00019-of-00046.safetensors",
156
+ "gpt_neox.layers.17.mlp.dense_h_to_4h.bias": "model-00019-of-00046.safetensors",
157
+ "gpt_neox.layers.17.mlp.dense_h_to_4h.weight": "model-00019-of-00046.safetensors",
158
+ "gpt_neox.layers.17.post_attention_layernorm.bias": "model-00018-of-00046.safetensors",
159
+ "gpt_neox.layers.17.post_attention_layernorm.weight": "model-00018-of-00046.safetensors",
160
+ "gpt_neox.layers.18.attention.bias": "model-00019-of-00046.safetensors",
161
+ "gpt_neox.layers.18.attention.dense.bias": "model-00019-of-00046.safetensors",
162
+ "gpt_neox.layers.18.attention.dense.weight": "model-00019-of-00046.safetensors",
163
+ "gpt_neox.layers.18.attention.masked_bias": "model-00019-of-00046.safetensors",
164
+ "gpt_neox.layers.18.attention.query_key_value.bias": "model-00019-of-00046.safetensors",
165
+ "gpt_neox.layers.18.attention.query_key_value.weight": "model-00019-of-00046.safetensors",
166
+ "gpt_neox.layers.18.attention.rotary_emb.inv_freq": "model-00019-of-00046.safetensors",
167
+ "gpt_neox.layers.18.input_layernorm.bias": "model-00019-of-00046.safetensors",
168
+ "gpt_neox.layers.18.input_layernorm.weight": "model-00019-of-00046.safetensors",
169
+ "gpt_neox.layers.18.mlp.dense_4h_to_h.bias": "model-00020-of-00046.safetensors",
170
+ "gpt_neox.layers.18.mlp.dense_4h_to_h.weight": "model-00020-of-00046.safetensors",
171
+ "gpt_neox.layers.18.mlp.dense_h_to_4h.bias": "model-00020-of-00046.safetensors",
172
+ "gpt_neox.layers.18.mlp.dense_h_to_4h.weight": "model-00020-of-00046.safetensors",
173
+ "gpt_neox.layers.18.post_attention_layernorm.bias": "model-00019-of-00046.safetensors",
174
+ "gpt_neox.layers.18.post_attention_layernorm.weight": "model-00019-of-00046.safetensors",
175
+ "gpt_neox.layers.19.attention.bias": "model-00020-of-00046.safetensors",
176
+ "gpt_neox.layers.19.attention.dense.bias": "model-00020-of-00046.safetensors",
177
+ "gpt_neox.layers.19.attention.dense.weight": "model-00020-of-00046.safetensors",
178
+ "gpt_neox.layers.19.attention.masked_bias": "model-00020-of-00046.safetensors",
179
+ "gpt_neox.layers.19.attention.query_key_value.bias": "model-00020-of-00046.safetensors",
180
+ "gpt_neox.layers.19.attention.query_key_value.weight": "model-00020-of-00046.safetensors",
181
+ "gpt_neox.layers.19.attention.rotary_emb.inv_freq": "model-00020-of-00046.safetensors",
182
+ "gpt_neox.layers.19.input_layernorm.bias": "model-00020-of-00046.safetensors",
183
+ "gpt_neox.layers.19.input_layernorm.weight": "model-00020-of-00046.safetensors",
184
+ "gpt_neox.layers.19.mlp.dense_4h_to_h.bias": "model-00021-of-00046.safetensors",
185
+ "gpt_neox.layers.19.mlp.dense_4h_to_h.weight": "model-00021-of-00046.safetensors",
186
+ "gpt_neox.layers.19.mlp.dense_h_to_4h.bias": "model-00021-of-00046.safetensors",
187
+ "gpt_neox.layers.19.mlp.dense_h_to_4h.weight": "model-00021-of-00046.safetensors",
188
+ "gpt_neox.layers.19.post_attention_layernorm.bias": "model-00020-of-00046.safetensors",
189
+ "gpt_neox.layers.19.post_attention_layernorm.weight": "model-00020-of-00046.safetensors",
190
+ "gpt_neox.layers.2.attention.bias": "model-00003-of-00046.safetensors",
191
+ "gpt_neox.layers.2.attention.dense.bias": "model-00003-of-00046.safetensors",
192
+ "gpt_neox.layers.2.attention.dense.weight": "model-00003-of-00046.safetensors",
193
+ "gpt_neox.layers.2.attention.masked_bias": "model-00003-of-00046.safetensors",
194
+ "gpt_neox.layers.2.attention.query_key_value.bias": "model-00003-of-00046.safetensors",
195
+ "gpt_neox.layers.2.attention.query_key_value.weight": "model-00003-of-00046.safetensors",
196
+ "gpt_neox.layers.2.attention.rotary_emb.inv_freq": "model-00003-of-00046.safetensors",
197
+ "gpt_neox.layers.2.input_layernorm.bias": "model-00003-of-00046.safetensors",
198
+ "gpt_neox.layers.2.input_layernorm.weight": "model-00003-of-00046.safetensors",
199
+ "gpt_neox.layers.2.mlp.dense_4h_to_h.bias": "model-00004-of-00046.safetensors",
200
+ "gpt_neox.layers.2.mlp.dense_4h_to_h.weight": "model-00004-of-00046.safetensors",
201
+ "gpt_neox.layers.2.mlp.dense_h_to_4h.bias": "model-00004-of-00046.safetensors",
202
+ "gpt_neox.layers.2.mlp.dense_h_to_4h.weight": "model-00004-of-00046.safetensors",
203
+ "gpt_neox.layers.2.post_attention_layernorm.bias": "model-00003-of-00046.safetensors",
204
+ "gpt_neox.layers.2.post_attention_layernorm.weight": "model-00003-of-00046.safetensors",
205
+ "gpt_neox.layers.20.attention.bias": "model-00021-of-00046.safetensors",
206
+ "gpt_neox.layers.20.attention.dense.bias": "model-00021-of-00046.safetensors",
207
+ "gpt_neox.layers.20.attention.dense.weight": "model-00021-of-00046.safetensors",
208
+ "gpt_neox.layers.20.attention.masked_bias": "model-00021-of-00046.safetensors",
209
+ "gpt_neox.layers.20.attention.query_key_value.bias": "model-00021-of-00046.safetensors",
210
+ "gpt_neox.layers.20.attention.query_key_value.weight": "model-00021-of-00046.safetensors",
211
+ "gpt_neox.layers.20.attention.rotary_emb.inv_freq": "model-00021-of-00046.safetensors",
212
+ "gpt_neox.layers.20.input_layernorm.bias": "model-00021-of-00046.safetensors",
213
+ "gpt_neox.layers.20.input_layernorm.weight": "model-00021-of-00046.safetensors",
214
+ "gpt_neox.layers.20.mlp.dense_4h_to_h.bias": "model-00022-of-00046.safetensors",
215
+ "gpt_neox.layers.20.mlp.dense_4h_to_h.weight": "model-00022-of-00046.safetensors",
216
+ "gpt_neox.layers.20.mlp.dense_h_to_4h.bias": "model-00022-of-00046.safetensors",
217
+ "gpt_neox.layers.20.mlp.dense_h_to_4h.weight": "model-00022-of-00046.safetensors",
218
+ "gpt_neox.layers.20.post_attention_layernorm.bias": "model-00021-of-00046.safetensors",
219
+ "gpt_neox.layers.20.post_attention_layernorm.weight": "model-00021-of-00046.safetensors",
220
+ "gpt_neox.layers.21.attention.bias": "model-00022-of-00046.safetensors",
221
+ "gpt_neox.layers.21.attention.dense.bias": "model-00022-of-00046.safetensors",
222
+ "gpt_neox.layers.21.attention.dense.weight": "model-00022-of-00046.safetensors",
223
+ "gpt_neox.layers.21.attention.masked_bias": "model-00022-of-00046.safetensors",
224
+ "gpt_neox.layers.21.attention.query_key_value.bias": "model-00022-of-00046.safetensors",
225
+ "gpt_neox.layers.21.attention.query_key_value.weight": "model-00022-of-00046.safetensors",
226
+ "gpt_neox.layers.21.attention.rotary_emb.inv_freq": "model-00022-of-00046.safetensors",
227
+ "gpt_neox.layers.21.input_layernorm.bias": "model-00022-of-00046.safetensors",
228
+ "gpt_neox.layers.21.input_layernorm.weight": "model-00022-of-00046.safetensors",
229
+ "gpt_neox.layers.21.mlp.dense_4h_to_h.bias": "model-00023-of-00046.safetensors",
230
+ "gpt_neox.layers.21.mlp.dense_4h_to_h.weight": "model-00023-of-00046.safetensors",
231
+ "gpt_neox.layers.21.mlp.dense_h_to_4h.bias": "model-00023-of-00046.safetensors",
232
+ "gpt_neox.layers.21.mlp.dense_h_to_4h.weight": "model-00023-of-00046.safetensors",
233
+ "gpt_neox.layers.21.post_attention_layernorm.bias": "model-00022-of-00046.safetensors",
234
+ "gpt_neox.layers.21.post_attention_layernorm.weight": "model-00022-of-00046.safetensors",
235
+ "gpt_neox.layers.22.attention.bias": "model-00023-of-00046.safetensors",
236
+ "gpt_neox.layers.22.attention.dense.bias": "model-00023-of-00046.safetensors",
237
+ "gpt_neox.layers.22.attention.dense.weight": "model-00023-of-00046.safetensors",
238
+ "gpt_neox.layers.22.attention.masked_bias": "model-00023-of-00046.safetensors",
239
+ "gpt_neox.layers.22.attention.query_key_value.bias": "model-00023-of-00046.safetensors",
240
+ "gpt_neox.layers.22.attention.query_key_value.weight": "model-00023-of-00046.safetensors",
241
+ "gpt_neox.layers.22.attention.rotary_emb.inv_freq": "model-00023-of-00046.safetensors",
242
+ "gpt_neox.layers.22.input_layernorm.bias": "model-00023-of-00046.safetensors",
243
+ "gpt_neox.layers.22.input_layernorm.weight": "model-00023-of-00046.safetensors",
244
+ "gpt_neox.layers.22.mlp.dense_4h_to_h.bias": "model-00024-of-00046.safetensors",
245
+ "gpt_neox.layers.22.mlp.dense_4h_to_h.weight": "model-00024-of-00046.safetensors",
246
+ "gpt_neox.layers.22.mlp.dense_h_to_4h.bias": "model-00024-of-00046.safetensors",
247
+ "gpt_neox.layers.22.mlp.dense_h_to_4h.weight": "model-00024-of-00046.safetensors",
248
+ "gpt_neox.layers.22.post_attention_layernorm.bias": "model-00023-of-00046.safetensors",
249
+ "gpt_neox.layers.22.post_attention_layernorm.weight": "model-00023-of-00046.safetensors",
250
+ "gpt_neox.layers.23.attention.bias": "model-00024-of-00046.safetensors",
251
+ "gpt_neox.layers.23.attention.dense.bias": "model-00024-of-00046.safetensors",
252
+ "gpt_neox.layers.23.attention.dense.weight": "model-00024-of-00046.safetensors",
253
+ "gpt_neox.layers.23.attention.masked_bias": "model-00024-of-00046.safetensors",
254
+ "gpt_neox.layers.23.attention.query_key_value.bias": "model-00024-of-00046.safetensors",
255
+ "gpt_neox.layers.23.attention.query_key_value.weight": "model-00024-of-00046.safetensors",
256
+ "gpt_neox.layers.23.attention.rotary_emb.inv_freq": "model-00024-of-00046.safetensors",
257
+ "gpt_neox.layers.23.input_layernorm.bias": "model-00024-of-00046.safetensors",
258
+ "gpt_neox.layers.23.input_layernorm.weight": "model-00024-of-00046.safetensors",
259
+ "gpt_neox.layers.23.mlp.dense_4h_to_h.bias": "model-00025-of-00046.safetensors",
260
+ "gpt_neox.layers.23.mlp.dense_4h_to_h.weight": "model-00025-of-00046.safetensors",
261
+ "gpt_neox.layers.23.mlp.dense_h_to_4h.bias": "model-00025-of-00046.safetensors",
262
+ "gpt_neox.layers.23.mlp.dense_h_to_4h.weight": "model-00025-of-00046.safetensors",
263
+ "gpt_neox.layers.23.post_attention_layernorm.bias": "model-00024-of-00046.safetensors",
264
+ "gpt_neox.layers.23.post_attention_layernorm.weight": "model-00024-of-00046.safetensors",
265
+ "gpt_neox.layers.24.attention.bias": "model-00025-of-00046.safetensors",
266
+ "gpt_neox.layers.24.attention.dense.bias": "model-00025-of-00046.safetensors",
267
+ "gpt_neox.layers.24.attention.dense.weight": "model-00025-of-00046.safetensors",
268
+ "gpt_neox.layers.24.attention.masked_bias": "model-00025-of-00046.safetensors",
269
+ "gpt_neox.layers.24.attention.query_key_value.bias": "model-00025-of-00046.safetensors",
270
+ "gpt_neox.layers.24.attention.query_key_value.weight": "model-00025-of-00046.safetensors",
271
+ "gpt_neox.layers.24.attention.rotary_emb.inv_freq": "model-00025-of-00046.safetensors",
272
+ "gpt_neox.layers.24.input_layernorm.bias": "model-00025-of-00046.safetensors",
273
+ "gpt_neox.layers.24.input_layernorm.weight": "model-00025-of-00046.safetensors",
274
+ "gpt_neox.layers.24.mlp.dense_4h_to_h.bias": "model-00026-of-00046.safetensors",
275
+ "gpt_neox.layers.24.mlp.dense_4h_to_h.weight": "model-00026-of-00046.safetensors",
276
+ "gpt_neox.layers.24.mlp.dense_h_to_4h.bias": "model-00026-of-00046.safetensors",
277
+ "gpt_neox.layers.24.mlp.dense_h_to_4h.weight": "model-00026-of-00046.safetensors",
278
+ "gpt_neox.layers.24.post_attention_layernorm.bias": "model-00025-of-00046.safetensors",
279
+ "gpt_neox.layers.24.post_attention_layernorm.weight": "model-00025-of-00046.safetensors",
280
+ "gpt_neox.layers.25.attention.bias": "model-00026-of-00046.safetensors",
281
+ "gpt_neox.layers.25.attention.dense.bias": "model-00026-of-00046.safetensors",
282
+ "gpt_neox.layers.25.attention.dense.weight": "model-00026-of-00046.safetensors",
283
+ "gpt_neox.layers.25.attention.masked_bias": "model-00026-of-00046.safetensors",
284
+ "gpt_neox.layers.25.attention.query_key_value.bias": "model-00026-of-00046.safetensors",
285
+ "gpt_neox.layers.25.attention.query_key_value.weight": "model-00026-of-00046.safetensors",
286
+ "gpt_neox.layers.25.attention.rotary_emb.inv_freq": "model-00026-of-00046.safetensors",
287
+ "gpt_neox.layers.25.input_layernorm.bias": "model-00026-of-00046.safetensors",
288
+ "gpt_neox.layers.25.input_layernorm.weight": "model-00026-of-00046.safetensors",
289
+ "gpt_neox.layers.25.mlp.dense_4h_to_h.bias": "model-00027-of-00046.safetensors",
290
+ "gpt_neox.layers.25.mlp.dense_4h_to_h.weight": "model-00027-of-00046.safetensors",
291
+ "gpt_neox.layers.25.mlp.dense_h_to_4h.bias": "model-00027-of-00046.safetensors",
292
+ "gpt_neox.layers.25.mlp.dense_h_to_4h.weight": "model-00027-of-00046.safetensors",
293
+ "gpt_neox.layers.25.post_attention_layernorm.bias": "model-00026-of-00046.safetensors",
294
+ "gpt_neox.layers.25.post_attention_layernorm.weight": "model-00026-of-00046.safetensors",
295
+ "gpt_neox.layers.26.attention.bias": "model-00027-of-00046.safetensors",
296
+ "gpt_neox.layers.26.attention.dense.bias": "model-00027-of-00046.safetensors",
297
+ "gpt_neox.layers.26.attention.dense.weight": "model-00027-of-00046.safetensors",
298
+ "gpt_neox.layers.26.attention.masked_bias": "model-00027-of-00046.safetensors",
299
+ "gpt_neox.layers.26.attention.query_key_value.bias": "model-00027-of-00046.safetensors",
300
+ "gpt_neox.layers.26.attention.query_key_value.weight": "model-00027-of-00046.safetensors",
301
+ "gpt_neox.layers.26.attention.rotary_emb.inv_freq": "model-00027-of-00046.safetensors",
302
+ "gpt_neox.layers.26.input_layernorm.bias": "model-00027-of-00046.safetensors",
303
+ "gpt_neox.layers.26.input_layernorm.weight": "model-00027-of-00046.safetensors",
304
+ "gpt_neox.layers.26.mlp.dense_4h_to_h.bias": "model-00028-of-00046.safetensors",
305
+ "gpt_neox.layers.26.mlp.dense_4h_to_h.weight": "model-00028-of-00046.safetensors",
306
+ "gpt_neox.layers.26.mlp.dense_h_to_4h.bias": "model-00028-of-00046.safetensors",
307
+ "gpt_neox.layers.26.mlp.dense_h_to_4h.weight": "model-00028-of-00046.safetensors",
308
+ "gpt_neox.layers.26.post_attention_layernorm.bias": "model-00027-of-00046.safetensors",
309
+ "gpt_neox.layers.26.post_attention_layernorm.weight": "model-00027-of-00046.safetensors",
310
+ "gpt_neox.layers.27.attention.bias": "model-00028-of-00046.safetensors",
311
+ "gpt_neox.layers.27.attention.dense.bias": "model-00028-of-00046.safetensors",
312
+ "gpt_neox.layers.27.attention.dense.weight": "model-00028-of-00046.safetensors",
313
+ "gpt_neox.layers.27.attention.masked_bias": "model-00028-of-00046.safetensors",
314
+ "gpt_neox.layers.27.attention.query_key_value.bias": "model-00028-of-00046.safetensors",
315
+ "gpt_neox.layers.27.attention.query_key_value.weight": "model-00028-of-00046.safetensors",
316
+ "gpt_neox.layers.27.attention.rotary_emb.inv_freq": "model-00028-of-00046.safetensors",
317
+ "gpt_neox.layers.27.input_layernorm.bias": "model-00028-of-00046.safetensors",
318
+ "gpt_neox.layers.27.input_layernorm.weight": "model-00028-of-00046.safetensors",
319
+ "gpt_neox.layers.27.mlp.dense_4h_to_h.bias": "model-00029-of-00046.safetensors",
320
+ "gpt_neox.layers.27.mlp.dense_4h_to_h.weight": "model-00029-of-00046.safetensors",
321
+ "gpt_neox.layers.27.mlp.dense_h_to_4h.bias": "model-00029-of-00046.safetensors",
322
+ "gpt_neox.layers.27.mlp.dense_h_to_4h.weight": "model-00029-of-00046.safetensors",
323
+ "gpt_neox.layers.27.post_attention_layernorm.bias": "model-00028-of-00046.safetensors",
324
+ "gpt_neox.layers.27.post_attention_layernorm.weight": "model-00028-of-00046.safetensors",
325
+ "gpt_neox.layers.28.attention.bias": "model-00029-of-00046.safetensors",
326
+ "gpt_neox.layers.28.attention.dense.bias": "model-00029-of-00046.safetensors",
327
+ "gpt_neox.layers.28.attention.dense.weight": "model-00029-of-00046.safetensors",
328
+ "gpt_neox.layers.28.attention.masked_bias": "model-00029-of-00046.safetensors",
329
+ "gpt_neox.layers.28.attention.query_key_value.bias": "model-00029-of-00046.safetensors",
330
+ "gpt_neox.layers.28.attention.query_key_value.weight": "model-00029-of-00046.safetensors",
331
+ "gpt_neox.layers.28.attention.rotary_emb.inv_freq": "model-00029-of-00046.safetensors",
332
+ "gpt_neox.layers.28.input_layernorm.bias": "model-00029-of-00046.safetensors",
333
+ "gpt_neox.layers.28.input_layernorm.weight": "model-00029-of-00046.safetensors",
334
+ "gpt_neox.layers.28.mlp.dense_4h_to_h.bias": "model-00030-of-00046.safetensors",
335
+ "gpt_neox.layers.28.mlp.dense_4h_to_h.weight": "model-00030-of-00046.safetensors",
336
+ "gpt_neox.layers.28.mlp.dense_h_to_4h.bias": "model-00030-of-00046.safetensors",
337
+ "gpt_neox.layers.28.mlp.dense_h_to_4h.weight": "model-00030-of-00046.safetensors",
338
+ "gpt_neox.layers.28.post_attention_layernorm.bias": "model-00029-of-00046.safetensors",
339
+ "gpt_neox.layers.28.post_attention_layernorm.weight": "model-00029-of-00046.safetensors",
340
+ "gpt_neox.layers.29.attention.bias": "model-00030-of-00046.safetensors",
341
+ "gpt_neox.layers.29.attention.dense.bias": "model-00030-of-00046.safetensors",
342
+ "gpt_neox.layers.29.attention.dense.weight": "model-00030-of-00046.safetensors",
343
+ "gpt_neox.layers.29.attention.masked_bias": "model-00030-of-00046.safetensors",
344
+ "gpt_neox.layers.29.attention.query_key_value.bias": "model-00030-of-00046.safetensors",
345
+ "gpt_neox.layers.29.attention.query_key_value.weight": "model-00030-of-00046.safetensors",
346
+ "gpt_neox.layers.29.attention.rotary_emb.inv_freq": "model-00030-of-00046.safetensors",
347
+ "gpt_neox.layers.29.input_layernorm.bias": "model-00030-of-00046.safetensors",
348
+ "gpt_neox.layers.29.input_layernorm.weight": "model-00030-of-00046.safetensors",
349
+ "gpt_neox.layers.29.mlp.dense_4h_to_h.bias": "model-00031-of-00046.safetensors",
350
+ "gpt_neox.layers.29.mlp.dense_4h_to_h.weight": "model-00031-of-00046.safetensors",
351
+ "gpt_neox.layers.29.mlp.dense_h_to_4h.bias": "model-00031-of-00046.safetensors",
352
+ "gpt_neox.layers.29.mlp.dense_h_to_4h.weight": "model-00031-of-00046.safetensors",
353
+ "gpt_neox.layers.29.post_attention_layernorm.bias": "model-00030-of-00046.safetensors",
354
+ "gpt_neox.layers.29.post_attention_layernorm.weight": "model-00030-of-00046.safetensors",
355
+ "gpt_neox.layers.3.attention.bias": "model-00004-of-00046.safetensors",
356
+ "gpt_neox.layers.3.attention.dense.bias": "model-00004-of-00046.safetensors",
357
+ "gpt_neox.layers.3.attention.dense.weight": "model-00004-of-00046.safetensors",
358
+ "gpt_neox.layers.3.attention.masked_bias": "model-00004-of-00046.safetensors",
359
+ "gpt_neox.layers.3.attention.query_key_value.bias": "model-00004-of-00046.safetensors",
360
+ "gpt_neox.layers.3.attention.query_key_value.weight": "model-00004-of-00046.safetensors",
361
+ "gpt_neox.layers.3.attention.rotary_emb.inv_freq": "model-00004-of-00046.safetensors",
362
+ "gpt_neox.layers.3.input_layernorm.bias": "model-00004-of-00046.safetensors",
363
+ "gpt_neox.layers.3.input_layernorm.weight": "model-00004-of-00046.safetensors",
364
+ "gpt_neox.layers.3.mlp.dense_4h_to_h.bias": "model-00005-of-00046.safetensors",
365
+ "gpt_neox.layers.3.mlp.dense_4h_to_h.weight": "model-00005-of-00046.safetensors",
366
+ "gpt_neox.layers.3.mlp.dense_h_to_4h.bias": "model-00005-of-00046.safetensors",
367
+ "gpt_neox.layers.3.mlp.dense_h_to_4h.weight": "model-00005-of-00046.safetensors",
368
+ "gpt_neox.layers.3.post_attention_layernorm.bias": "model-00004-of-00046.safetensors",
369
+ "gpt_neox.layers.3.post_attention_layernorm.weight": "model-00004-of-00046.safetensors",
370
+ "gpt_neox.layers.30.attention.bias": "model-00031-of-00046.safetensors",
371
+ "gpt_neox.layers.30.attention.dense.bias": "model-00031-of-00046.safetensors",
372
+ "gpt_neox.layers.30.attention.dense.weight": "model-00031-of-00046.safetensors",
373
+ "gpt_neox.layers.30.attention.masked_bias": "model-00031-of-00046.safetensors",
374
+ "gpt_neox.layers.30.attention.query_key_value.bias": "model-00031-of-00046.safetensors",
375
+ "gpt_neox.layers.30.attention.query_key_value.weight": "model-00031-of-00046.safetensors",
376
+ "gpt_neox.layers.30.attention.rotary_emb.inv_freq": "model-00031-of-00046.safetensors",
377
+ "gpt_neox.layers.30.input_layernorm.bias": "model-00031-of-00046.safetensors",
378
+ "gpt_neox.layers.30.input_layernorm.weight": "model-00031-of-00046.safetensors",
379
+ "gpt_neox.layers.30.mlp.dense_4h_to_h.bias": "model-00032-of-00046.safetensors",
380
+ "gpt_neox.layers.30.mlp.dense_4h_to_h.weight": "model-00032-of-00046.safetensors",
381
+ "gpt_neox.layers.30.mlp.dense_h_to_4h.bias": "model-00032-of-00046.safetensors",
382
+ "gpt_neox.layers.30.mlp.dense_h_to_4h.weight": "model-00032-of-00046.safetensors",
383
+ "gpt_neox.layers.30.post_attention_layernorm.bias": "model-00031-of-00046.safetensors",
384
+ "gpt_neox.layers.30.post_attention_layernorm.weight": "model-00031-of-00046.safetensors",
385
+ "gpt_neox.layers.31.attention.bias": "model-00032-of-00046.safetensors",
386
+ "gpt_neox.layers.31.attention.dense.bias": "model-00032-of-00046.safetensors",
387
+ "gpt_neox.layers.31.attention.dense.weight": "model-00032-of-00046.safetensors",
388
+ "gpt_neox.layers.31.attention.masked_bias": "model-00032-of-00046.safetensors",
389
+ "gpt_neox.layers.31.attention.query_key_value.bias": "model-00032-of-00046.safetensors",
390
+ "gpt_neox.layers.31.attention.query_key_value.weight": "model-00032-of-00046.safetensors",
391
+ "gpt_neox.layers.31.attention.rotary_emb.inv_freq": "model-00032-of-00046.safetensors",
392
+ "gpt_neox.layers.31.input_layernorm.bias": "model-00032-of-00046.safetensors",
393
+ "gpt_neox.layers.31.input_layernorm.weight": "model-00032-of-00046.safetensors",
394
+ "gpt_neox.layers.31.mlp.dense_4h_to_h.bias": "model-00033-of-00046.safetensors",
395
+ "gpt_neox.layers.31.mlp.dense_4h_to_h.weight": "model-00033-of-00046.safetensors",
396
+ "gpt_neox.layers.31.mlp.dense_h_to_4h.bias": "model-00033-of-00046.safetensors",
397
+ "gpt_neox.layers.31.mlp.dense_h_to_4h.weight": "model-00033-of-00046.safetensors",
398
+ "gpt_neox.layers.31.post_attention_layernorm.bias": "model-00032-of-00046.safetensors",
399
+ "gpt_neox.layers.31.post_attention_layernorm.weight": "model-00032-of-00046.safetensors",
400
+ "gpt_neox.layers.32.attention.bias": "model-00033-of-00046.safetensors",
401
+ "gpt_neox.layers.32.attention.dense.bias": "model-00033-of-00046.safetensors",
402
+ "gpt_neox.layers.32.attention.dense.weight": "model-00033-of-00046.safetensors",
403
+ "gpt_neox.layers.32.attention.masked_bias": "model-00033-of-00046.safetensors",
404
+ "gpt_neox.layers.32.attention.query_key_value.bias": "model-00033-of-00046.safetensors",
405
+ "gpt_neox.layers.32.attention.query_key_value.weight": "model-00033-of-00046.safetensors",
406
+ "gpt_neox.layers.32.attention.rotary_emb.inv_freq": "model-00033-of-00046.safetensors",
407
+ "gpt_neox.layers.32.input_layernorm.bias": "model-00033-of-00046.safetensors",
408
+ "gpt_neox.layers.32.input_layernorm.weight": "model-00033-of-00046.safetensors",
409
+ "gpt_neox.layers.32.mlp.dense_4h_to_h.bias": "model-00034-of-00046.safetensors",
410
+ "gpt_neox.layers.32.mlp.dense_4h_to_h.weight": "model-00034-of-00046.safetensors",
411
+ "gpt_neox.layers.32.mlp.dense_h_to_4h.bias": "model-00034-of-00046.safetensors",
412
+ "gpt_neox.layers.32.mlp.dense_h_to_4h.weight": "model-00034-of-00046.safetensors",
413
+ "gpt_neox.layers.32.post_attention_layernorm.bias": "model-00033-of-00046.safetensors",
414
+ "gpt_neox.layers.32.post_attention_layernorm.weight": "model-00033-of-00046.safetensors",
415
+ "gpt_neox.layers.33.attention.bias": "model-00034-of-00046.safetensors",
416
+ "gpt_neox.layers.33.attention.dense.bias": "model-00034-of-00046.safetensors",
417
+ "gpt_neox.layers.33.attention.dense.weight": "model-00034-of-00046.safetensors",
418
+ "gpt_neox.layers.33.attention.masked_bias": "model-00034-of-00046.safetensors",
419
+ "gpt_neox.layers.33.attention.query_key_value.bias": "model-00034-of-00046.safetensors",
420
+ "gpt_neox.layers.33.attention.query_key_value.weight": "model-00034-of-00046.safetensors",
421
+ "gpt_neox.layers.33.attention.rotary_emb.inv_freq": "model-00034-of-00046.safetensors",
422
+ "gpt_neox.layers.33.input_layernorm.bias": "model-00034-of-00046.safetensors",
423
+ "gpt_neox.layers.33.input_layernorm.weight": "model-00034-of-00046.safetensors",
424
+ "gpt_neox.layers.33.mlp.dense_4h_to_h.bias": "model-00035-of-00046.safetensors",
425
+ "gpt_neox.layers.33.mlp.dense_4h_to_h.weight": "model-00035-of-00046.safetensors",
426
+ "gpt_neox.layers.33.mlp.dense_h_to_4h.bias": "model-00035-of-00046.safetensors",
427
+ "gpt_neox.layers.33.mlp.dense_h_to_4h.weight": "model-00035-of-00046.safetensors",
428
+ "gpt_neox.layers.33.post_attention_layernorm.bias": "model-00034-of-00046.safetensors",
429
+ "gpt_neox.layers.33.post_attention_layernorm.weight": "model-00034-of-00046.safetensors",
430
+ "gpt_neox.layers.34.attention.bias": "model-00035-of-00046.safetensors",
431
+ "gpt_neox.layers.34.attention.dense.bias": "model-00035-of-00046.safetensors",
432
+ "gpt_neox.layers.34.attention.dense.weight": "model-00035-of-00046.safetensors",
433
+ "gpt_neox.layers.34.attention.masked_bias": "model-00035-of-00046.safetensors",
434
+ "gpt_neox.layers.34.attention.query_key_value.bias": "model-00035-of-00046.safetensors",
435
+ "gpt_neox.layers.34.attention.query_key_value.weight": "model-00035-of-00046.safetensors",
436
+ "gpt_neox.layers.34.attention.rotary_emb.inv_freq": "model-00035-of-00046.safetensors",
437
+ "gpt_neox.layers.34.input_layernorm.bias": "model-00035-of-00046.safetensors",
438
+ "gpt_neox.layers.34.input_layernorm.weight": "model-00035-of-00046.safetensors",
439
+ "gpt_neox.layers.34.mlp.dense_4h_to_h.bias": "model-00036-of-00046.safetensors",
440
+ "gpt_neox.layers.34.mlp.dense_4h_to_h.weight": "model-00036-of-00046.safetensors",
441
+ "gpt_neox.layers.34.mlp.dense_h_to_4h.bias": "model-00036-of-00046.safetensors",
442
+ "gpt_neox.layers.34.mlp.dense_h_to_4h.weight": "model-00036-of-00046.safetensors",
443
+ "gpt_neox.layers.34.post_attention_layernorm.bias": "model-00035-of-00046.safetensors",
444
+ "gpt_neox.layers.34.post_attention_layernorm.weight": "model-00035-of-00046.safetensors",
445
+ "gpt_neox.layers.35.attention.bias": "model-00036-of-00046.safetensors",
446
+ "gpt_neox.layers.35.attention.dense.bias": "model-00036-of-00046.safetensors",
447
+ "gpt_neox.layers.35.attention.dense.weight": "model-00036-of-00046.safetensors",
448
+ "gpt_neox.layers.35.attention.masked_bias": "model-00036-of-00046.safetensors",
449
+ "gpt_neox.layers.35.attention.query_key_value.bias": "model-00036-of-00046.safetensors",
450
+ "gpt_neox.layers.35.attention.query_key_value.weight": "model-00036-of-00046.safetensors",
451
+ "gpt_neox.layers.35.attention.rotary_emb.inv_freq": "model-00036-of-00046.safetensors",
452
+ "gpt_neox.layers.35.input_layernorm.bias": "model-00036-of-00046.safetensors",
453
+ "gpt_neox.layers.35.input_layernorm.weight": "model-00036-of-00046.safetensors",
454
+ "gpt_neox.layers.35.mlp.dense_4h_to_h.bias": "model-00037-of-00046.safetensors",
455
+ "gpt_neox.layers.35.mlp.dense_4h_to_h.weight": "model-00037-of-00046.safetensors",
456
+ "gpt_neox.layers.35.mlp.dense_h_to_4h.bias": "model-00037-of-00046.safetensors",
457
+ "gpt_neox.layers.35.mlp.dense_h_to_4h.weight": "model-00037-of-00046.safetensors",
458
+ "gpt_neox.layers.35.post_attention_layernorm.bias": "model-00036-of-00046.safetensors",
459
+ "gpt_neox.layers.35.post_attention_layernorm.weight": "model-00036-of-00046.safetensors",
460
+ "gpt_neox.layers.36.attention.bias": "model-00037-of-00046.safetensors",
461
+ "gpt_neox.layers.36.attention.dense.bias": "model-00037-of-00046.safetensors",
462
+ "gpt_neox.layers.36.attention.dense.weight": "model-00037-of-00046.safetensors",
463
+ "gpt_neox.layers.36.attention.masked_bias": "model-00037-of-00046.safetensors",
464
+ "gpt_neox.layers.36.attention.query_key_value.bias": "model-00037-of-00046.safetensors",
465
+ "gpt_neox.layers.36.attention.query_key_value.weight": "model-00037-of-00046.safetensors",
466
+ "gpt_neox.layers.36.attention.rotary_emb.inv_freq": "model-00037-of-00046.safetensors",
467
+ "gpt_neox.layers.36.input_layernorm.bias": "model-00037-of-00046.safetensors",
468
+ "gpt_neox.layers.36.input_layernorm.weight": "model-00037-of-00046.safetensors",
469
+ "gpt_neox.layers.36.mlp.dense_4h_to_h.bias": "model-00038-of-00046.safetensors",
470
+ "gpt_neox.layers.36.mlp.dense_4h_to_h.weight": "model-00038-of-00046.safetensors",
471
+ "gpt_neox.layers.36.mlp.dense_h_to_4h.bias": "model-00038-of-00046.safetensors",
472
+ "gpt_neox.layers.36.mlp.dense_h_to_4h.weight": "model-00038-of-00046.safetensors",
473
+ "gpt_neox.layers.36.post_attention_layernorm.bias": "model-00037-of-00046.safetensors",
474
+ "gpt_neox.layers.36.post_attention_layernorm.weight": "model-00037-of-00046.safetensors",
475
+ "gpt_neox.layers.37.attention.bias": "model-00038-of-00046.safetensors",
476
+ "gpt_neox.layers.37.attention.dense.bias": "model-00038-of-00046.safetensors",
477
+ "gpt_neox.layers.37.attention.dense.weight": "model-00038-of-00046.safetensors",
478
+ "gpt_neox.layers.37.attention.masked_bias": "model-00038-of-00046.safetensors",
479
+ "gpt_neox.layers.37.attention.query_key_value.bias": "model-00038-of-00046.safetensors",
480
+ "gpt_neox.layers.37.attention.query_key_value.weight": "model-00038-of-00046.safetensors",
481
+ "gpt_neox.layers.37.attention.rotary_emb.inv_freq": "model-00038-of-00046.safetensors",
482
+ "gpt_neox.layers.37.input_layernorm.bias": "model-00038-of-00046.safetensors",
483
+ "gpt_neox.layers.37.input_layernorm.weight": "model-00038-of-00046.safetensors",
484
+ "gpt_neox.layers.37.mlp.dense_4h_to_h.bias": "model-00039-of-00046.safetensors",
485
+ "gpt_neox.layers.37.mlp.dense_4h_to_h.weight": "model-00039-of-00046.safetensors",
486
+ "gpt_neox.layers.37.mlp.dense_h_to_4h.bias": "model-00039-of-00046.safetensors",
487
+ "gpt_neox.layers.37.mlp.dense_h_to_4h.weight": "model-00039-of-00046.safetensors",
488
+ "gpt_neox.layers.37.post_attention_layernorm.bias": "model-00038-of-00046.safetensors",
489
+ "gpt_neox.layers.37.post_attention_layernorm.weight": "model-00038-of-00046.safetensors",
490
+ "gpt_neox.layers.38.attention.bias": "model-00039-of-00046.safetensors",
491
+ "gpt_neox.layers.38.attention.dense.bias": "model-00039-of-00046.safetensors",
492
+ "gpt_neox.layers.38.attention.dense.weight": "model-00039-of-00046.safetensors",
493
+ "gpt_neox.layers.38.attention.masked_bias": "model-00039-of-00046.safetensors",
494
+ "gpt_neox.layers.38.attention.query_key_value.bias": "model-00039-of-00046.safetensors",
495
+ "gpt_neox.layers.38.attention.query_key_value.weight": "model-00039-of-00046.safetensors",
496
+ "gpt_neox.layers.38.attention.rotary_emb.inv_freq": "model-00039-of-00046.safetensors",
497
+ "gpt_neox.layers.38.input_layernorm.bias": "model-00039-of-00046.safetensors",
498
+ "gpt_neox.layers.38.input_layernorm.weight": "model-00039-of-00046.safetensors",
499
+ "gpt_neox.layers.38.mlp.dense_4h_to_h.bias": "model-00040-of-00046.safetensors",
500
+ "gpt_neox.layers.38.mlp.dense_4h_to_h.weight": "model-00040-of-00046.safetensors",
501
+ "gpt_neox.layers.38.mlp.dense_h_to_4h.bias": "model-00040-of-00046.safetensors",
502
+ "gpt_neox.layers.38.mlp.dense_h_to_4h.weight": "model-00040-of-00046.safetensors",
503
+ "gpt_neox.layers.38.post_attention_layernorm.bias": "model-00039-of-00046.safetensors",
504
+ "gpt_neox.layers.38.post_attention_layernorm.weight": "model-00039-of-00046.safetensors",
505
+ "gpt_neox.layers.39.attention.bias": "model-00040-of-00046.safetensors",
506
+ "gpt_neox.layers.39.attention.dense.bias": "model-00040-of-00046.safetensors",
507
+ "gpt_neox.layers.39.attention.dense.weight": "model-00040-of-00046.safetensors",
508
+ "gpt_neox.layers.39.attention.masked_bias": "model-00040-of-00046.safetensors",
509
+ "gpt_neox.layers.39.attention.query_key_value.bias": "model-00040-of-00046.safetensors",
510
+ "gpt_neox.layers.39.attention.query_key_value.weight": "model-00040-of-00046.safetensors",
511
+ "gpt_neox.layers.39.attention.rotary_emb.inv_freq": "model-00040-of-00046.safetensors",
512
+ "gpt_neox.layers.39.input_layernorm.bias": "model-00040-of-00046.safetensors",
513
+ "gpt_neox.layers.39.input_layernorm.weight": "model-00040-of-00046.safetensors",
514
+ "gpt_neox.layers.39.mlp.dense_4h_to_h.bias": "model-00041-of-00046.safetensors",
515
+ "gpt_neox.layers.39.mlp.dense_4h_to_h.weight": "model-00041-of-00046.safetensors",
516
+ "gpt_neox.layers.39.mlp.dense_h_to_4h.bias": "model-00041-of-00046.safetensors",
517
+ "gpt_neox.layers.39.mlp.dense_h_to_4h.weight": "model-00041-of-00046.safetensors",
518
+ "gpt_neox.layers.39.post_attention_layernorm.bias": "model-00040-of-00046.safetensors",
519
+ "gpt_neox.layers.39.post_attention_layernorm.weight": "model-00040-of-00046.safetensors",
520
+ "gpt_neox.layers.4.attention.bias": "model-00005-of-00046.safetensors",
521
+ "gpt_neox.layers.4.attention.dense.bias": "model-00005-of-00046.safetensors",
522
+ "gpt_neox.layers.4.attention.dense.weight": "model-00005-of-00046.safetensors",
523
+ "gpt_neox.layers.4.attention.masked_bias": "model-00005-of-00046.safetensors",
524
+ "gpt_neox.layers.4.attention.query_key_value.bias": "model-00005-of-00046.safetensors",
525
+ "gpt_neox.layers.4.attention.query_key_value.weight": "model-00005-of-00046.safetensors",
526
+ "gpt_neox.layers.4.attention.rotary_emb.inv_freq": "model-00005-of-00046.safetensors",
527
+ "gpt_neox.layers.4.input_layernorm.bias": "model-00005-of-00046.safetensors",
528
+ "gpt_neox.layers.4.input_layernorm.weight": "model-00005-of-00046.safetensors",
529
+ "gpt_neox.layers.4.mlp.dense_4h_to_h.bias": "model-00006-of-00046.safetensors",
530
+ "gpt_neox.layers.4.mlp.dense_4h_to_h.weight": "model-00006-of-00046.safetensors",
531
+ "gpt_neox.layers.4.mlp.dense_h_to_4h.bias": "model-00006-of-00046.safetensors",
532
+ "gpt_neox.layers.4.mlp.dense_h_to_4h.weight": "model-00006-of-00046.safetensors",
533
+ "gpt_neox.layers.4.post_attention_layernorm.bias": "model-00005-of-00046.safetensors",
534
+ "gpt_neox.layers.4.post_attention_layernorm.weight": "model-00005-of-00046.safetensors",
535
+ "gpt_neox.layers.40.attention.bias": "model-00041-of-00046.safetensors",
536
+ "gpt_neox.layers.40.attention.dense.bias": "model-00041-of-00046.safetensors",
537
+ "gpt_neox.layers.40.attention.dense.weight": "model-00041-of-00046.safetensors",
538
+ "gpt_neox.layers.40.attention.masked_bias": "model-00041-of-00046.safetensors",
539
+ "gpt_neox.layers.40.attention.query_key_value.bias": "model-00041-of-00046.safetensors",
540
+ "gpt_neox.layers.40.attention.query_key_value.weight": "model-00041-of-00046.safetensors",
541
+ "gpt_neox.layers.40.attention.rotary_emb.inv_freq": "model-00041-of-00046.safetensors",
542
+ "gpt_neox.layers.40.input_layernorm.bias": "model-00041-of-00046.safetensors",
543
+ "gpt_neox.layers.40.input_layernorm.weight": "model-00041-of-00046.safetensors",
544
+ "gpt_neox.layers.40.mlp.dense_4h_to_h.bias": "model-00042-of-00046.safetensors",
545
+ "gpt_neox.layers.40.mlp.dense_4h_to_h.weight": "model-00042-of-00046.safetensors",
546
+ "gpt_neox.layers.40.mlp.dense_h_to_4h.bias": "model-00042-of-00046.safetensors",
547
+ "gpt_neox.layers.40.mlp.dense_h_to_4h.weight": "model-00042-of-00046.safetensors",
548
+ "gpt_neox.layers.40.post_attention_layernorm.bias": "model-00041-of-00046.safetensors",
549
+ "gpt_neox.layers.40.post_attention_layernorm.weight": "model-00041-of-00046.safetensors",
550
+ "gpt_neox.layers.41.attention.bias": "model-00042-of-00046.safetensors",
551
+ "gpt_neox.layers.41.attention.dense.bias": "model-00042-of-00046.safetensors",
552
+ "gpt_neox.layers.41.attention.dense.weight": "model-00042-of-00046.safetensors",
553
+ "gpt_neox.layers.41.attention.masked_bias": "model-00042-of-00046.safetensors",
554
+ "gpt_neox.layers.41.attention.query_key_value.bias": "model-00042-of-00046.safetensors",
555
+ "gpt_neox.layers.41.attention.query_key_value.weight": "model-00042-of-00046.safetensors",
556
+ "gpt_neox.layers.41.attention.rotary_emb.inv_freq": "model-00042-of-00046.safetensors",
557
+ "gpt_neox.layers.41.input_layernorm.bias": "model-00042-of-00046.safetensors",
558
+ "gpt_neox.layers.41.input_layernorm.weight": "model-00042-of-00046.safetensors",
559
+ "gpt_neox.layers.41.mlp.dense_4h_to_h.bias": "model-00043-of-00046.safetensors",
560
+ "gpt_neox.layers.41.mlp.dense_4h_to_h.weight": "model-00043-of-00046.safetensors",
561
+ "gpt_neox.layers.41.mlp.dense_h_to_4h.bias": "model-00043-of-00046.safetensors",
562
+ "gpt_neox.layers.41.mlp.dense_h_to_4h.weight": "model-00043-of-00046.safetensors",
563
+ "gpt_neox.layers.41.post_attention_layernorm.bias": "model-00042-of-00046.safetensors",
564
+ "gpt_neox.layers.41.post_attention_layernorm.weight": "model-00042-of-00046.safetensors",
565
+ "gpt_neox.layers.42.attention.bias": "model-00043-of-00046.safetensors",
566
+ "gpt_neox.layers.42.attention.dense.bias": "model-00043-of-00046.safetensors",
567
+ "gpt_neox.layers.42.attention.dense.weight": "model-00043-of-00046.safetensors",
568
+ "gpt_neox.layers.42.attention.masked_bias": "model-00043-of-00046.safetensors",
569
+ "gpt_neox.layers.42.attention.query_key_value.bias": "model-00043-of-00046.safetensors",
570
+ "gpt_neox.layers.42.attention.query_key_value.weight": "model-00043-of-00046.safetensors",
571
+ "gpt_neox.layers.42.attention.rotary_emb.inv_freq": "model-00043-of-00046.safetensors",
572
+ "gpt_neox.layers.42.input_layernorm.bias": "model-00043-of-00046.safetensors",
573
+ "gpt_neox.layers.42.input_layernorm.weight": "model-00043-of-00046.safetensors",
574
+ "gpt_neox.layers.42.mlp.dense_4h_to_h.bias": "model-00044-of-00046.safetensors",
575
+ "gpt_neox.layers.42.mlp.dense_4h_to_h.weight": "model-00044-of-00046.safetensors",
576
+ "gpt_neox.layers.42.mlp.dense_h_to_4h.bias": "model-00044-of-00046.safetensors",
577
+ "gpt_neox.layers.42.mlp.dense_h_to_4h.weight": "model-00044-of-00046.safetensors",
578
+ "gpt_neox.layers.42.post_attention_layernorm.bias": "model-00043-of-00046.safetensors",
579
+ "gpt_neox.layers.42.post_attention_layernorm.weight": "model-00043-of-00046.safetensors",
580
+ "gpt_neox.layers.43.attention.bias": "model-00044-of-00046.safetensors",
581
+ "gpt_neox.layers.43.attention.dense.bias": "model-00044-of-00046.safetensors",
582
+ "gpt_neox.layers.43.attention.dense.weight": "model-00044-of-00046.safetensors",
583
+ "gpt_neox.layers.43.attention.masked_bias": "model-00044-of-00046.safetensors",
584
+ "gpt_neox.layers.43.attention.query_key_value.bias": "model-00044-of-00046.safetensors",
585
+ "gpt_neox.layers.43.attention.query_key_value.weight": "model-00044-of-00046.safetensors",
586
+ "gpt_neox.layers.43.attention.rotary_emb.inv_freq": "model-00044-of-00046.safetensors",
587
+ "gpt_neox.layers.43.input_layernorm.bias": "model-00044-of-00046.safetensors",
588
+ "gpt_neox.layers.43.input_layernorm.weight": "model-00044-of-00046.safetensors",
589
+ "gpt_neox.layers.43.mlp.dense_4h_to_h.bias": "model-00045-of-00046.safetensors",
590
+ "gpt_neox.layers.43.mlp.dense_4h_to_h.weight": "model-00045-of-00046.safetensors",
591
+ "gpt_neox.layers.43.mlp.dense_h_to_4h.bias": "model-00045-of-00046.safetensors",
592
+ "gpt_neox.layers.43.mlp.dense_h_to_4h.weight": "model-00045-of-00046.safetensors",
593
+ "gpt_neox.layers.43.post_attention_layernorm.bias": "model-00044-of-00046.safetensors",
594
+ "gpt_neox.layers.43.post_attention_layernorm.weight": "model-00044-of-00046.safetensors",
595
+ "gpt_neox.layers.5.attention.bias": "model-00006-of-00046.safetensors",
596
+ "gpt_neox.layers.5.attention.dense.bias": "model-00006-of-00046.safetensors",
597
+ "gpt_neox.layers.5.attention.dense.weight": "model-00006-of-00046.safetensors",
598
+ "gpt_neox.layers.5.attention.masked_bias": "model-00006-of-00046.safetensors",
599
+ "gpt_neox.layers.5.attention.query_key_value.bias": "model-00006-of-00046.safetensors",
600
+ "gpt_neox.layers.5.attention.query_key_value.weight": "model-00006-of-00046.safetensors",
601
+ "gpt_neox.layers.5.attention.rotary_emb.inv_freq": "model-00006-of-00046.safetensors",
602
+ "gpt_neox.layers.5.input_layernorm.bias": "model-00006-of-00046.safetensors",
603
+ "gpt_neox.layers.5.input_layernorm.weight": "model-00006-of-00046.safetensors",
604
+ "gpt_neox.layers.5.mlp.dense_4h_to_h.bias": "model-00007-of-00046.safetensors",
605
+ "gpt_neox.layers.5.mlp.dense_4h_to_h.weight": "model-00007-of-00046.safetensors",
606
+ "gpt_neox.layers.5.mlp.dense_h_to_4h.bias": "model-00007-of-00046.safetensors",
607
+ "gpt_neox.layers.5.mlp.dense_h_to_4h.weight": "model-00007-of-00046.safetensors",
608
+ "gpt_neox.layers.5.post_attention_layernorm.bias": "model-00006-of-00046.safetensors",
609
+ "gpt_neox.layers.5.post_attention_layernorm.weight": "model-00006-of-00046.safetensors",
610
+ "gpt_neox.layers.6.attention.bias": "model-00007-of-00046.safetensors",
611
+ "gpt_neox.layers.6.attention.dense.bias": "model-00007-of-00046.safetensors",
612
+ "gpt_neox.layers.6.attention.dense.weight": "model-00007-of-00046.safetensors",
613
+ "gpt_neox.layers.6.attention.masked_bias": "model-00007-of-00046.safetensors",
614
+ "gpt_neox.layers.6.attention.query_key_value.bias": "model-00007-of-00046.safetensors",
615
+ "gpt_neox.layers.6.attention.query_key_value.weight": "model-00007-of-00046.safetensors",
616
+ "gpt_neox.layers.6.attention.rotary_emb.inv_freq": "model-00007-of-00046.safetensors",
617
+ "gpt_neox.layers.6.input_layernorm.bias": "model-00007-of-00046.safetensors",
618
+ "gpt_neox.layers.6.input_layernorm.weight": "model-00007-of-00046.safetensors",
619
+ "gpt_neox.layers.6.mlp.dense_4h_to_h.bias": "model-00008-of-00046.safetensors",
620
+ "gpt_neox.layers.6.mlp.dense_4h_to_h.weight": "model-00008-of-00046.safetensors",
621
+ "gpt_neox.layers.6.mlp.dense_h_to_4h.bias": "model-00008-of-00046.safetensors",
622
+ "gpt_neox.layers.6.mlp.dense_h_to_4h.weight": "model-00008-of-00046.safetensors",
623
+ "gpt_neox.layers.6.post_attention_layernorm.bias": "model-00007-of-00046.safetensors",
624
+ "gpt_neox.layers.6.post_attention_layernorm.weight": "model-00007-of-00046.safetensors",
625
+ "gpt_neox.layers.7.attention.bias": "model-00008-of-00046.safetensors",
626
+ "gpt_neox.layers.7.attention.dense.bias": "model-00008-of-00046.safetensors",
627
+ "gpt_neox.layers.7.attention.dense.weight": "model-00008-of-00046.safetensors",
628
+ "gpt_neox.layers.7.attention.masked_bias": "model-00008-of-00046.safetensors",
629
+ "gpt_neox.layers.7.attention.query_key_value.bias": "model-00008-of-00046.safetensors",
630
+ "gpt_neox.layers.7.attention.query_key_value.weight": "model-00008-of-00046.safetensors",
631
+ "gpt_neox.layers.7.attention.rotary_emb.inv_freq": "model-00008-of-00046.safetensors",
632
+ "gpt_neox.layers.7.input_layernorm.bias": "model-00008-of-00046.safetensors",
633
+ "gpt_neox.layers.7.input_layernorm.weight": "model-00008-of-00046.safetensors",
634
+ "gpt_neox.layers.7.mlp.dense_4h_to_h.bias": "model-00009-of-00046.safetensors",
635
+ "gpt_neox.layers.7.mlp.dense_4h_to_h.weight": "model-00009-of-00046.safetensors",
636
+ "gpt_neox.layers.7.mlp.dense_h_to_4h.bias": "model-00009-of-00046.safetensors",
637
+ "gpt_neox.layers.7.mlp.dense_h_to_4h.weight": "model-00009-of-00046.safetensors",
638
+ "gpt_neox.layers.7.post_attention_layernorm.bias": "model-00008-of-00046.safetensors",
639
+ "gpt_neox.layers.7.post_attention_layernorm.weight": "model-00008-of-00046.safetensors",
640
+ "gpt_neox.layers.8.attention.bias": "model-00009-of-00046.safetensors",
641
+ "gpt_neox.layers.8.attention.dense.bias": "model-00009-of-00046.safetensors",
642
+ "gpt_neox.layers.8.attention.dense.weight": "model-00009-of-00046.safetensors",
643
+ "gpt_neox.layers.8.attention.masked_bias": "model-00009-of-00046.safetensors",
644
+ "gpt_neox.layers.8.attention.query_key_value.bias": "model-00009-of-00046.safetensors",
645
+ "gpt_neox.layers.8.attention.query_key_value.weight": "model-00009-of-00046.safetensors",
646
+ "gpt_neox.layers.8.attention.rotary_emb.inv_freq": "model-00009-of-00046.safetensors",
647
+ "gpt_neox.layers.8.input_layernorm.bias": "model-00009-of-00046.safetensors",
648
+ "gpt_neox.layers.8.input_layernorm.weight": "model-00009-of-00046.safetensors",
649
+ "gpt_neox.layers.8.mlp.dense_4h_to_h.bias": "model-00010-of-00046.safetensors",
650
+ "gpt_neox.layers.8.mlp.dense_4h_to_h.weight": "model-00010-of-00046.safetensors",
651
+ "gpt_neox.layers.8.mlp.dense_h_to_4h.bias": "model-00010-of-00046.safetensors",
652
+ "gpt_neox.layers.8.mlp.dense_h_to_4h.weight": "model-00010-of-00046.safetensors",
653
+ "gpt_neox.layers.8.post_attention_layernorm.bias": "model-00009-of-00046.safetensors",
654
+ "gpt_neox.layers.8.post_attention_layernorm.weight": "model-00009-of-00046.safetensors",
655
+ "gpt_neox.layers.9.attention.bias": "model-00010-of-00046.safetensors",
656
+ "gpt_neox.layers.9.attention.dense.bias": "model-00010-of-00046.safetensors",
657
+ "gpt_neox.layers.9.attention.dense.weight": "model-00010-of-00046.safetensors",
658
+ "gpt_neox.layers.9.attention.masked_bias": "model-00010-of-00046.safetensors",
659
+ "gpt_neox.layers.9.attention.query_key_value.bias": "model-00010-of-00046.safetensors",
660
+ "gpt_neox.layers.9.attention.query_key_value.weight": "model-00010-of-00046.safetensors",
661
+ "gpt_neox.layers.9.attention.rotary_emb.inv_freq": "model-00010-of-00046.safetensors",
662
+ "gpt_neox.layers.9.input_layernorm.bias": "model-00010-of-00046.safetensors",
663
+ "gpt_neox.layers.9.input_layernorm.weight": "model-00010-of-00046.safetensors",
664
+ "gpt_neox.layers.9.mlp.dense_4h_to_h.bias": "model-00011-of-00046.safetensors",
665
+ "gpt_neox.layers.9.mlp.dense_4h_to_h.weight": "model-00011-of-00046.safetensors",
666
+ "gpt_neox.layers.9.mlp.dense_h_to_4h.bias": "model-00011-of-00046.safetensors",
667
+ "gpt_neox.layers.9.mlp.dense_h_to_4h.weight": "model-00011-of-00046.safetensors",
668
+ "gpt_neox.layers.9.post_attention_layernorm.bias": "model-00010-of-00046.safetensors",
669
+ "gpt_neox.layers.9.post_attention_layernorm.weight": "model-00010-of-00046.safetensors"
670
+ }
671
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:406433211f34a36e63ba1c54c2b8997c39b14d9b0c5a31a40622a74e11f8d992
3
+ size 17436485
pytorch_model.bin.index.json ADDED
@@ -0,0 +1,671 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 41293685880
4
+ },
5
+ "weight_map": {
6
+ "embed_out.weight": "pytorch_model-00046-of-00046.bin",
7
+ "gpt_neox.embed_in.weight": "pytorch_model-00001-of-00046.bin",
8
+ "gpt_neox.final_layer_norm.bias": "pytorch_model-00045-of-00046.bin",
9
+ "gpt_neox.final_layer_norm.weight": "pytorch_model-00045-of-00046.bin",
10
+ "gpt_neox.layers.0.attention.bias": "pytorch_model-00001-of-00046.bin",
11
+ "gpt_neox.layers.0.attention.dense.bias": "pytorch_model-00001-of-00046.bin",
12
+ "gpt_neox.layers.0.attention.dense.weight": "pytorch_model-00001-of-00046.bin",
13
+ "gpt_neox.layers.0.attention.masked_bias": "pytorch_model-00001-of-00046.bin",
14
+ "gpt_neox.layers.0.attention.query_key_value.bias": "pytorch_model-00001-of-00046.bin",
15
+ "gpt_neox.layers.0.attention.query_key_value.weight": "pytorch_model-00001-of-00046.bin",
16
+ "gpt_neox.layers.0.attention.rotary_emb.inv_freq": "pytorch_model-00001-of-00046.bin",
17
+ "gpt_neox.layers.0.input_layernorm.bias": "pytorch_model-00001-of-00046.bin",
18
+ "gpt_neox.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00046.bin",
19
+ "gpt_neox.layers.0.mlp.dense_4h_to_h.bias": "pytorch_model-00002-of-00046.bin",
20
+ "gpt_neox.layers.0.mlp.dense_4h_to_h.weight": "pytorch_model-00002-of-00046.bin",
21
+ "gpt_neox.layers.0.mlp.dense_h_to_4h.bias": "pytorch_model-00002-of-00046.bin",
22
+ "gpt_neox.layers.0.mlp.dense_h_to_4h.weight": "pytorch_model-00002-of-00046.bin",
23
+ "gpt_neox.layers.0.post_attention_layernorm.bias": "pytorch_model-00001-of-00046.bin",
24
+ "gpt_neox.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00046.bin",
25
+ "gpt_neox.layers.1.attention.bias": "pytorch_model-00002-of-00046.bin",
26
+ "gpt_neox.layers.1.attention.dense.bias": "pytorch_model-00002-of-00046.bin",
27
+ "gpt_neox.layers.1.attention.dense.weight": "pytorch_model-00002-of-00046.bin",
28
+ "gpt_neox.layers.1.attention.masked_bias": "pytorch_model-00002-of-00046.bin",
29
+ "gpt_neox.layers.1.attention.query_key_value.bias": "pytorch_model-00002-of-00046.bin",
30
+ "gpt_neox.layers.1.attention.query_key_value.weight": "pytorch_model-00002-of-00046.bin",
31
+ "gpt_neox.layers.1.attention.rotary_emb.inv_freq": "pytorch_model-00002-of-00046.bin",
32
+ "gpt_neox.layers.1.input_layernorm.bias": "pytorch_model-00002-of-00046.bin",
33
+ "gpt_neox.layers.1.input_layernorm.weight": "pytorch_model-00002-of-00046.bin",
34
+ "gpt_neox.layers.1.mlp.dense_4h_to_h.bias": "pytorch_model-00003-of-00046.bin",
35
+ "gpt_neox.layers.1.mlp.dense_4h_to_h.weight": "pytorch_model-00003-of-00046.bin",
36
+ "gpt_neox.layers.1.mlp.dense_h_to_4h.bias": "pytorch_model-00003-of-00046.bin",
37
+ "gpt_neox.layers.1.mlp.dense_h_to_4h.weight": "pytorch_model-00003-of-00046.bin",
38
+ "gpt_neox.layers.1.post_attention_layernorm.bias": "pytorch_model-00002-of-00046.bin",
39
+ "gpt_neox.layers.1.post_attention_layernorm.weight": "pytorch_model-00002-of-00046.bin",
40
+ "gpt_neox.layers.10.attention.bias": "pytorch_model-00011-of-00046.bin",
41
+ "gpt_neox.layers.10.attention.dense.bias": "pytorch_model-00011-of-00046.bin",
42
+ "gpt_neox.layers.10.attention.dense.weight": "pytorch_model-00011-of-00046.bin",
43
+ "gpt_neox.layers.10.attention.masked_bias": "pytorch_model-00011-of-00046.bin",
44
+ "gpt_neox.layers.10.attention.query_key_value.bias": "pytorch_model-00011-of-00046.bin",
45
+ "gpt_neox.layers.10.attention.query_key_value.weight": "pytorch_model-00011-of-00046.bin",
46
+ "gpt_neox.layers.10.attention.rotary_emb.inv_freq": "pytorch_model-00011-of-00046.bin",
47
+ "gpt_neox.layers.10.input_layernorm.bias": "pytorch_model-00011-of-00046.bin",
48
+ "gpt_neox.layers.10.input_layernorm.weight": "pytorch_model-00011-of-00046.bin",
49
+ "gpt_neox.layers.10.mlp.dense_4h_to_h.bias": "pytorch_model-00012-of-00046.bin",
50
+ "gpt_neox.layers.10.mlp.dense_4h_to_h.weight": "pytorch_model-00012-of-00046.bin",
51
+ "gpt_neox.layers.10.mlp.dense_h_to_4h.bias": "pytorch_model-00012-of-00046.bin",
52
+ "gpt_neox.layers.10.mlp.dense_h_to_4h.weight": "pytorch_model-00012-of-00046.bin",
53
+ "gpt_neox.layers.10.post_attention_layernorm.bias": "pytorch_model-00011-of-00046.bin",
54
+ "gpt_neox.layers.10.post_attention_layernorm.weight": "pytorch_model-00011-of-00046.bin",
55
+ "gpt_neox.layers.11.attention.bias": "pytorch_model-00012-of-00046.bin",
56
+ "gpt_neox.layers.11.attention.dense.bias": "pytorch_model-00012-of-00046.bin",
57
+ "gpt_neox.layers.11.attention.dense.weight": "pytorch_model-00012-of-00046.bin",
58
+ "gpt_neox.layers.11.attention.masked_bias": "pytorch_model-00012-of-00046.bin",
59
+ "gpt_neox.layers.11.attention.query_key_value.bias": "pytorch_model-00012-of-00046.bin",
60
+ "gpt_neox.layers.11.attention.query_key_value.weight": "pytorch_model-00012-of-00046.bin",
61
+ "gpt_neox.layers.11.attention.rotary_emb.inv_freq": "pytorch_model-00012-of-00046.bin",
62
+ "gpt_neox.layers.11.input_layernorm.bias": "pytorch_model-00012-of-00046.bin",
63
+ "gpt_neox.layers.11.input_layernorm.weight": "pytorch_model-00012-of-00046.bin",
64
+ "gpt_neox.layers.11.mlp.dense_4h_to_h.bias": "pytorch_model-00013-of-00046.bin",
65
+ "gpt_neox.layers.11.mlp.dense_4h_to_h.weight": "pytorch_model-00013-of-00046.bin",
66
+ "gpt_neox.layers.11.mlp.dense_h_to_4h.bias": "pytorch_model-00013-of-00046.bin",
67
+ "gpt_neox.layers.11.mlp.dense_h_to_4h.weight": "pytorch_model-00013-of-00046.bin",
68
+ "gpt_neox.layers.11.post_attention_layernorm.bias": "pytorch_model-00012-of-00046.bin",
69
+ "gpt_neox.layers.11.post_attention_layernorm.weight": "pytorch_model-00012-of-00046.bin",
70
+ "gpt_neox.layers.12.attention.bias": "pytorch_model-00013-of-00046.bin",
71
+ "gpt_neox.layers.12.attention.dense.bias": "pytorch_model-00013-of-00046.bin",
72
+ "gpt_neox.layers.12.attention.dense.weight": "pytorch_model-00013-of-00046.bin",
73
+ "gpt_neox.layers.12.attention.masked_bias": "pytorch_model-00013-of-00046.bin",
74
+ "gpt_neox.layers.12.attention.query_key_value.bias": "pytorch_model-00013-of-00046.bin",
75
+ "gpt_neox.layers.12.attention.query_key_value.weight": "pytorch_model-00013-of-00046.bin",
76
+ "gpt_neox.layers.12.attention.rotary_emb.inv_freq": "pytorch_model-00013-of-00046.bin",
77
+ "gpt_neox.layers.12.input_layernorm.bias": "pytorch_model-00013-of-00046.bin",
78
+ "gpt_neox.layers.12.input_layernorm.weight": "pytorch_model-00013-of-00046.bin",
79
+ "gpt_neox.layers.12.mlp.dense_4h_to_h.bias": "pytorch_model-00014-of-00046.bin",
80
+ "gpt_neox.layers.12.mlp.dense_4h_to_h.weight": "pytorch_model-00014-of-00046.bin",
81
+ "gpt_neox.layers.12.mlp.dense_h_to_4h.bias": "pytorch_model-00014-of-00046.bin",
82
+ "gpt_neox.layers.12.mlp.dense_h_to_4h.weight": "pytorch_model-00014-of-00046.bin",
83
+ "gpt_neox.layers.12.post_attention_layernorm.bias": "pytorch_model-00013-of-00046.bin",
84
+ "gpt_neox.layers.12.post_attention_layernorm.weight": "pytorch_model-00013-of-00046.bin",
85
+ "gpt_neox.layers.13.attention.bias": "pytorch_model-00014-of-00046.bin",
86
+ "gpt_neox.layers.13.attention.dense.bias": "pytorch_model-00014-of-00046.bin",
87
+ "gpt_neox.layers.13.attention.dense.weight": "pytorch_model-00014-of-00046.bin",
88
+ "gpt_neox.layers.13.attention.masked_bias": "pytorch_model-00014-of-00046.bin",
89
+ "gpt_neox.layers.13.attention.query_key_value.bias": "pytorch_model-00014-of-00046.bin",
90
+ "gpt_neox.layers.13.attention.query_key_value.weight": "pytorch_model-00014-of-00046.bin",
91
+ "gpt_neox.layers.13.attention.rotary_emb.inv_freq": "pytorch_model-00014-of-00046.bin",
92
+ "gpt_neox.layers.13.input_layernorm.bias": "pytorch_model-00014-of-00046.bin",
93
+ "gpt_neox.layers.13.input_layernorm.weight": "pytorch_model-00014-of-00046.bin",
94
+ "gpt_neox.layers.13.mlp.dense_4h_to_h.bias": "pytorch_model-00015-of-00046.bin",
95
+ "gpt_neox.layers.13.mlp.dense_4h_to_h.weight": "pytorch_model-00015-of-00046.bin",
96
+ "gpt_neox.layers.13.mlp.dense_h_to_4h.bias": "pytorch_model-00015-of-00046.bin",
97
+ "gpt_neox.layers.13.mlp.dense_h_to_4h.weight": "pytorch_model-00015-of-00046.bin",
98
+ "gpt_neox.layers.13.post_attention_layernorm.bias": "pytorch_model-00014-of-00046.bin",
99
+ "gpt_neox.layers.13.post_attention_layernorm.weight": "pytorch_model-00014-of-00046.bin",
100
+ "gpt_neox.layers.14.attention.bias": "pytorch_model-00015-of-00046.bin",
101
+ "gpt_neox.layers.14.attention.dense.bias": "pytorch_model-00015-of-00046.bin",
102
+ "gpt_neox.layers.14.attention.dense.weight": "pytorch_model-00015-of-00046.bin",
103
+ "gpt_neox.layers.14.attention.masked_bias": "pytorch_model-00015-of-00046.bin",
104
+ "gpt_neox.layers.14.attention.query_key_value.bias": "pytorch_model-00015-of-00046.bin",
105
+ "gpt_neox.layers.14.attention.query_key_value.weight": "pytorch_model-00015-of-00046.bin",
106
+ "gpt_neox.layers.14.attention.rotary_emb.inv_freq": "pytorch_model-00015-of-00046.bin",
107
+ "gpt_neox.layers.14.input_layernorm.bias": "pytorch_model-00015-of-00046.bin",
108
+ "gpt_neox.layers.14.input_layernorm.weight": "pytorch_model-00015-of-00046.bin",
109
+ "gpt_neox.layers.14.mlp.dense_4h_to_h.bias": "pytorch_model-00016-of-00046.bin",
110
+ "gpt_neox.layers.14.mlp.dense_4h_to_h.weight": "pytorch_model-00016-of-00046.bin",
111
+ "gpt_neox.layers.14.mlp.dense_h_to_4h.bias": "pytorch_model-00016-of-00046.bin",
112
+ "gpt_neox.layers.14.mlp.dense_h_to_4h.weight": "pytorch_model-00016-of-00046.bin",
113
+ "gpt_neox.layers.14.post_attention_layernorm.bias": "pytorch_model-00015-of-00046.bin",
114
+ "gpt_neox.layers.14.post_attention_layernorm.weight": "pytorch_model-00015-of-00046.bin",
115
+ "gpt_neox.layers.15.attention.bias": "pytorch_model-00016-of-00046.bin",
116
+ "gpt_neox.layers.15.attention.dense.bias": "pytorch_model-00016-of-00046.bin",
117
+ "gpt_neox.layers.15.attention.dense.weight": "pytorch_model-00016-of-00046.bin",
118
+ "gpt_neox.layers.15.attention.masked_bias": "pytorch_model-00016-of-00046.bin",
119
+ "gpt_neox.layers.15.attention.query_key_value.bias": "pytorch_model-00016-of-00046.bin",
120
+ "gpt_neox.layers.15.attention.query_key_value.weight": "pytorch_model-00016-of-00046.bin",
121
+ "gpt_neox.layers.15.attention.rotary_emb.inv_freq": "pytorch_model-00016-of-00046.bin",
122
+ "gpt_neox.layers.15.input_layernorm.bias": "pytorch_model-00016-of-00046.bin",
123
+ "gpt_neox.layers.15.input_layernorm.weight": "pytorch_model-00016-of-00046.bin",
124
+ "gpt_neox.layers.15.mlp.dense_4h_to_h.bias": "pytorch_model-00017-of-00046.bin",
125
+ "gpt_neox.layers.15.mlp.dense_4h_to_h.weight": "pytorch_model-00017-of-00046.bin",
126
+ "gpt_neox.layers.15.mlp.dense_h_to_4h.bias": "pytorch_model-00017-of-00046.bin",
127
+ "gpt_neox.layers.15.mlp.dense_h_to_4h.weight": "pytorch_model-00017-of-00046.bin",
128
+ "gpt_neox.layers.15.post_attention_layernorm.bias": "pytorch_model-00016-of-00046.bin",
129
+ "gpt_neox.layers.15.post_attention_layernorm.weight": "pytorch_model-00016-of-00046.bin",
130
+ "gpt_neox.layers.16.attention.bias": "pytorch_model-00017-of-00046.bin",
131
+ "gpt_neox.layers.16.attention.dense.bias": "pytorch_model-00017-of-00046.bin",
132
+ "gpt_neox.layers.16.attention.dense.weight": "pytorch_model-00017-of-00046.bin",
133
+ "gpt_neox.layers.16.attention.masked_bias": "pytorch_model-00017-of-00046.bin",
134
+ "gpt_neox.layers.16.attention.query_key_value.bias": "pytorch_model-00017-of-00046.bin",
135
+ "gpt_neox.layers.16.attention.query_key_value.weight": "pytorch_model-00017-of-00046.bin",
136
+ "gpt_neox.layers.16.attention.rotary_emb.inv_freq": "pytorch_model-00017-of-00046.bin",
137
+ "gpt_neox.layers.16.input_layernorm.bias": "pytorch_model-00017-of-00046.bin",
138
+ "gpt_neox.layers.16.input_layernorm.weight": "pytorch_model-00017-of-00046.bin",
139
+ "gpt_neox.layers.16.mlp.dense_4h_to_h.bias": "pytorch_model-00018-of-00046.bin",
140
+ "gpt_neox.layers.16.mlp.dense_4h_to_h.weight": "pytorch_model-00018-of-00046.bin",
141
+ "gpt_neox.layers.16.mlp.dense_h_to_4h.bias": "pytorch_model-00018-of-00046.bin",
142
+ "gpt_neox.layers.16.mlp.dense_h_to_4h.weight": "pytorch_model-00018-of-00046.bin",
143
+ "gpt_neox.layers.16.post_attention_layernorm.bias": "pytorch_model-00017-of-00046.bin",
144
+ "gpt_neox.layers.16.post_attention_layernorm.weight": "pytorch_model-00017-of-00046.bin",
145
+ "gpt_neox.layers.17.attention.bias": "pytorch_model-00018-of-00046.bin",
146
+ "gpt_neox.layers.17.attention.dense.bias": "pytorch_model-00018-of-00046.bin",
147
+ "gpt_neox.layers.17.attention.dense.weight": "pytorch_model-00018-of-00046.bin",
148
+ "gpt_neox.layers.17.attention.masked_bias": "pytorch_model-00018-of-00046.bin",
149
+ "gpt_neox.layers.17.attention.query_key_value.bias": "pytorch_model-00018-of-00046.bin",
150
+ "gpt_neox.layers.17.attention.query_key_value.weight": "pytorch_model-00018-of-00046.bin",
151
+ "gpt_neox.layers.17.attention.rotary_emb.inv_freq": "pytorch_model-00018-of-00046.bin",
152
+ "gpt_neox.layers.17.input_layernorm.bias": "pytorch_model-00018-of-00046.bin",
153
+ "gpt_neox.layers.17.input_layernorm.weight": "pytorch_model-00018-of-00046.bin",
154
+ "gpt_neox.layers.17.mlp.dense_4h_to_h.bias": "pytorch_model-00019-of-00046.bin",
155
+ "gpt_neox.layers.17.mlp.dense_4h_to_h.weight": "pytorch_model-00019-of-00046.bin",
156
+ "gpt_neox.layers.17.mlp.dense_h_to_4h.bias": "pytorch_model-00019-of-00046.bin",
157
+ "gpt_neox.layers.17.mlp.dense_h_to_4h.weight": "pytorch_model-00019-of-00046.bin",
158
+ "gpt_neox.layers.17.post_attention_layernorm.bias": "pytorch_model-00018-of-00046.bin",
159
+ "gpt_neox.layers.17.post_attention_layernorm.weight": "pytorch_model-00018-of-00046.bin",
160
+ "gpt_neox.layers.18.attention.bias": "pytorch_model-00019-of-00046.bin",
161
+ "gpt_neox.layers.18.attention.dense.bias": "pytorch_model-00019-of-00046.bin",
162
+ "gpt_neox.layers.18.attention.dense.weight": "pytorch_model-00019-of-00046.bin",
163
+ "gpt_neox.layers.18.attention.masked_bias": "pytorch_model-00019-of-00046.bin",
164
+ "gpt_neox.layers.18.attention.query_key_value.bias": "pytorch_model-00019-of-00046.bin",
165
+ "gpt_neox.layers.18.attention.query_key_value.weight": "pytorch_model-00019-of-00046.bin",
166
+ "gpt_neox.layers.18.attention.rotary_emb.inv_freq": "pytorch_model-00019-of-00046.bin",
167
+ "gpt_neox.layers.18.input_layernorm.bias": "pytorch_model-00019-of-00046.bin",
168
+ "gpt_neox.layers.18.input_layernorm.weight": "pytorch_model-00019-of-00046.bin",
169
+ "gpt_neox.layers.18.mlp.dense_4h_to_h.bias": "pytorch_model-00020-of-00046.bin",
170
+ "gpt_neox.layers.18.mlp.dense_4h_to_h.weight": "pytorch_model-00020-of-00046.bin",
171
+ "gpt_neox.layers.18.mlp.dense_h_to_4h.bias": "pytorch_model-00020-of-00046.bin",
172
+ "gpt_neox.layers.18.mlp.dense_h_to_4h.weight": "pytorch_model-00020-of-00046.bin",
173
+ "gpt_neox.layers.18.post_attention_layernorm.bias": "pytorch_model-00019-of-00046.bin",
174
+ "gpt_neox.layers.18.post_attention_layernorm.weight": "pytorch_model-00019-of-00046.bin",
175
+ "gpt_neox.layers.19.attention.bias": "pytorch_model-00020-of-00046.bin",
176
+ "gpt_neox.layers.19.attention.dense.bias": "pytorch_model-00020-of-00046.bin",
177
+ "gpt_neox.layers.19.attention.dense.weight": "pytorch_model-00020-of-00046.bin",
178
+ "gpt_neox.layers.19.attention.masked_bias": "pytorch_model-00020-of-00046.bin",
179
+ "gpt_neox.layers.19.attention.query_key_value.bias": "pytorch_model-00020-of-00046.bin",
180
+ "gpt_neox.layers.19.attention.query_key_value.weight": "pytorch_model-00020-of-00046.bin",
181
+ "gpt_neox.layers.19.attention.rotary_emb.inv_freq": "pytorch_model-00020-of-00046.bin",
182
+ "gpt_neox.layers.19.input_layernorm.bias": "pytorch_model-00020-of-00046.bin",
183
+ "gpt_neox.layers.19.input_layernorm.weight": "pytorch_model-00020-of-00046.bin",
184
+ "gpt_neox.layers.19.mlp.dense_4h_to_h.bias": "pytorch_model-00021-of-00046.bin",
185
+ "gpt_neox.layers.19.mlp.dense_4h_to_h.weight": "pytorch_model-00021-of-00046.bin",
186
+ "gpt_neox.layers.19.mlp.dense_h_to_4h.bias": "pytorch_model-00021-of-00046.bin",
187
+ "gpt_neox.layers.19.mlp.dense_h_to_4h.weight": "pytorch_model-00021-of-00046.bin",
188
+ "gpt_neox.layers.19.post_attention_layernorm.bias": "pytorch_model-00020-of-00046.bin",
189
+ "gpt_neox.layers.19.post_attention_layernorm.weight": "pytorch_model-00020-of-00046.bin",
190
+ "gpt_neox.layers.2.attention.bias": "pytorch_model-00003-of-00046.bin",
191
+ "gpt_neox.layers.2.attention.dense.bias": "pytorch_model-00003-of-00046.bin",
192
+ "gpt_neox.layers.2.attention.dense.weight": "pytorch_model-00003-of-00046.bin",
193
+ "gpt_neox.layers.2.attention.masked_bias": "pytorch_model-00003-of-00046.bin",
194
+ "gpt_neox.layers.2.attention.query_key_value.bias": "pytorch_model-00003-of-00046.bin",
195
+ "gpt_neox.layers.2.attention.query_key_value.weight": "pytorch_model-00003-of-00046.bin",
196
+ "gpt_neox.layers.2.attention.rotary_emb.inv_freq": "pytorch_model-00003-of-00046.bin",
197
+ "gpt_neox.layers.2.input_layernorm.bias": "pytorch_model-00003-of-00046.bin",
198
+ "gpt_neox.layers.2.input_layernorm.weight": "pytorch_model-00003-of-00046.bin",
199
+ "gpt_neox.layers.2.mlp.dense_4h_to_h.bias": "pytorch_model-00004-of-00046.bin",
200
+ "gpt_neox.layers.2.mlp.dense_4h_to_h.weight": "pytorch_model-00004-of-00046.bin",
201
+ "gpt_neox.layers.2.mlp.dense_h_to_4h.bias": "pytorch_model-00004-of-00046.bin",
202
+ "gpt_neox.layers.2.mlp.dense_h_to_4h.weight": "pytorch_model-00004-of-00046.bin",
203
+ "gpt_neox.layers.2.post_attention_layernorm.bias": "pytorch_model-00003-of-00046.bin",
204
+ "gpt_neox.layers.2.post_attention_layernorm.weight": "pytorch_model-00003-of-00046.bin",
205
+ "gpt_neox.layers.20.attention.bias": "pytorch_model-00021-of-00046.bin",
206
+ "gpt_neox.layers.20.attention.dense.bias": "pytorch_model-00021-of-00046.bin",
207
+ "gpt_neox.layers.20.attention.dense.weight": "pytorch_model-00021-of-00046.bin",
208
+ "gpt_neox.layers.20.attention.masked_bias": "pytorch_model-00021-of-00046.bin",
209
+ "gpt_neox.layers.20.attention.query_key_value.bias": "pytorch_model-00021-of-00046.bin",
210
+ "gpt_neox.layers.20.attention.query_key_value.weight": "pytorch_model-00021-of-00046.bin",
211
+ "gpt_neox.layers.20.attention.rotary_emb.inv_freq": "pytorch_model-00021-of-00046.bin",
212
+ "gpt_neox.layers.20.input_layernorm.bias": "pytorch_model-00021-of-00046.bin",
213
+ "gpt_neox.layers.20.input_layernorm.weight": "pytorch_model-00021-of-00046.bin",
214
+ "gpt_neox.layers.20.mlp.dense_4h_to_h.bias": "pytorch_model-00022-of-00046.bin",
215
+ "gpt_neox.layers.20.mlp.dense_4h_to_h.weight": "pytorch_model-00022-of-00046.bin",
216
+ "gpt_neox.layers.20.mlp.dense_h_to_4h.bias": "pytorch_model-00022-of-00046.bin",
217
+ "gpt_neox.layers.20.mlp.dense_h_to_4h.weight": "pytorch_model-00022-of-00046.bin",
218
+ "gpt_neox.layers.20.post_attention_layernorm.bias": "pytorch_model-00021-of-00046.bin",
219
+ "gpt_neox.layers.20.post_attention_layernorm.weight": "pytorch_model-00021-of-00046.bin",
220
+ "gpt_neox.layers.21.attention.bias": "pytorch_model-00022-of-00046.bin",
221
+ "gpt_neox.layers.21.attention.dense.bias": "pytorch_model-00022-of-00046.bin",
222
+ "gpt_neox.layers.21.attention.dense.weight": "pytorch_model-00022-of-00046.bin",
223
+ "gpt_neox.layers.21.attention.masked_bias": "pytorch_model-00022-of-00046.bin",
224
+ "gpt_neox.layers.21.attention.query_key_value.bias": "pytorch_model-00022-of-00046.bin",
225
+ "gpt_neox.layers.21.attention.query_key_value.weight": "pytorch_model-00022-of-00046.bin",
226
+ "gpt_neox.layers.21.attention.rotary_emb.inv_freq": "pytorch_model-00022-of-00046.bin",
227
+ "gpt_neox.layers.21.input_layernorm.bias": "pytorch_model-00022-of-00046.bin",
228
+ "gpt_neox.layers.21.input_layernorm.weight": "pytorch_model-00022-of-00046.bin",
229
+ "gpt_neox.layers.21.mlp.dense_4h_to_h.bias": "pytorch_model-00023-of-00046.bin",
230
+ "gpt_neox.layers.21.mlp.dense_4h_to_h.weight": "pytorch_model-00023-of-00046.bin",
231
+ "gpt_neox.layers.21.mlp.dense_h_to_4h.bias": "pytorch_model-00023-of-00046.bin",
232
+ "gpt_neox.layers.21.mlp.dense_h_to_4h.weight": "pytorch_model-00023-of-00046.bin",
233
+ "gpt_neox.layers.21.post_attention_layernorm.bias": "pytorch_model-00022-of-00046.bin",
234
+ "gpt_neox.layers.21.post_attention_layernorm.weight": "pytorch_model-00022-of-00046.bin",
235
+ "gpt_neox.layers.22.attention.bias": "pytorch_model-00023-of-00046.bin",
236
+ "gpt_neox.layers.22.attention.dense.bias": "pytorch_model-00023-of-00046.bin",
237
+ "gpt_neox.layers.22.attention.dense.weight": "pytorch_model-00023-of-00046.bin",
238
+ "gpt_neox.layers.22.attention.masked_bias": "pytorch_model-00023-of-00046.bin",
239
+ "gpt_neox.layers.22.attention.query_key_value.bias": "pytorch_model-00023-of-00046.bin",
240
+ "gpt_neox.layers.22.attention.query_key_value.weight": "pytorch_model-00023-of-00046.bin",
241
+ "gpt_neox.layers.22.attention.rotary_emb.inv_freq": "pytorch_model-00023-of-00046.bin",
242
+ "gpt_neox.layers.22.input_layernorm.bias": "pytorch_model-00023-of-00046.bin",
243
+ "gpt_neox.layers.22.input_layernorm.weight": "pytorch_model-00023-of-00046.bin",
244
+ "gpt_neox.layers.22.mlp.dense_4h_to_h.bias": "pytorch_model-00024-of-00046.bin",
245
+ "gpt_neox.layers.22.mlp.dense_4h_to_h.weight": "pytorch_model-00024-of-00046.bin",
246
+ "gpt_neox.layers.22.mlp.dense_h_to_4h.bias": "pytorch_model-00024-of-00046.bin",
247
+ "gpt_neox.layers.22.mlp.dense_h_to_4h.weight": "pytorch_model-00024-of-00046.bin",
248
+ "gpt_neox.layers.22.post_attention_layernorm.bias": "pytorch_model-00023-of-00046.bin",
249
+ "gpt_neox.layers.22.post_attention_layernorm.weight": "pytorch_model-00023-of-00046.bin",
250
+ "gpt_neox.layers.23.attention.bias": "pytorch_model-00024-of-00046.bin",
251
+ "gpt_neox.layers.23.attention.dense.bias": "pytorch_model-00024-of-00046.bin",
252
+ "gpt_neox.layers.23.attention.dense.weight": "pytorch_model-00024-of-00046.bin",
253
+ "gpt_neox.layers.23.attention.masked_bias": "pytorch_model-00024-of-00046.bin",
254
+ "gpt_neox.layers.23.attention.query_key_value.bias": "pytorch_model-00024-of-00046.bin",
255
+ "gpt_neox.layers.23.attention.query_key_value.weight": "pytorch_model-00024-of-00046.bin",
256
+ "gpt_neox.layers.23.attention.rotary_emb.inv_freq": "pytorch_model-00024-of-00046.bin",
257
+ "gpt_neox.layers.23.input_layernorm.bias": "pytorch_model-00024-of-00046.bin",
258
+ "gpt_neox.layers.23.input_layernorm.weight": "pytorch_model-00024-of-00046.bin",
259
+ "gpt_neox.layers.23.mlp.dense_4h_to_h.bias": "pytorch_model-00025-of-00046.bin",
260
+ "gpt_neox.layers.23.mlp.dense_4h_to_h.weight": "pytorch_model-00025-of-00046.bin",
261
+ "gpt_neox.layers.23.mlp.dense_h_to_4h.bias": "pytorch_model-00025-of-00046.bin",
262
+ "gpt_neox.layers.23.mlp.dense_h_to_4h.weight": "pytorch_model-00025-of-00046.bin",
263
+ "gpt_neox.layers.23.post_attention_layernorm.bias": "pytorch_model-00024-of-00046.bin",
264
+ "gpt_neox.layers.23.post_attention_layernorm.weight": "pytorch_model-00024-of-00046.bin",
265
+ "gpt_neox.layers.24.attention.bias": "pytorch_model-00025-of-00046.bin",
266
+ "gpt_neox.layers.24.attention.dense.bias": "pytorch_model-00025-of-00046.bin",
267
+ "gpt_neox.layers.24.attention.dense.weight": "pytorch_model-00025-of-00046.bin",
268
+ "gpt_neox.layers.24.attention.masked_bias": "pytorch_model-00025-of-00046.bin",
269
+ "gpt_neox.layers.24.attention.query_key_value.bias": "pytorch_model-00025-of-00046.bin",
270
+ "gpt_neox.layers.24.attention.query_key_value.weight": "pytorch_model-00025-of-00046.bin",
271
+ "gpt_neox.layers.24.attention.rotary_emb.inv_freq": "pytorch_model-00025-of-00046.bin",
272
+ "gpt_neox.layers.24.input_layernorm.bias": "pytorch_model-00025-of-00046.bin",
273
+ "gpt_neox.layers.24.input_layernorm.weight": "pytorch_model-00025-of-00046.bin",
274
+ "gpt_neox.layers.24.mlp.dense_4h_to_h.bias": "pytorch_model-00026-of-00046.bin",
275
+ "gpt_neox.layers.24.mlp.dense_4h_to_h.weight": "pytorch_model-00026-of-00046.bin",
276
+ "gpt_neox.layers.24.mlp.dense_h_to_4h.bias": "pytorch_model-00026-of-00046.bin",
277
+ "gpt_neox.layers.24.mlp.dense_h_to_4h.weight": "pytorch_model-00026-of-00046.bin",
278
+ "gpt_neox.layers.24.post_attention_layernorm.bias": "pytorch_model-00025-of-00046.bin",
279
+ "gpt_neox.layers.24.post_attention_layernorm.weight": "pytorch_model-00025-of-00046.bin",
280
+ "gpt_neox.layers.25.attention.bias": "pytorch_model-00026-of-00046.bin",
281
+ "gpt_neox.layers.25.attention.dense.bias": "pytorch_model-00026-of-00046.bin",
282
+ "gpt_neox.layers.25.attention.dense.weight": "pytorch_model-00026-of-00046.bin",
283
+ "gpt_neox.layers.25.attention.masked_bias": "pytorch_model-00026-of-00046.bin",
284
+ "gpt_neox.layers.25.attention.query_key_value.bias": "pytorch_model-00026-of-00046.bin",
285
+ "gpt_neox.layers.25.attention.query_key_value.weight": "pytorch_model-00026-of-00046.bin",
286
+ "gpt_neox.layers.25.attention.rotary_emb.inv_freq": "pytorch_model-00026-of-00046.bin",
287
+ "gpt_neox.layers.25.input_layernorm.bias": "pytorch_model-00026-of-00046.bin",
288
+ "gpt_neox.layers.25.input_layernorm.weight": "pytorch_model-00026-of-00046.bin",
289
+ "gpt_neox.layers.25.mlp.dense_4h_to_h.bias": "pytorch_model-00027-of-00046.bin",
290
+ "gpt_neox.layers.25.mlp.dense_4h_to_h.weight": "pytorch_model-00027-of-00046.bin",
291
+ "gpt_neox.layers.25.mlp.dense_h_to_4h.bias": "pytorch_model-00027-of-00046.bin",
292
+ "gpt_neox.layers.25.mlp.dense_h_to_4h.weight": "pytorch_model-00027-of-00046.bin",
293
+ "gpt_neox.layers.25.post_attention_layernorm.bias": "pytorch_model-00026-of-00046.bin",
294
+ "gpt_neox.layers.25.post_attention_layernorm.weight": "pytorch_model-00026-of-00046.bin",
295
+ "gpt_neox.layers.26.attention.bias": "pytorch_model-00027-of-00046.bin",
296
+ "gpt_neox.layers.26.attention.dense.bias": "pytorch_model-00027-of-00046.bin",
297
+ "gpt_neox.layers.26.attention.dense.weight": "pytorch_model-00027-of-00046.bin",
298
+ "gpt_neox.layers.26.attention.masked_bias": "pytorch_model-00027-of-00046.bin",
299
+ "gpt_neox.layers.26.attention.query_key_value.bias": "pytorch_model-00027-of-00046.bin",
300
+ "gpt_neox.layers.26.attention.query_key_value.weight": "pytorch_model-00027-of-00046.bin",
301
+ "gpt_neox.layers.26.attention.rotary_emb.inv_freq": "pytorch_model-00027-of-00046.bin",
302
+ "gpt_neox.layers.26.input_layernorm.bias": "pytorch_model-00027-of-00046.bin",
303
+ "gpt_neox.layers.26.input_layernorm.weight": "pytorch_model-00027-of-00046.bin",
304
+ "gpt_neox.layers.26.mlp.dense_4h_to_h.bias": "pytorch_model-00028-of-00046.bin",
305
+ "gpt_neox.layers.26.mlp.dense_4h_to_h.weight": "pytorch_model-00028-of-00046.bin",
306
+ "gpt_neox.layers.26.mlp.dense_h_to_4h.bias": "pytorch_model-00028-of-00046.bin",
307
+ "gpt_neox.layers.26.mlp.dense_h_to_4h.weight": "pytorch_model-00028-of-00046.bin",
308
+ "gpt_neox.layers.26.post_attention_layernorm.bias": "pytorch_model-00027-of-00046.bin",
309
+ "gpt_neox.layers.26.post_attention_layernorm.weight": "pytorch_model-00027-of-00046.bin",
310
+ "gpt_neox.layers.27.attention.bias": "pytorch_model-00028-of-00046.bin",
311
+ "gpt_neox.layers.27.attention.dense.bias": "pytorch_model-00028-of-00046.bin",
312
+ "gpt_neox.layers.27.attention.dense.weight": "pytorch_model-00028-of-00046.bin",
313
+ "gpt_neox.layers.27.attention.masked_bias": "pytorch_model-00028-of-00046.bin",
314
+ "gpt_neox.layers.27.attention.query_key_value.bias": "pytorch_model-00028-of-00046.bin",
315
+ "gpt_neox.layers.27.attention.query_key_value.weight": "pytorch_model-00028-of-00046.bin",
316
+ "gpt_neox.layers.27.attention.rotary_emb.inv_freq": "pytorch_model-00028-of-00046.bin",
317
+ "gpt_neox.layers.27.input_layernorm.bias": "pytorch_model-00028-of-00046.bin",
318
+ "gpt_neox.layers.27.input_layernorm.weight": "pytorch_model-00028-of-00046.bin",
319
+ "gpt_neox.layers.27.mlp.dense_4h_to_h.bias": "pytorch_model-00029-of-00046.bin",
320
+ "gpt_neox.layers.27.mlp.dense_4h_to_h.weight": "pytorch_model-00029-of-00046.bin",
321
+ "gpt_neox.layers.27.mlp.dense_h_to_4h.bias": "pytorch_model-00029-of-00046.bin",
322
+ "gpt_neox.layers.27.mlp.dense_h_to_4h.weight": "pytorch_model-00029-of-00046.bin",
323
+ "gpt_neox.layers.27.post_attention_layernorm.bias": "pytorch_model-00028-of-00046.bin",
324
+ "gpt_neox.layers.27.post_attention_layernorm.weight": "pytorch_model-00028-of-00046.bin",
325
+ "gpt_neox.layers.28.attention.bias": "pytorch_model-00029-of-00046.bin",
326
+ "gpt_neox.layers.28.attention.dense.bias": "pytorch_model-00029-of-00046.bin",
327
+ "gpt_neox.layers.28.attention.dense.weight": "pytorch_model-00029-of-00046.bin",
328
+ "gpt_neox.layers.28.attention.masked_bias": "pytorch_model-00029-of-00046.bin",
329
+ "gpt_neox.layers.28.attention.query_key_value.bias": "pytorch_model-00029-of-00046.bin",
330
+ "gpt_neox.layers.28.attention.query_key_value.weight": "pytorch_model-00029-of-00046.bin",
331
+ "gpt_neox.layers.28.attention.rotary_emb.inv_freq": "pytorch_model-00029-of-00046.bin",
332
+ "gpt_neox.layers.28.input_layernorm.bias": "pytorch_model-00029-of-00046.bin",
333
+ "gpt_neox.layers.28.input_layernorm.weight": "pytorch_model-00029-of-00046.bin",
334
+ "gpt_neox.layers.28.mlp.dense_4h_to_h.bias": "pytorch_model-00030-of-00046.bin",
335
+ "gpt_neox.layers.28.mlp.dense_4h_to_h.weight": "pytorch_model-00030-of-00046.bin",
336
+ "gpt_neox.layers.28.mlp.dense_h_to_4h.bias": "pytorch_model-00030-of-00046.bin",
337
+ "gpt_neox.layers.28.mlp.dense_h_to_4h.weight": "pytorch_model-00030-of-00046.bin",
338
+ "gpt_neox.layers.28.post_attention_layernorm.bias": "pytorch_model-00029-of-00046.bin",
339
+ "gpt_neox.layers.28.post_attention_layernorm.weight": "pytorch_model-00029-of-00046.bin",
340
+ "gpt_neox.layers.29.attention.bias": "pytorch_model-00030-of-00046.bin",
341
+ "gpt_neox.layers.29.attention.dense.bias": "pytorch_model-00030-of-00046.bin",
342
+ "gpt_neox.layers.29.attention.dense.weight": "pytorch_model-00030-of-00046.bin",
343
+ "gpt_neox.layers.29.attention.masked_bias": "pytorch_model-00030-of-00046.bin",
344
+ "gpt_neox.layers.29.attention.query_key_value.bias": "pytorch_model-00030-of-00046.bin",
345
+ "gpt_neox.layers.29.attention.query_key_value.weight": "pytorch_model-00030-of-00046.bin",
346
+ "gpt_neox.layers.29.attention.rotary_emb.inv_freq": "pytorch_model-00030-of-00046.bin",
347
+ "gpt_neox.layers.29.input_layernorm.bias": "pytorch_model-00030-of-00046.bin",
348
+ "gpt_neox.layers.29.input_layernorm.weight": "pytorch_model-00030-of-00046.bin",
349
+ "gpt_neox.layers.29.mlp.dense_4h_to_h.bias": "pytorch_model-00031-of-00046.bin",
350
+ "gpt_neox.layers.29.mlp.dense_4h_to_h.weight": "pytorch_model-00031-of-00046.bin",
351
+ "gpt_neox.layers.29.mlp.dense_h_to_4h.bias": "pytorch_model-00031-of-00046.bin",
352
+ "gpt_neox.layers.29.mlp.dense_h_to_4h.weight": "pytorch_model-00031-of-00046.bin",
353
+ "gpt_neox.layers.29.post_attention_layernorm.bias": "pytorch_model-00030-of-00046.bin",
354
+ "gpt_neox.layers.29.post_attention_layernorm.weight": "pytorch_model-00030-of-00046.bin",
355
+ "gpt_neox.layers.3.attention.bias": "pytorch_model-00004-of-00046.bin",
356
+ "gpt_neox.layers.3.attention.dense.bias": "pytorch_model-00004-of-00046.bin",
357
+ "gpt_neox.layers.3.attention.dense.weight": "pytorch_model-00004-of-00046.bin",
358
+ "gpt_neox.layers.3.attention.masked_bias": "pytorch_model-00004-of-00046.bin",
359
+ "gpt_neox.layers.3.attention.query_key_value.bias": "pytorch_model-00004-of-00046.bin",
360
+ "gpt_neox.layers.3.attention.query_key_value.weight": "pytorch_model-00004-of-00046.bin",
361
+ "gpt_neox.layers.3.attention.rotary_emb.inv_freq": "pytorch_model-00004-of-00046.bin",
362
+ "gpt_neox.layers.3.input_layernorm.bias": "pytorch_model-00004-of-00046.bin",
363
+ "gpt_neox.layers.3.input_layernorm.weight": "pytorch_model-00004-of-00046.bin",
364
+ "gpt_neox.layers.3.mlp.dense_4h_to_h.bias": "pytorch_model-00005-of-00046.bin",
365
+ "gpt_neox.layers.3.mlp.dense_4h_to_h.weight": "pytorch_model-00005-of-00046.bin",
366
+ "gpt_neox.layers.3.mlp.dense_h_to_4h.bias": "pytorch_model-00005-of-00046.bin",
367
+ "gpt_neox.layers.3.mlp.dense_h_to_4h.weight": "pytorch_model-00005-of-00046.bin",
368
+ "gpt_neox.layers.3.post_attention_layernorm.bias": "pytorch_model-00004-of-00046.bin",
369
+ "gpt_neox.layers.3.post_attention_layernorm.weight": "pytorch_model-00004-of-00046.bin",
370
+ "gpt_neox.layers.30.attention.bias": "pytorch_model-00031-of-00046.bin",
371
+ "gpt_neox.layers.30.attention.dense.bias": "pytorch_model-00031-of-00046.bin",
372
+ "gpt_neox.layers.30.attention.dense.weight": "pytorch_model-00031-of-00046.bin",
373
+ "gpt_neox.layers.30.attention.masked_bias": "pytorch_model-00031-of-00046.bin",
374
+ "gpt_neox.layers.30.attention.query_key_value.bias": "pytorch_model-00031-of-00046.bin",
375
+ "gpt_neox.layers.30.attention.query_key_value.weight": "pytorch_model-00031-of-00046.bin",
376
+ "gpt_neox.layers.30.attention.rotary_emb.inv_freq": "pytorch_model-00031-of-00046.bin",
377
+ "gpt_neox.layers.30.input_layernorm.bias": "pytorch_model-00031-of-00046.bin",
378
+ "gpt_neox.layers.30.input_layernorm.weight": "pytorch_model-00031-of-00046.bin",
379
+ "gpt_neox.layers.30.mlp.dense_4h_to_h.bias": "pytorch_model-00032-of-00046.bin",
380
+ "gpt_neox.layers.30.mlp.dense_4h_to_h.weight": "pytorch_model-00032-of-00046.bin",
381
+ "gpt_neox.layers.30.mlp.dense_h_to_4h.bias": "pytorch_model-00032-of-00046.bin",
382
+ "gpt_neox.layers.30.mlp.dense_h_to_4h.weight": "pytorch_model-00032-of-00046.bin",
383
+ "gpt_neox.layers.30.post_attention_layernorm.bias": "pytorch_model-00031-of-00046.bin",
384
+ "gpt_neox.layers.30.post_attention_layernorm.weight": "pytorch_model-00031-of-00046.bin",
385
+ "gpt_neox.layers.31.attention.bias": "pytorch_model-00032-of-00046.bin",
386
+ "gpt_neox.layers.31.attention.dense.bias": "pytorch_model-00032-of-00046.bin",
387
+ "gpt_neox.layers.31.attention.dense.weight": "pytorch_model-00032-of-00046.bin",
388
+ "gpt_neox.layers.31.attention.masked_bias": "pytorch_model-00032-of-00046.bin",
389
+ "gpt_neox.layers.31.attention.query_key_value.bias": "pytorch_model-00032-of-00046.bin",
390
+ "gpt_neox.layers.31.attention.query_key_value.weight": "pytorch_model-00032-of-00046.bin",
391
+ "gpt_neox.layers.31.attention.rotary_emb.inv_freq": "pytorch_model-00032-of-00046.bin",
392
+ "gpt_neox.layers.31.input_layernorm.bias": "pytorch_model-00032-of-00046.bin",
393
+ "gpt_neox.layers.31.input_layernorm.weight": "pytorch_model-00032-of-00046.bin",
394
+ "gpt_neox.layers.31.mlp.dense_4h_to_h.bias": "pytorch_model-00033-of-00046.bin",
395
+ "gpt_neox.layers.31.mlp.dense_4h_to_h.weight": "pytorch_model-00033-of-00046.bin",
396
+ "gpt_neox.layers.31.mlp.dense_h_to_4h.bias": "pytorch_model-00033-of-00046.bin",
397
+ "gpt_neox.layers.31.mlp.dense_h_to_4h.weight": "pytorch_model-00033-of-00046.bin",
398
+ "gpt_neox.layers.31.post_attention_layernorm.bias": "pytorch_model-00032-of-00046.bin",
399
+ "gpt_neox.layers.31.post_attention_layernorm.weight": "pytorch_model-00032-of-00046.bin",
400
+ "gpt_neox.layers.32.attention.bias": "pytorch_model-00033-of-00046.bin",
401
+ "gpt_neox.layers.32.attention.dense.bias": "pytorch_model-00033-of-00046.bin",
402
+ "gpt_neox.layers.32.attention.dense.weight": "pytorch_model-00033-of-00046.bin",
403
+ "gpt_neox.layers.32.attention.masked_bias": "pytorch_model-00033-of-00046.bin",
404
+ "gpt_neox.layers.32.attention.query_key_value.bias": "pytorch_model-00033-of-00046.bin",
405
+ "gpt_neox.layers.32.attention.query_key_value.weight": "pytorch_model-00033-of-00046.bin",
406
+ "gpt_neox.layers.32.attention.rotary_emb.inv_freq": "pytorch_model-00033-of-00046.bin",
407
+ "gpt_neox.layers.32.input_layernorm.bias": "pytorch_model-00033-of-00046.bin",
408
+ "gpt_neox.layers.32.input_layernorm.weight": "pytorch_model-00033-of-00046.bin",
409
+ "gpt_neox.layers.32.mlp.dense_4h_to_h.bias": "pytorch_model-00034-of-00046.bin",
410
+ "gpt_neox.layers.32.mlp.dense_4h_to_h.weight": "pytorch_model-00034-of-00046.bin",
411
+ "gpt_neox.layers.32.mlp.dense_h_to_4h.bias": "pytorch_model-00034-of-00046.bin",
412
+ "gpt_neox.layers.32.mlp.dense_h_to_4h.weight": "pytorch_model-00034-of-00046.bin",
413
+ "gpt_neox.layers.32.post_attention_layernorm.bias": "pytorch_model-00033-of-00046.bin",
414
+ "gpt_neox.layers.32.post_attention_layernorm.weight": "pytorch_model-00033-of-00046.bin",
415
+ "gpt_neox.layers.33.attention.bias": "pytorch_model-00034-of-00046.bin",
416
+ "gpt_neox.layers.33.attention.dense.bias": "pytorch_model-00034-of-00046.bin",
417
+ "gpt_neox.layers.33.attention.dense.weight": "pytorch_model-00034-of-00046.bin",
418
+ "gpt_neox.layers.33.attention.masked_bias": "pytorch_model-00034-of-00046.bin",
419
+ "gpt_neox.layers.33.attention.query_key_value.bias": "pytorch_model-00034-of-00046.bin",
420
+ "gpt_neox.layers.33.attention.query_key_value.weight": "pytorch_model-00034-of-00046.bin",
421
+ "gpt_neox.layers.33.attention.rotary_emb.inv_freq": "pytorch_model-00034-of-00046.bin",
422
+ "gpt_neox.layers.33.input_layernorm.bias": "pytorch_model-00034-of-00046.bin",
423
+ "gpt_neox.layers.33.input_layernorm.weight": "pytorch_model-00034-of-00046.bin",
424
+ "gpt_neox.layers.33.mlp.dense_4h_to_h.bias": "pytorch_model-00035-of-00046.bin",
425
+ "gpt_neox.layers.33.mlp.dense_4h_to_h.weight": "pytorch_model-00035-of-00046.bin",
426
+ "gpt_neox.layers.33.mlp.dense_h_to_4h.bias": "pytorch_model-00035-of-00046.bin",
427
+ "gpt_neox.layers.33.mlp.dense_h_to_4h.weight": "pytorch_model-00035-of-00046.bin",
428
+ "gpt_neox.layers.33.post_attention_layernorm.bias": "pytorch_model-00034-of-00046.bin",
429
+ "gpt_neox.layers.33.post_attention_layernorm.weight": "pytorch_model-00034-of-00046.bin",
430
+ "gpt_neox.layers.34.attention.bias": "pytorch_model-00035-of-00046.bin",
431
+ "gpt_neox.layers.34.attention.dense.bias": "pytorch_model-00035-of-00046.bin",
432
+ "gpt_neox.layers.34.attention.dense.weight": "pytorch_model-00035-of-00046.bin",
433
+ "gpt_neox.layers.34.attention.masked_bias": "pytorch_model-00035-of-00046.bin",
434
+ "gpt_neox.layers.34.attention.query_key_value.bias": "pytorch_model-00035-of-00046.bin",
435
+ "gpt_neox.layers.34.attention.query_key_value.weight": "pytorch_model-00035-of-00046.bin",
436
+ "gpt_neox.layers.34.attention.rotary_emb.inv_freq": "pytorch_model-00035-of-00046.bin",
437
+ "gpt_neox.layers.34.input_layernorm.bias": "pytorch_model-00035-of-00046.bin",
438
+ "gpt_neox.layers.34.input_layernorm.weight": "pytorch_model-00035-of-00046.bin",
439
+ "gpt_neox.layers.34.mlp.dense_4h_to_h.bias": "pytorch_model-00036-of-00046.bin",
440
+ "gpt_neox.layers.34.mlp.dense_4h_to_h.weight": "pytorch_model-00036-of-00046.bin",
441
+ "gpt_neox.layers.34.mlp.dense_h_to_4h.bias": "pytorch_model-00036-of-00046.bin",
442
+ "gpt_neox.layers.34.mlp.dense_h_to_4h.weight": "pytorch_model-00036-of-00046.bin",
443
+ "gpt_neox.layers.34.post_attention_layernorm.bias": "pytorch_model-00035-of-00046.bin",
444
+ "gpt_neox.layers.34.post_attention_layernorm.weight": "pytorch_model-00035-of-00046.bin",
445
+ "gpt_neox.layers.35.attention.bias": "pytorch_model-00036-of-00046.bin",
446
+ "gpt_neox.layers.35.attention.dense.bias": "pytorch_model-00036-of-00046.bin",
447
+ "gpt_neox.layers.35.attention.dense.weight": "pytorch_model-00036-of-00046.bin",
448
+ "gpt_neox.layers.35.attention.masked_bias": "pytorch_model-00036-of-00046.bin",
449
+ "gpt_neox.layers.35.attention.query_key_value.bias": "pytorch_model-00036-of-00046.bin",
450
+ "gpt_neox.layers.35.attention.query_key_value.weight": "pytorch_model-00036-of-00046.bin",
451
+ "gpt_neox.layers.35.attention.rotary_emb.inv_freq": "pytorch_model-00036-of-00046.bin",
452
+ "gpt_neox.layers.35.input_layernorm.bias": "pytorch_model-00036-of-00046.bin",
453
+ "gpt_neox.layers.35.input_layernorm.weight": "pytorch_model-00036-of-00046.bin",
454
+ "gpt_neox.layers.35.mlp.dense_4h_to_h.bias": "pytorch_model-00037-of-00046.bin",
455
+ "gpt_neox.layers.35.mlp.dense_4h_to_h.weight": "pytorch_model-00037-of-00046.bin",
456
+ "gpt_neox.layers.35.mlp.dense_h_to_4h.bias": "pytorch_model-00037-of-00046.bin",
457
+ "gpt_neox.layers.35.mlp.dense_h_to_4h.weight": "pytorch_model-00037-of-00046.bin",
458
+ "gpt_neox.layers.35.post_attention_layernorm.bias": "pytorch_model-00036-of-00046.bin",
459
+ "gpt_neox.layers.35.post_attention_layernorm.weight": "pytorch_model-00036-of-00046.bin",
460
+ "gpt_neox.layers.36.attention.bias": "pytorch_model-00037-of-00046.bin",
461
+ "gpt_neox.layers.36.attention.dense.bias": "pytorch_model-00037-of-00046.bin",
462
+ "gpt_neox.layers.36.attention.dense.weight": "pytorch_model-00037-of-00046.bin",
463
+ "gpt_neox.layers.36.attention.masked_bias": "pytorch_model-00037-of-00046.bin",
464
+ "gpt_neox.layers.36.attention.query_key_value.bias": "pytorch_model-00037-of-00046.bin",
465
+ "gpt_neox.layers.36.attention.query_key_value.weight": "pytorch_model-00037-of-00046.bin",
466
+ "gpt_neox.layers.36.attention.rotary_emb.inv_freq": "pytorch_model-00037-of-00046.bin",
467
+ "gpt_neox.layers.36.input_layernorm.bias": "pytorch_model-00037-of-00046.bin",
468
+ "gpt_neox.layers.36.input_layernorm.weight": "pytorch_model-00037-of-00046.bin",
469
+ "gpt_neox.layers.36.mlp.dense_4h_to_h.bias": "pytorch_model-00038-of-00046.bin",
470
+ "gpt_neox.layers.36.mlp.dense_4h_to_h.weight": "pytorch_model-00038-of-00046.bin",
471
+ "gpt_neox.layers.36.mlp.dense_h_to_4h.bias": "pytorch_model-00038-of-00046.bin",
472
+ "gpt_neox.layers.36.mlp.dense_h_to_4h.weight": "pytorch_model-00038-of-00046.bin",
473
+ "gpt_neox.layers.36.post_attention_layernorm.bias": "pytorch_model-00037-of-00046.bin",
474
+ "gpt_neox.layers.36.post_attention_layernorm.weight": "pytorch_model-00037-of-00046.bin",
475
+ "gpt_neox.layers.37.attention.bias": "pytorch_model-00038-of-00046.bin",
476
+ "gpt_neox.layers.37.attention.dense.bias": "pytorch_model-00038-of-00046.bin",
477
+ "gpt_neox.layers.37.attention.dense.weight": "pytorch_model-00038-of-00046.bin",
478
+ "gpt_neox.layers.37.attention.masked_bias": "pytorch_model-00038-of-00046.bin",
479
+ "gpt_neox.layers.37.attention.query_key_value.bias": "pytorch_model-00038-of-00046.bin",
480
+ "gpt_neox.layers.37.attention.query_key_value.weight": "pytorch_model-00038-of-00046.bin",
481
+ "gpt_neox.layers.37.attention.rotary_emb.inv_freq": "pytorch_model-00038-of-00046.bin",
482
+ "gpt_neox.layers.37.input_layernorm.bias": "pytorch_model-00038-of-00046.bin",
483
+ "gpt_neox.layers.37.input_layernorm.weight": "pytorch_model-00038-of-00046.bin",
484
+ "gpt_neox.layers.37.mlp.dense_4h_to_h.bias": "pytorch_model-00039-of-00046.bin",
485
+ "gpt_neox.layers.37.mlp.dense_4h_to_h.weight": "pytorch_model-00039-of-00046.bin",
486
+ "gpt_neox.layers.37.mlp.dense_h_to_4h.bias": "pytorch_model-00039-of-00046.bin",
487
+ "gpt_neox.layers.37.mlp.dense_h_to_4h.weight": "pytorch_model-00039-of-00046.bin",
488
+ "gpt_neox.layers.37.post_attention_layernorm.bias": "pytorch_model-00038-of-00046.bin",
489
+ "gpt_neox.layers.37.post_attention_layernorm.weight": "pytorch_model-00038-of-00046.bin",
490
+ "gpt_neox.layers.38.attention.bias": "pytorch_model-00039-of-00046.bin",
491
+ "gpt_neox.layers.38.attention.dense.bias": "pytorch_model-00039-of-00046.bin",
492
+ "gpt_neox.layers.38.attention.dense.weight": "pytorch_model-00039-of-00046.bin",
493
+ "gpt_neox.layers.38.attention.masked_bias": "pytorch_model-00039-of-00046.bin",
494
+ "gpt_neox.layers.38.attention.query_key_value.bias": "pytorch_model-00039-of-00046.bin",
495
+ "gpt_neox.layers.38.attention.query_key_value.weight": "pytorch_model-00039-of-00046.bin",
496
+ "gpt_neox.layers.38.attention.rotary_emb.inv_freq": "pytorch_model-00039-of-00046.bin",
497
+ "gpt_neox.layers.38.input_layernorm.bias": "pytorch_model-00039-of-00046.bin",
498
+ "gpt_neox.layers.38.input_layernorm.weight": "pytorch_model-00039-of-00046.bin",
499
+ "gpt_neox.layers.38.mlp.dense_4h_to_h.bias": "pytorch_model-00040-of-00046.bin",
500
+ "gpt_neox.layers.38.mlp.dense_4h_to_h.weight": "pytorch_model-00040-of-00046.bin",
501
+ "gpt_neox.layers.38.mlp.dense_h_to_4h.bias": "pytorch_model-00040-of-00046.bin",
502
+ "gpt_neox.layers.38.mlp.dense_h_to_4h.weight": "pytorch_model-00040-of-00046.bin",
503
+ "gpt_neox.layers.38.post_attention_layernorm.bias": "pytorch_model-00039-of-00046.bin",
504
+ "gpt_neox.layers.38.post_attention_layernorm.weight": "pytorch_model-00039-of-00046.bin",
505
+ "gpt_neox.layers.39.attention.bias": "pytorch_model-00040-of-00046.bin",
506
+ "gpt_neox.layers.39.attention.dense.bias": "pytorch_model-00040-of-00046.bin",
507
+ "gpt_neox.layers.39.attention.dense.weight": "pytorch_model-00040-of-00046.bin",
508
+ "gpt_neox.layers.39.attention.masked_bias": "pytorch_model-00040-of-00046.bin",
509
+ "gpt_neox.layers.39.attention.query_key_value.bias": "pytorch_model-00040-of-00046.bin",
510
+ "gpt_neox.layers.39.attention.query_key_value.weight": "pytorch_model-00040-of-00046.bin",
511
+ "gpt_neox.layers.39.attention.rotary_emb.inv_freq": "pytorch_model-00040-of-00046.bin",
512
+ "gpt_neox.layers.39.input_layernorm.bias": "pytorch_model-00040-of-00046.bin",
513
+ "gpt_neox.layers.39.input_layernorm.weight": "pytorch_model-00040-of-00046.bin",
514
+ "gpt_neox.layers.39.mlp.dense_4h_to_h.bias": "pytorch_model-00041-of-00046.bin",
515
+ "gpt_neox.layers.39.mlp.dense_4h_to_h.weight": "pytorch_model-00041-of-00046.bin",
516
+ "gpt_neox.layers.39.mlp.dense_h_to_4h.bias": "pytorch_model-00041-of-00046.bin",
517
+ "gpt_neox.layers.39.mlp.dense_h_to_4h.weight": "pytorch_model-00041-of-00046.bin",
518
+ "gpt_neox.layers.39.post_attention_layernorm.bias": "pytorch_model-00040-of-00046.bin",
519
+ "gpt_neox.layers.39.post_attention_layernorm.weight": "pytorch_model-00040-of-00046.bin",
520
+ "gpt_neox.layers.4.attention.bias": "pytorch_model-00005-of-00046.bin",
521
+ "gpt_neox.layers.4.attention.dense.bias": "pytorch_model-00005-of-00046.bin",
522
+ "gpt_neox.layers.4.attention.dense.weight": "pytorch_model-00005-of-00046.bin",
523
+ "gpt_neox.layers.4.attention.masked_bias": "pytorch_model-00005-of-00046.bin",
524
+ "gpt_neox.layers.4.attention.query_key_value.bias": "pytorch_model-00005-of-00046.bin",
525
+ "gpt_neox.layers.4.attention.query_key_value.weight": "pytorch_model-00005-of-00046.bin",
526
+ "gpt_neox.layers.4.attention.rotary_emb.inv_freq": "pytorch_model-00005-of-00046.bin",
527
+ "gpt_neox.layers.4.input_layernorm.bias": "pytorch_model-00005-of-00046.bin",
528
+ "gpt_neox.layers.4.input_layernorm.weight": "pytorch_model-00005-of-00046.bin",
529
+ "gpt_neox.layers.4.mlp.dense_4h_to_h.bias": "pytorch_model-00006-of-00046.bin",
530
+ "gpt_neox.layers.4.mlp.dense_4h_to_h.weight": "pytorch_model-00006-of-00046.bin",
531
+ "gpt_neox.layers.4.mlp.dense_h_to_4h.bias": "pytorch_model-00006-of-00046.bin",
532
+ "gpt_neox.layers.4.mlp.dense_h_to_4h.weight": "pytorch_model-00006-of-00046.bin",
533
+ "gpt_neox.layers.4.post_attention_layernorm.bias": "pytorch_model-00005-of-00046.bin",
534
+ "gpt_neox.layers.4.post_attention_layernorm.weight": "pytorch_model-00005-of-00046.bin",
535
+ "gpt_neox.layers.40.attention.bias": "pytorch_model-00041-of-00046.bin",
536
+ "gpt_neox.layers.40.attention.dense.bias": "pytorch_model-00041-of-00046.bin",
537
+ "gpt_neox.layers.40.attention.dense.weight": "pytorch_model-00041-of-00046.bin",
538
+ "gpt_neox.layers.40.attention.masked_bias": "pytorch_model-00041-of-00046.bin",
539
+ "gpt_neox.layers.40.attention.query_key_value.bias": "pytorch_model-00041-of-00046.bin",
540
+ "gpt_neox.layers.40.attention.query_key_value.weight": "pytorch_model-00041-of-00046.bin",
541
+ "gpt_neox.layers.40.attention.rotary_emb.inv_freq": "pytorch_model-00041-of-00046.bin",
542
+ "gpt_neox.layers.40.input_layernorm.bias": "pytorch_model-00041-of-00046.bin",
543
+ "gpt_neox.layers.40.input_layernorm.weight": "pytorch_model-00041-of-00046.bin",
544
+ "gpt_neox.layers.40.mlp.dense_4h_to_h.bias": "pytorch_model-00042-of-00046.bin",
545
+ "gpt_neox.layers.40.mlp.dense_4h_to_h.weight": "pytorch_model-00042-of-00046.bin",
546
+ "gpt_neox.layers.40.mlp.dense_h_to_4h.bias": "pytorch_model-00042-of-00046.bin",
547
+ "gpt_neox.layers.40.mlp.dense_h_to_4h.weight": "pytorch_model-00042-of-00046.bin",
548
+ "gpt_neox.layers.40.post_attention_layernorm.bias": "pytorch_model-00041-of-00046.bin",
549
+ "gpt_neox.layers.40.post_attention_layernorm.weight": "pytorch_model-00041-of-00046.bin",
550
+ "gpt_neox.layers.41.attention.bias": "pytorch_model-00042-of-00046.bin",
551
+ "gpt_neox.layers.41.attention.dense.bias": "pytorch_model-00042-of-00046.bin",
552
+ "gpt_neox.layers.41.attention.dense.weight": "pytorch_model-00042-of-00046.bin",
553
+ "gpt_neox.layers.41.attention.masked_bias": "pytorch_model-00042-of-00046.bin",
554
+ "gpt_neox.layers.41.attention.query_key_value.bias": "pytorch_model-00042-of-00046.bin",
555
+ "gpt_neox.layers.41.attention.query_key_value.weight": "pytorch_model-00042-of-00046.bin",
556
+ "gpt_neox.layers.41.attention.rotary_emb.inv_freq": "pytorch_model-00042-of-00046.bin",
557
+ "gpt_neox.layers.41.input_layernorm.bias": "pytorch_model-00042-of-00046.bin",
558
+ "gpt_neox.layers.41.input_layernorm.weight": "pytorch_model-00042-of-00046.bin",
559
+ "gpt_neox.layers.41.mlp.dense_4h_to_h.bias": "pytorch_model-00043-of-00046.bin",
560
+ "gpt_neox.layers.41.mlp.dense_4h_to_h.weight": "pytorch_model-00043-of-00046.bin",
561
+ "gpt_neox.layers.41.mlp.dense_h_to_4h.bias": "pytorch_model-00043-of-00046.bin",
562
+ "gpt_neox.layers.41.mlp.dense_h_to_4h.weight": "pytorch_model-00043-of-00046.bin",
563
+ "gpt_neox.layers.41.post_attention_layernorm.bias": "pytorch_model-00042-of-00046.bin",
564
+ "gpt_neox.layers.41.post_attention_layernorm.weight": "pytorch_model-00042-of-00046.bin",
565
+ "gpt_neox.layers.42.attention.bias": "pytorch_model-00043-of-00046.bin",
566
+ "gpt_neox.layers.42.attention.dense.bias": "pytorch_model-00043-of-00046.bin",
567
+ "gpt_neox.layers.42.attention.dense.weight": "pytorch_model-00043-of-00046.bin",
568
+ "gpt_neox.layers.42.attention.masked_bias": "pytorch_model-00043-of-00046.bin",
569
+ "gpt_neox.layers.42.attention.query_key_value.bias": "pytorch_model-00043-of-00046.bin",
570
+ "gpt_neox.layers.42.attention.query_key_value.weight": "pytorch_model-00043-of-00046.bin",
571
+ "gpt_neox.layers.42.attention.rotary_emb.inv_freq": "pytorch_model-00043-of-00046.bin",
572
+ "gpt_neox.layers.42.input_layernorm.bias": "pytorch_model-00043-of-00046.bin",
573
+ "gpt_neox.layers.42.input_layernorm.weight": "pytorch_model-00043-of-00046.bin",
574
+ "gpt_neox.layers.42.mlp.dense_4h_to_h.bias": "pytorch_model-00044-of-00046.bin",
575
+ "gpt_neox.layers.42.mlp.dense_4h_to_h.weight": "pytorch_model-00044-of-00046.bin",
576
+ "gpt_neox.layers.42.mlp.dense_h_to_4h.bias": "pytorch_model-00044-of-00046.bin",
577
+ "gpt_neox.layers.42.mlp.dense_h_to_4h.weight": "pytorch_model-00044-of-00046.bin",
578
+ "gpt_neox.layers.42.post_attention_layernorm.bias": "pytorch_model-00043-of-00046.bin",
579
+ "gpt_neox.layers.42.post_attention_layernorm.weight": "pytorch_model-00043-of-00046.bin",
580
+ "gpt_neox.layers.43.attention.bias": "pytorch_model-00044-of-00046.bin",
581
+ "gpt_neox.layers.43.attention.dense.bias": "pytorch_model-00044-of-00046.bin",
582
+ "gpt_neox.layers.43.attention.dense.weight": "pytorch_model-00044-of-00046.bin",
583
+ "gpt_neox.layers.43.attention.masked_bias": "pytorch_model-00044-of-00046.bin",
584
+ "gpt_neox.layers.43.attention.query_key_value.bias": "pytorch_model-00044-of-00046.bin",
585
+ "gpt_neox.layers.43.attention.query_key_value.weight": "pytorch_model-00044-of-00046.bin",
586
+ "gpt_neox.layers.43.attention.rotary_emb.inv_freq": "pytorch_model-00044-of-00046.bin",
587
+ "gpt_neox.layers.43.input_layernorm.bias": "pytorch_model-00044-of-00046.bin",
588
+ "gpt_neox.layers.43.input_layernorm.weight": "pytorch_model-00044-of-00046.bin",
589
+ "gpt_neox.layers.43.mlp.dense_4h_to_h.bias": "pytorch_model-00045-of-00046.bin",
590
+ "gpt_neox.layers.43.mlp.dense_4h_to_h.weight": "pytorch_model-00045-of-00046.bin",
591
+ "gpt_neox.layers.43.mlp.dense_h_to_4h.bias": "pytorch_model-00045-of-00046.bin",
592
+ "gpt_neox.layers.43.mlp.dense_h_to_4h.weight": "pytorch_model-00045-of-00046.bin",
593
+ "gpt_neox.layers.43.post_attention_layernorm.bias": "pytorch_model-00044-of-00046.bin",
594
+ "gpt_neox.layers.43.post_attention_layernorm.weight": "pytorch_model-00044-of-00046.bin",
595
+ "gpt_neox.layers.5.attention.bias": "pytorch_model-00006-of-00046.bin",
596
+ "gpt_neox.layers.5.attention.dense.bias": "pytorch_model-00006-of-00046.bin",
597
+ "gpt_neox.layers.5.attention.dense.weight": "pytorch_model-00006-of-00046.bin",
598
+ "gpt_neox.layers.5.attention.masked_bias": "pytorch_model-00006-of-00046.bin",
599
+ "gpt_neox.layers.5.attention.query_key_value.bias": "pytorch_model-00006-of-00046.bin",
600
+ "gpt_neox.layers.5.attention.query_key_value.weight": "pytorch_model-00006-of-00046.bin",
601
+ "gpt_neox.layers.5.attention.rotary_emb.inv_freq": "pytorch_model-00006-of-00046.bin",
602
+ "gpt_neox.layers.5.input_layernorm.bias": "pytorch_model-00006-of-00046.bin",
603
+ "gpt_neox.layers.5.input_layernorm.weight": "pytorch_model-00006-of-00046.bin",
604
+ "gpt_neox.layers.5.mlp.dense_4h_to_h.bias": "pytorch_model-00007-of-00046.bin",
605
+ "gpt_neox.layers.5.mlp.dense_4h_to_h.weight": "pytorch_model-00007-of-00046.bin",
606
+ "gpt_neox.layers.5.mlp.dense_h_to_4h.bias": "pytorch_model-00007-of-00046.bin",
607
+ "gpt_neox.layers.5.mlp.dense_h_to_4h.weight": "pytorch_model-00007-of-00046.bin",
608
+ "gpt_neox.layers.5.post_attention_layernorm.bias": "pytorch_model-00006-of-00046.bin",
609
+ "gpt_neox.layers.5.post_attention_layernorm.weight": "pytorch_model-00006-of-00046.bin",
610
+ "gpt_neox.layers.6.attention.bias": "pytorch_model-00007-of-00046.bin",
611
+ "gpt_neox.layers.6.attention.dense.bias": "pytorch_model-00007-of-00046.bin",
612
+ "gpt_neox.layers.6.attention.dense.weight": "pytorch_model-00007-of-00046.bin",
613
+ "gpt_neox.layers.6.attention.masked_bias": "pytorch_model-00007-of-00046.bin",
614
+ "gpt_neox.layers.6.attention.query_key_value.bias": "pytorch_model-00007-of-00046.bin",
615
+ "gpt_neox.layers.6.attention.query_key_value.weight": "pytorch_model-00007-of-00046.bin",
616
+ "gpt_neox.layers.6.attention.rotary_emb.inv_freq": "pytorch_model-00007-of-00046.bin",
617
+ "gpt_neox.layers.6.input_layernorm.bias": "pytorch_model-00007-of-00046.bin",
618
+ "gpt_neox.layers.6.input_layernorm.weight": "pytorch_model-00007-of-00046.bin",
619
+ "gpt_neox.layers.6.mlp.dense_4h_to_h.bias": "pytorch_model-00008-of-00046.bin",
620
+ "gpt_neox.layers.6.mlp.dense_4h_to_h.weight": "pytorch_model-00008-of-00046.bin",
621
+ "gpt_neox.layers.6.mlp.dense_h_to_4h.bias": "pytorch_model-00008-of-00046.bin",
622
+ "gpt_neox.layers.6.mlp.dense_h_to_4h.weight": "pytorch_model-00008-of-00046.bin",
623
+ "gpt_neox.layers.6.post_attention_layernorm.bias": "pytorch_model-00007-of-00046.bin",
624
+ "gpt_neox.layers.6.post_attention_layernorm.weight": "pytorch_model-00007-of-00046.bin",
625
+ "gpt_neox.layers.7.attention.bias": "pytorch_model-00008-of-00046.bin",
626
+ "gpt_neox.layers.7.attention.dense.bias": "pytorch_model-00008-of-00046.bin",
627
+ "gpt_neox.layers.7.attention.dense.weight": "pytorch_model-00008-of-00046.bin",
628
+ "gpt_neox.layers.7.attention.masked_bias": "pytorch_model-00008-of-00046.bin",
629
+ "gpt_neox.layers.7.attention.query_key_value.bias": "pytorch_model-00008-of-00046.bin",
630
+ "gpt_neox.layers.7.attention.query_key_value.weight": "pytorch_model-00008-of-00046.bin",
631
+ "gpt_neox.layers.7.attention.rotary_emb.inv_freq": "pytorch_model-00008-of-00046.bin",
632
+ "gpt_neox.layers.7.input_layernorm.bias": "pytorch_model-00008-of-00046.bin",
633
+ "gpt_neox.layers.7.input_layernorm.weight": "pytorch_model-00008-of-00046.bin",
634
+ "gpt_neox.layers.7.mlp.dense_4h_to_h.bias": "pytorch_model-00009-of-00046.bin",
635
+ "gpt_neox.layers.7.mlp.dense_4h_to_h.weight": "pytorch_model-00009-of-00046.bin",
636
+ "gpt_neox.layers.7.mlp.dense_h_to_4h.bias": "pytorch_model-00009-of-00046.bin",
637
+ "gpt_neox.layers.7.mlp.dense_h_to_4h.weight": "pytorch_model-00009-of-00046.bin",
638
+ "gpt_neox.layers.7.post_attention_layernorm.bias": "pytorch_model-00008-of-00046.bin",
639
+ "gpt_neox.layers.7.post_attention_layernorm.weight": "pytorch_model-00008-of-00046.bin",
640
+ "gpt_neox.layers.8.attention.bias": "pytorch_model-00009-of-00046.bin",
641
+ "gpt_neox.layers.8.attention.dense.bias": "pytorch_model-00009-of-00046.bin",
642
+ "gpt_neox.layers.8.attention.dense.weight": "pytorch_model-00009-of-00046.bin",
643
+ "gpt_neox.layers.8.attention.masked_bias": "pytorch_model-00009-of-00046.bin",
644
+ "gpt_neox.layers.8.attention.query_key_value.bias": "pytorch_model-00009-of-00046.bin",
645
+ "gpt_neox.layers.8.attention.query_key_value.weight": "pytorch_model-00009-of-00046.bin",
646
+ "gpt_neox.layers.8.attention.rotary_emb.inv_freq": "pytorch_model-00009-of-00046.bin",
647
+ "gpt_neox.layers.8.input_layernorm.bias": "pytorch_model-00009-of-00046.bin",
648
+ "gpt_neox.layers.8.input_layernorm.weight": "pytorch_model-00009-of-00046.bin",
649
+ "gpt_neox.layers.8.mlp.dense_4h_to_h.bias": "pytorch_model-00010-of-00046.bin",
650
+ "gpt_neox.layers.8.mlp.dense_4h_to_h.weight": "pytorch_model-00010-of-00046.bin",
651
+ "gpt_neox.layers.8.mlp.dense_h_to_4h.bias": "pytorch_model-00010-of-00046.bin",
652
+ "gpt_neox.layers.8.mlp.dense_h_to_4h.weight": "pytorch_model-00010-of-00046.bin",
653
+ "gpt_neox.layers.8.post_attention_layernorm.bias": "pytorch_model-00009-of-00046.bin",
654
+ "gpt_neox.layers.8.post_attention_layernorm.weight": "pytorch_model-00009-of-00046.bin",
655
+ "gpt_neox.layers.9.attention.bias": "pytorch_model-00010-of-00046.bin",
656
+ "gpt_neox.layers.9.attention.dense.bias": "pytorch_model-00010-of-00046.bin",
657
+ "gpt_neox.layers.9.attention.dense.weight": "pytorch_model-00010-of-00046.bin",
658
+ "gpt_neox.layers.9.attention.masked_bias": "pytorch_model-00010-of-00046.bin",
659
+ "gpt_neox.layers.9.attention.query_key_value.bias": "pytorch_model-00010-of-00046.bin",
660
+ "gpt_neox.layers.9.attention.query_key_value.weight": "pytorch_model-00010-of-00046.bin",
661
+ "gpt_neox.layers.9.attention.rotary_emb.inv_freq": "pytorch_model-00010-of-00046.bin",
662
+ "gpt_neox.layers.9.input_layernorm.bias": "pytorch_model-00010-of-00046.bin",
663
+ "gpt_neox.layers.9.input_layernorm.weight": "pytorch_model-00010-of-00046.bin",
664
+ "gpt_neox.layers.9.mlp.dense_4h_to_h.bias": "pytorch_model-00011-of-00046.bin",
665
+ "gpt_neox.layers.9.mlp.dense_4h_to_h.weight": "pytorch_model-00011-of-00046.bin",
666
+ "gpt_neox.layers.9.mlp.dense_h_to_4h.bias": "pytorch_model-00011-of-00046.bin",
667
+ "gpt_neox.layers.9.mlp.dense_h_to_4h.weight": "pytorch_model-00011-of-00046.bin",
668
+ "gpt_neox.layers.9.post_attention_layernorm.bias": "pytorch_model-00010-of-00046.bin",
669
+ "gpt_neox.layers.9.post_attention_layernorm.weight": "pytorch_model-00010-of-00046.bin"
670
+ }
671
+ }
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c1c533d9f9369841d58c7b49b873072baf9da45f0668396625eb4eb0e791c8e
3
+ size 14575
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0df7ca13bd2d05e2f4fe8b42e46b381ac116909f5329aa8a852d1eff61106519
3
+ size 627
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "tokenizer_class": "GPTNeoXTokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,3640 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 33.982300884955755,
5
+ "global_step": 600,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.06,
12
+ "learning_rate": 2.0000000000000003e-06,
13
+ "loss": 2.2431,
14
+ "step": 1
15
+ },
16
+ {
17
+ "epoch": 0.11,
18
+ "learning_rate": 4.000000000000001e-06,
19
+ "loss": 2.5758,
20
+ "step": 2
21
+ },
22
+ {
23
+ "epoch": 0.17,
24
+ "learning_rate": 6e-06,
25
+ "loss": 2.3017,
26
+ "step": 3
27
+ },
28
+ {
29
+ "epoch": 0.23,
30
+ "learning_rate": 8.000000000000001e-06,
31
+ "loss": 2.3675,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 0.28,
36
+ "learning_rate": 1e-05,
37
+ "loss": 2.3984,
38
+ "step": 5
39
+ },
40
+ {
41
+ "epoch": 0.34,
42
+ "learning_rate": 1.2e-05,
43
+ "loss": 2.2344,
44
+ "step": 6
45
+ },
46
+ {
47
+ "epoch": 0.4,
48
+ "learning_rate": 1.4000000000000001e-05,
49
+ "loss": 2.3358,
50
+ "step": 7
51
+ },
52
+ {
53
+ "epoch": 0.45,
54
+ "learning_rate": 1.6000000000000003e-05,
55
+ "loss": 2.183,
56
+ "step": 8
57
+ },
58
+ {
59
+ "epoch": 0.51,
60
+ "learning_rate": 1.8e-05,
61
+ "loss": 2.2045,
62
+ "step": 9
63
+ },
64
+ {
65
+ "epoch": 0.57,
66
+ "learning_rate": 2e-05,
67
+ "loss": 2.146,
68
+ "step": 10
69
+ },
70
+ {
71
+ "epoch": 0.62,
72
+ "learning_rate": 2.2000000000000003e-05,
73
+ "loss": 2.0256,
74
+ "step": 11
75
+ },
76
+ {
77
+ "epoch": 0.68,
78
+ "learning_rate": 2.4e-05,
79
+ "loss": 1.961,
80
+ "step": 12
81
+ },
82
+ {
83
+ "epoch": 0.74,
84
+ "learning_rate": 2.6000000000000002e-05,
85
+ "loss": 2.121,
86
+ "step": 13
87
+ },
88
+ {
89
+ "epoch": 0.79,
90
+ "learning_rate": 2.8000000000000003e-05,
91
+ "loss": 2.0502,
92
+ "step": 14
93
+ },
94
+ {
95
+ "epoch": 0.85,
96
+ "learning_rate": 3e-05,
97
+ "loss": 1.9754,
98
+ "step": 15
99
+ },
100
+ {
101
+ "epoch": 0.91,
102
+ "learning_rate": 3.2000000000000005e-05,
103
+ "loss": 1.8381,
104
+ "step": 16
105
+ },
106
+ {
107
+ "epoch": 0.96,
108
+ "learning_rate": 3.4000000000000007e-05,
109
+ "loss": 1.8665,
110
+ "step": 17
111
+ },
112
+ {
113
+ "epoch": 1.02,
114
+ "learning_rate": 3.6e-05,
115
+ "loss": 2.0878,
116
+ "step": 18
117
+ },
118
+ {
119
+ "epoch": 1.08,
120
+ "learning_rate": 3.8e-05,
121
+ "loss": 1.7907,
122
+ "step": 19
123
+ },
124
+ {
125
+ "epoch": 1.13,
126
+ "learning_rate": 4e-05,
127
+ "loss": 2.0663,
128
+ "step": 20
129
+ },
130
+ {
131
+ "epoch": 1.19,
132
+ "learning_rate": 4.2e-05,
133
+ "loss": 1.7996,
134
+ "step": 21
135
+ },
136
+ {
137
+ "epoch": 1.25,
138
+ "learning_rate": 4.4000000000000006e-05,
139
+ "loss": 2.0797,
140
+ "step": 22
141
+ },
142
+ {
143
+ "epoch": 1.3,
144
+ "learning_rate": 4.600000000000001e-05,
145
+ "loss": 1.8656,
146
+ "step": 23
147
+ },
148
+ {
149
+ "epoch": 1.36,
150
+ "learning_rate": 4.8e-05,
151
+ "loss": 1.8085,
152
+ "step": 24
153
+ },
154
+ {
155
+ "epoch": 1.42,
156
+ "learning_rate": 5e-05,
157
+ "loss": 1.8279,
158
+ "step": 25
159
+ },
160
+ {
161
+ "epoch": 1.47,
162
+ "learning_rate": 5.2000000000000004e-05,
163
+ "loss": 1.7922,
164
+ "step": 26
165
+ },
166
+ {
167
+ "epoch": 1.53,
168
+ "learning_rate": 5.4000000000000005e-05,
169
+ "loss": 1.8851,
170
+ "step": 27
171
+ },
172
+ {
173
+ "epoch": 1.59,
174
+ "learning_rate": 5.6000000000000006e-05,
175
+ "loss": 1.6355,
176
+ "step": 28
177
+ },
178
+ {
179
+ "epoch": 1.64,
180
+ "learning_rate": 5.8e-05,
181
+ "loss": 1.7099,
182
+ "step": 29
183
+ },
184
+ {
185
+ "epoch": 1.7,
186
+ "learning_rate": 6e-05,
187
+ "loss": 1.7664,
188
+ "step": 30
189
+ },
190
+ {
191
+ "epoch": 1.76,
192
+ "learning_rate": 6.2e-05,
193
+ "loss": 1.7728,
194
+ "step": 31
195
+ },
196
+ {
197
+ "epoch": 1.81,
198
+ "learning_rate": 6.400000000000001e-05,
199
+ "loss": 1.8335,
200
+ "step": 32
201
+ },
202
+ {
203
+ "epoch": 1.87,
204
+ "learning_rate": 6.6e-05,
205
+ "loss": 1.7173,
206
+ "step": 33
207
+ },
208
+ {
209
+ "epoch": 1.93,
210
+ "learning_rate": 6.800000000000001e-05,
211
+ "loss": 1.5278,
212
+ "step": 34
213
+ },
214
+ {
215
+ "epoch": 1.98,
216
+ "learning_rate": 7e-05,
217
+ "loss": 1.8614,
218
+ "step": 35
219
+ },
220
+ {
221
+ "epoch": 2.04,
222
+ "learning_rate": 7.2e-05,
223
+ "loss": 1.7847,
224
+ "step": 36
225
+ },
226
+ {
227
+ "epoch": 2.1,
228
+ "learning_rate": 7.4e-05,
229
+ "loss": 1.7366,
230
+ "step": 37
231
+ },
232
+ {
233
+ "epoch": 2.15,
234
+ "learning_rate": 7.6e-05,
235
+ "loss": 1.5379,
236
+ "step": 38
237
+ },
238
+ {
239
+ "epoch": 2.21,
240
+ "learning_rate": 7.800000000000001e-05,
241
+ "loss": 1.6856,
242
+ "step": 39
243
+ },
244
+ {
245
+ "epoch": 2.27,
246
+ "learning_rate": 8e-05,
247
+ "loss": 1.6452,
248
+ "step": 40
249
+ },
250
+ {
251
+ "epoch": 2.32,
252
+ "learning_rate": 8.2e-05,
253
+ "loss": 1.8233,
254
+ "step": 41
255
+ },
256
+ {
257
+ "epoch": 2.38,
258
+ "learning_rate": 8.4e-05,
259
+ "loss": 1.8079,
260
+ "step": 42
261
+ },
262
+ {
263
+ "epoch": 2.44,
264
+ "learning_rate": 8.6e-05,
265
+ "loss": 1.7128,
266
+ "step": 43
267
+ },
268
+ {
269
+ "epoch": 2.49,
270
+ "learning_rate": 8.800000000000001e-05,
271
+ "loss": 1.8746,
272
+ "step": 44
273
+ },
274
+ {
275
+ "epoch": 2.55,
276
+ "learning_rate": 9e-05,
277
+ "loss": 1.6117,
278
+ "step": 45
279
+ },
280
+ {
281
+ "epoch": 2.61,
282
+ "learning_rate": 9.200000000000001e-05,
283
+ "loss": 1.6552,
284
+ "step": 46
285
+ },
286
+ {
287
+ "epoch": 2.66,
288
+ "learning_rate": 9.4e-05,
289
+ "loss": 1.6429,
290
+ "step": 47
291
+ },
292
+ {
293
+ "epoch": 2.72,
294
+ "learning_rate": 9.6e-05,
295
+ "loss": 1.5429,
296
+ "step": 48
297
+ },
298
+ {
299
+ "epoch": 2.78,
300
+ "learning_rate": 9.8e-05,
301
+ "loss": 1.6358,
302
+ "step": 49
303
+ },
304
+ {
305
+ "epoch": 2.83,
306
+ "learning_rate": 0.0001,
307
+ "loss": 1.8581,
308
+ "step": 50
309
+ },
310
+ {
311
+ "epoch": 2.89,
312
+ "learning_rate": 0.00010200000000000001,
313
+ "loss": 1.6794,
314
+ "step": 51
315
+ },
316
+ {
317
+ "epoch": 2.95,
318
+ "learning_rate": 0.00010400000000000001,
319
+ "loss": 1.683,
320
+ "step": 52
321
+ },
322
+ {
323
+ "epoch": 3.0,
324
+ "learning_rate": 0.00010600000000000002,
325
+ "loss": 1.7204,
326
+ "step": 53
327
+ },
328
+ {
329
+ "epoch": 3.06,
330
+ "learning_rate": 0.00010800000000000001,
331
+ "loss": 1.7068,
332
+ "step": 54
333
+ },
334
+ {
335
+ "epoch": 3.12,
336
+ "learning_rate": 0.00011000000000000002,
337
+ "loss": 1.5245,
338
+ "step": 55
339
+ },
340
+ {
341
+ "epoch": 3.17,
342
+ "learning_rate": 0.00011200000000000001,
343
+ "loss": 1.5683,
344
+ "step": 56
345
+ },
346
+ {
347
+ "epoch": 3.23,
348
+ "learning_rate": 0.00011399999999999999,
349
+ "loss": 1.8291,
350
+ "step": 57
351
+ },
352
+ {
353
+ "epoch": 3.28,
354
+ "learning_rate": 0.000116,
355
+ "loss": 1.625,
356
+ "step": 58
357
+ },
358
+ {
359
+ "epoch": 3.34,
360
+ "learning_rate": 0.000118,
361
+ "loss": 1.7945,
362
+ "step": 59
363
+ },
364
+ {
365
+ "epoch": 3.4,
366
+ "learning_rate": 0.00012,
367
+ "loss": 1.5244,
368
+ "step": 60
369
+ },
370
+ {
371
+ "epoch": 3.45,
372
+ "learning_rate": 0.000122,
373
+ "loss": 1.6436,
374
+ "step": 61
375
+ },
376
+ {
377
+ "epoch": 3.51,
378
+ "learning_rate": 0.000124,
379
+ "loss": 1.6679,
380
+ "step": 62
381
+ },
382
+ {
383
+ "epoch": 3.57,
384
+ "learning_rate": 0.000126,
385
+ "loss": 1.7644,
386
+ "step": 63
387
+ },
388
+ {
389
+ "epoch": 3.62,
390
+ "learning_rate": 0.00012800000000000002,
391
+ "loss": 1.6776,
392
+ "step": 64
393
+ },
394
+ {
395
+ "epoch": 3.68,
396
+ "learning_rate": 0.00013000000000000002,
397
+ "loss": 1.627,
398
+ "step": 65
399
+ },
400
+ {
401
+ "epoch": 3.74,
402
+ "learning_rate": 0.000132,
403
+ "loss": 1.5084,
404
+ "step": 66
405
+ },
406
+ {
407
+ "epoch": 3.79,
408
+ "learning_rate": 0.000134,
409
+ "loss": 1.6967,
410
+ "step": 67
411
+ },
412
+ {
413
+ "epoch": 3.85,
414
+ "learning_rate": 0.00013600000000000003,
415
+ "loss": 1.5249,
416
+ "step": 68
417
+ },
418
+ {
419
+ "epoch": 3.91,
420
+ "learning_rate": 0.000138,
421
+ "loss": 1.6821,
422
+ "step": 69
423
+ },
424
+ {
425
+ "epoch": 3.96,
426
+ "learning_rate": 0.00014,
427
+ "loss": 1.5071,
428
+ "step": 70
429
+ },
430
+ {
431
+ "epoch": 4.02,
432
+ "learning_rate": 0.000142,
433
+ "loss": 1.7624,
434
+ "step": 71
435
+ },
436
+ {
437
+ "epoch": 4.08,
438
+ "learning_rate": 0.000144,
439
+ "loss": 1.6826,
440
+ "step": 72
441
+ },
442
+ {
443
+ "epoch": 4.13,
444
+ "learning_rate": 0.000146,
445
+ "loss": 1.6488,
446
+ "step": 73
447
+ },
448
+ {
449
+ "epoch": 4.19,
450
+ "learning_rate": 0.000148,
451
+ "loss": 1.5138,
452
+ "step": 74
453
+ },
454
+ {
455
+ "epoch": 4.25,
456
+ "learning_rate": 0.00015000000000000001,
457
+ "loss": 1.6149,
458
+ "step": 75
459
+ },
460
+ {
461
+ "epoch": 4.3,
462
+ "learning_rate": 0.000152,
463
+ "loss": 1.4464,
464
+ "step": 76
465
+ },
466
+ {
467
+ "epoch": 4.36,
468
+ "learning_rate": 0.000154,
469
+ "loss": 1.4529,
470
+ "step": 77
471
+ },
472
+ {
473
+ "epoch": 4.42,
474
+ "learning_rate": 0.00015600000000000002,
475
+ "loss": 1.3768,
476
+ "step": 78
477
+ },
478
+ {
479
+ "epoch": 4.47,
480
+ "learning_rate": 0.00015800000000000002,
481
+ "loss": 1.4852,
482
+ "step": 79
483
+ },
484
+ {
485
+ "epoch": 4.53,
486
+ "learning_rate": 0.00016,
487
+ "loss": 1.6161,
488
+ "step": 80
489
+ },
490
+ {
491
+ "epoch": 4.59,
492
+ "learning_rate": 0.000162,
493
+ "loss": 1.749,
494
+ "step": 81
495
+ },
496
+ {
497
+ "epoch": 4.64,
498
+ "learning_rate": 0.000164,
499
+ "loss": 1.7293,
500
+ "step": 82
501
+ },
502
+ {
503
+ "epoch": 4.7,
504
+ "learning_rate": 0.000166,
505
+ "loss": 1.6095,
506
+ "step": 83
507
+ },
508
+ {
509
+ "epoch": 4.76,
510
+ "learning_rate": 0.000168,
511
+ "loss": 1.6558,
512
+ "step": 84
513
+ },
514
+ {
515
+ "epoch": 4.81,
516
+ "learning_rate": 0.00017,
517
+ "loss": 1.7317,
518
+ "step": 85
519
+ },
520
+ {
521
+ "epoch": 4.87,
522
+ "learning_rate": 0.000172,
523
+ "loss": 1.6435,
524
+ "step": 86
525
+ },
526
+ {
527
+ "epoch": 4.93,
528
+ "learning_rate": 0.000174,
529
+ "loss": 1.8289,
530
+ "step": 87
531
+ },
532
+ {
533
+ "epoch": 4.98,
534
+ "learning_rate": 0.00017600000000000002,
535
+ "loss": 1.4689,
536
+ "step": 88
537
+ },
538
+ {
539
+ "epoch": 5.04,
540
+ "learning_rate": 0.00017800000000000002,
541
+ "loss": 1.728,
542
+ "step": 89
543
+ },
544
+ {
545
+ "epoch": 5.1,
546
+ "learning_rate": 0.00018,
547
+ "loss": 1.5481,
548
+ "step": 90
549
+ },
550
+ {
551
+ "epoch": 5.15,
552
+ "learning_rate": 0.000182,
553
+ "loss": 1.6498,
554
+ "step": 91
555
+ },
556
+ {
557
+ "epoch": 5.21,
558
+ "learning_rate": 0.00018400000000000003,
559
+ "loss": 1.5319,
560
+ "step": 92
561
+ },
562
+ {
563
+ "epoch": 5.27,
564
+ "learning_rate": 0.00018600000000000002,
565
+ "loss": 1.6678,
566
+ "step": 93
567
+ },
568
+ {
569
+ "epoch": 5.32,
570
+ "learning_rate": 0.000188,
571
+ "loss": 1.4058,
572
+ "step": 94
573
+ },
574
+ {
575
+ "epoch": 5.38,
576
+ "learning_rate": 0.00019,
577
+ "loss": 1.6551,
578
+ "step": 95
579
+ },
580
+ {
581
+ "epoch": 5.44,
582
+ "learning_rate": 0.000192,
583
+ "loss": 1.5329,
584
+ "step": 96
585
+ },
586
+ {
587
+ "epoch": 5.49,
588
+ "learning_rate": 0.000194,
589
+ "loss": 1.4911,
590
+ "step": 97
591
+ },
592
+ {
593
+ "epoch": 5.55,
594
+ "learning_rate": 0.000196,
595
+ "loss": 1.796,
596
+ "step": 98
597
+ },
598
+ {
599
+ "epoch": 5.61,
600
+ "learning_rate": 0.00019800000000000002,
601
+ "loss": 1.6878,
602
+ "step": 99
603
+ },
604
+ {
605
+ "epoch": 5.66,
606
+ "learning_rate": 0.0002,
607
+ "loss": 1.6483,
608
+ "step": 100
609
+ },
610
+ {
611
+ "epoch": 5.72,
612
+ "learning_rate": 0.00019987500000000002,
613
+ "loss": 1.4458,
614
+ "step": 101
615
+ },
616
+ {
617
+ "epoch": 5.78,
618
+ "learning_rate": 0.00019975,
619
+ "loss": 1.4612,
620
+ "step": 102
621
+ },
622
+ {
623
+ "epoch": 5.83,
624
+ "learning_rate": 0.00019962500000000001,
625
+ "loss": 1.6,
626
+ "step": 103
627
+ },
628
+ {
629
+ "epoch": 5.89,
630
+ "learning_rate": 0.00019950000000000002,
631
+ "loss": 1.5416,
632
+ "step": 104
633
+ },
634
+ {
635
+ "epoch": 5.95,
636
+ "learning_rate": 0.000199375,
637
+ "loss": 1.614,
638
+ "step": 105
639
+ },
640
+ {
641
+ "epoch": 6.0,
642
+ "learning_rate": 0.00019925,
643
+ "loss": 1.5744,
644
+ "step": 106
645
+ },
646
+ {
647
+ "epoch": 6.06,
648
+ "learning_rate": 0.000199125,
649
+ "loss": 1.4321,
650
+ "step": 107
651
+ },
652
+ {
653
+ "epoch": 6.12,
654
+ "learning_rate": 0.000199,
655
+ "loss": 1.5335,
656
+ "step": 108
657
+ },
658
+ {
659
+ "epoch": 6.17,
660
+ "learning_rate": 0.00019887500000000002,
661
+ "loss": 1.4739,
662
+ "step": 109
663
+ },
664
+ {
665
+ "epoch": 6.23,
666
+ "learning_rate": 0.00019875,
667
+ "loss": 1.4072,
668
+ "step": 110
669
+ },
670
+ {
671
+ "epoch": 6.29,
672
+ "learning_rate": 0.00019862500000000002,
673
+ "loss": 1.3836,
674
+ "step": 111
675
+ },
676
+ {
677
+ "epoch": 6.34,
678
+ "learning_rate": 0.00019850000000000003,
679
+ "loss": 1.5784,
680
+ "step": 112
681
+ },
682
+ {
683
+ "epoch": 6.4,
684
+ "learning_rate": 0.000198375,
685
+ "loss": 1.7039,
686
+ "step": 113
687
+ },
688
+ {
689
+ "epoch": 6.46,
690
+ "learning_rate": 0.00019825,
691
+ "loss": 1.587,
692
+ "step": 114
693
+ },
694
+ {
695
+ "epoch": 6.51,
696
+ "learning_rate": 0.000198125,
697
+ "loss": 1.4426,
698
+ "step": 115
699
+ },
700
+ {
701
+ "epoch": 6.57,
702
+ "learning_rate": 0.00019800000000000002,
703
+ "loss": 1.514,
704
+ "step": 116
705
+ },
706
+ {
707
+ "epoch": 6.63,
708
+ "learning_rate": 0.000197875,
709
+ "loss": 1.4913,
710
+ "step": 117
711
+ },
712
+ {
713
+ "epoch": 6.68,
714
+ "learning_rate": 0.00019775,
715
+ "loss": 1.5138,
716
+ "step": 118
717
+ },
718
+ {
719
+ "epoch": 6.74,
720
+ "learning_rate": 0.00019762500000000002,
721
+ "loss": 1.4991,
722
+ "step": 119
723
+ },
724
+ {
725
+ "epoch": 6.8,
726
+ "learning_rate": 0.00019750000000000003,
727
+ "loss": 1.6112,
728
+ "step": 120
729
+ },
730
+ {
731
+ "epoch": 6.85,
732
+ "learning_rate": 0.00019737499999999999,
733
+ "loss": 1.6294,
734
+ "step": 121
735
+ },
736
+ {
737
+ "epoch": 6.91,
738
+ "learning_rate": 0.00019725,
739
+ "loss": 1.6799,
740
+ "step": 122
741
+ },
742
+ {
743
+ "epoch": 6.97,
744
+ "learning_rate": 0.000197125,
745
+ "loss": 1.6058,
746
+ "step": 123
747
+ },
748
+ {
749
+ "epoch": 7.02,
750
+ "learning_rate": 0.00019700000000000002,
751
+ "loss": 1.7086,
752
+ "step": 124
753
+ },
754
+ {
755
+ "epoch": 7.08,
756
+ "learning_rate": 0.000196875,
757
+ "loss": 1.659,
758
+ "step": 125
759
+ },
760
+ {
761
+ "epoch": 7.14,
762
+ "learning_rate": 0.00019675,
763
+ "loss": 1.4066,
764
+ "step": 126
765
+ },
766
+ {
767
+ "epoch": 7.19,
768
+ "learning_rate": 0.00019662500000000002,
769
+ "loss": 1.4265,
770
+ "step": 127
771
+ },
772
+ {
773
+ "epoch": 7.25,
774
+ "learning_rate": 0.0001965,
775
+ "loss": 1.5891,
776
+ "step": 128
777
+ },
778
+ {
779
+ "epoch": 7.31,
780
+ "learning_rate": 0.00019637500000000002,
781
+ "loss": 1.6088,
782
+ "step": 129
783
+ },
784
+ {
785
+ "epoch": 7.36,
786
+ "learning_rate": 0.00019625,
787
+ "loss": 1.5274,
788
+ "step": 130
789
+ },
790
+ {
791
+ "epoch": 7.42,
792
+ "learning_rate": 0.000196125,
793
+ "loss": 1.5988,
794
+ "step": 131
795
+ },
796
+ {
797
+ "epoch": 7.48,
798
+ "learning_rate": 0.000196,
799
+ "loss": 1.6922,
800
+ "step": 132
801
+ },
802
+ {
803
+ "epoch": 7.53,
804
+ "learning_rate": 0.000195875,
805
+ "loss": 1.4708,
806
+ "step": 133
807
+ },
808
+ {
809
+ "epoch": 7.59,
810
+ "learning_rate": 0.00019575000000000001,
811
+ "loss": 1.3907,
812
+ "step": 134
813
+ },
814
+ {
815
+ "epoch": 7.65,
816
+ "learning_rate": 0.00019562500000000003,
817
+ "loss": 1.5097,
818
+ "step": 135
819
+ },
820
+ {
821
+ "epoch": 7.7,
822
+ "learning_rate": 0.0001955,
823
+ "loss": 1.5843,
824
+ "step": 136
825
+ },
826
+ {
827
+ "epoch": 7.76,
828
+ "learning_rate": 0.00019537500000000002,
829
+ "loss": 1.5949,
830
+ "step": 137
831
+ },
832
+ {
833
+ "epoch": 7.82,
834
+ "learning_rate": 0.00019525,
835
+ "loss": 1.5275,
836
+ "step": 138
837
+ },
838
+ {
839
+ "epoch": 7.87,
840
+ "learning_rate": 0.000195125,
841
+ "loss": 1.3971,
842
+ "step": 139
843
+ },
844
+ {
845
+ "epoch": 7.93,
846
+ "learning_rate": 0.000195,
847
+ "loss": 1.6457,
848
+ "step": 140
849
+ },
850
+ {
851
+ "epoch": 7.99,
852
+ "learning_rate": 0.000194875,
853
+ "loss": 1.4983,
854
+ "step": 141
855
+ },
856
+ {
857
+ "epoch": 8.04,
858
+ "learning_rate": 0.00019475000000000002,
859
+ "loss": 1.3942,
860
+ "step": 142
861
+ },
862
+ {
863
+ "epoch": 8.1,
864
+ "learning_rate": 0.000194625,
865
+ "loss": 1.3633,
866
+ "step": 143
867
+ },
868
+ {
869
+ "epoch": 8.16,
870
+ "learning_rate": 0.0001945,
871
+ "loss": 1.3908,
872
+ "step": 144
873
+ },
874
+ {
875
+ "epoch": 8.21,
876
+ "learning_rate": 0.00019437500000000002,
877
+ "loss": 1.4746,
878
+ "step": 145
879
+ },
880
+ {
881
+ "epoch": 8.27,
882
+ "learning_rate": 0.00019425,
883
+ "loss": 1.4776,
884
+ "step": 146
885
+ },
886
+ {
887
+ "epoch": 8.33,
888
+ "learning_rate": 0.000194125,
889
+ "loss": 1.5559,
890
+ "step": 147
891
+ },
892
+ {
893
+ "epoch": 8.38,
894
+ "learning_rate": 0.000194,
895
+ "loss": 1.5798,
896
+ "step": 148
897
+ },
898
+ {
899
+ "epoch": 8.44,
900
+ "learning_rate": 0.000193875,
901
+ "loss": 1.5173,
902
+ "step": 149
903
+ },
904
+ {
905
+ "epoch": 8.5,
906
+ "learning_rate": 0.00019375000000000002,
907
+ "loss": 1.4286,
908
+ "step": 150
909
+ },
910
+ {
911
+ "epoch": 8.55,
912
+ "learning_rate": 0.000193625,
913
+ "loss": 1.4609,
914
+ "step": 151
915
+ },
916
+ {
917
+ "epoch": 8.61,
918
+ "learning_rate": 0.00019350000000000001,
919
+ "loss": 1.361,
920
+ "step": 152
921
+ },
922
+ {
923
+ "epoch": 8.67,
924
+ "learning_rate": 0.00019337500000000002,
925
+ "loss": 1.6866,
926
+ "step": 153
927
+ },
928
+ {
929
+ "epoch": 8.72,
930
+ "learning_rate": 0.00019325,
931
+ "loss": 1.5571,
932
+ "step": 154
933
+ },
934
+ {
935
+ "epoch": 8.78,
936
+ "learning_rate": 0.000193125,
937
+ "loss": 1.4351,
938
+ "step": 155
939
+ },
940
+ {
941
+ "epoch": 8.84,
942
+ "learning_rate": 0.000193,
943
+ "loss": 1.444,
944
+ "step": 156
945
+ },
946
+ {
947
+ "epoch": 8.89,
948
+ "learning_rate": 0.000192875,
949
+ "loss": 1.5379,
950
+ "step": 157
951
+ },
952
+ {
953
+ "epoch": 8.95,
954
+ "learning_rate": 0.00019275,
955
+ "loss": 1.4775,
956
+ "step": 158
957
+ },
958
+ {
959
+ "epoch": 9.01,
960
+ "learning_rate": 0.000192625,
961
+ "loss": 1.5141,
962
+ "step": 159
963
+ },
964
+ {
965
+ "epoch": 9.06,
966
+ "learning_rate": 0.00019250000000000002,
967
+ "loss": 1.491,
968
+ "step": 160
969
+ },
970
+ {
971
+ "epoch": 9.12,
972
+ "learning_rate": 0.00019237500000000003,
973
+ "loss": 1.6459,
974
+ "step": 161
975
+ },
976
+ {
977
+ "epoch": 9.18,
978
+ "learning_rate": 0.00019225,
979
+ "loss": 1.4853,
980
+ "step": 162
981
+ },
982
+ {
983
+ "epoch": 9.23,
984
+ "learning_rate": 0.000192125,
985
+ "loss": 1.5958,
986
+ "step": 163
987
+ },
988
+ {
989
+ "epoch": 9.29,
990
+ "learning_rate": 0.000192,
991
+ "loss": 1.228,
992
+ "step": 164
993
+ },
994
+ {
995
+ "epoch": 9.35,
996
+ "learning_rate": 0.00019187500000000002,
997
+ "loss": 1.3625,
998
+ "step": 165
999
+ },
1000
+ {
1001
+ "epoch": 9.4,
1002
+ "learning_rate": 0.00019175,
1003
+ "loss": 1.3645,
1004
+ "step": 166
1005
+ },
1006
+ {
1007
+ "epoch": 9.46,
1008
+ "learning_rate": 0.000191625,
1009
+ "loss": 1.4625,
1010
+ "step": 167
1011
+ },
1012
+ {
1013
+ "epoch": 9.52,
1014
+ "learning_rate": 0.00019150000000000002,
1015
+ "loss": 1.4005,
1016
+ "step": 168
1017
+ },
1018
+ {
1019
+ "epoch": 9.57,
1020
+ "learning_rate": 0.000191375,
1021
+ "loss": 1.4099,
1022
+ "step": 169
1023
+ },
1024
+ {
1025
+ "epoch": 9.63,
1026
+ "learning_rate": 0.00019125000000000001,
1027
+ "loss": 1.5099,
1028
+ "step": 170
1029
+ },
1030
+ {
1031
+ "epoch": 9.68,
1032
+ "learning_rate": 0.000191125,
1033
+ "loss": 1.5183,
1034
+ "step": 171
1035
+ },
1036
+ {
1037
+ "epoch": 9.74,
1038
+ "learning_rate": 0.000191,
1039
+ "loss": 1.5149,
1040
+ "step": 172
1041
+ },
1042
+ {
1043
+ "epoch": 9.8,
1044
+ "learning_rate": 0.000190875,
1045
+ "loss": 1.2841,
1046
+ "step": 173
1047
+ },
1048
+ {
1049
+ "epoch": 9.85,
1050
+ "learning_rate": 0.00019075,
1051
+ "loss": 1.6436,
1052
+ "step": 174
1053
+ },
1054
+ {
1055
+ "epoch": 9.91,
1056
+ "learning_rate": 0.000190625,
1057
+ "loss": 1.4587,
1058
+ "step": 175
1059
+ },
1060
+ {
1061
+ "epoch": 9.97,
1062
+ "learning_rate": 0.00019050000000000002,
1063
+ "loss": 1.7131,
1064
+ "step": 176
1065
+ },
1066
+ {
1067
+ "epoch": 10.02,
1068
+ "learning_rate": 0.000190375,
1069
+ "loss": 1.5204,
1070
+ "step": 177
1071
+ },
1072
+ {
1073
+ "epoch": 10.08,
1074
+ "learning_rate": 0.00019025000000000002,
1075
+ "loss": 1.2174,
1076
+ "step": 178
1077
+ },
1078
+ {
1079
+ "epoch": 10.14,
1080
+ "learning_rate": 0.00019012500000000003,
1081
+ "loss": 1.3831,
1082
+ "step": 179
1083
+ },
1084
+ {
1085
+ "epoch": 10.19,
1086
+ "learning_rate": 0.00019,
1087
+ "loss": 1.1915,
1088
+ "step": 180
1089
+ },
1090
+ {
1091
+ "epoch": 10.25,
1092
+ "learning_rate": 0.000189875,
1093
+ "loss": 1.4282,
1094
+ "step": 181
1095
+ },
1096
+ {
1097
+ "epoch": 10.31,
1098
+ "learning_rate": 0.00018975,
1099
+ "loss": 1.4675,
1100
+ "step": 182
1101
+ },
1102
+ {
1103
+ "epoch": 10.36,
1104
+ "learning_rate": 0.00018962500000000001,
1105
+ "loss": 1.4183,
1106
+ "step": 183
1107
+ },
1108
+ {
1109
+ "epoch": 10.42,
1110
+ "learning_rate": 0.0001895,
1111
+ "loss": 1.6309,
1112
+ "step": 184
1113
+ },
1114
+ {
1115
+ "epoch": 10.48,
1116
+ "learning_rate": 0.000189375,
1117
+ "loss": 1.4368,
1118
+ "step": 185
1119
+ },
1120
+ {
1121
+ "epoch": 10.53,
1122
+ "learning_rate": 0.00018925000000000002,
1123
+ "loss": 1.507,
1124
+ "step": 186
1125
+ },
1126
+ {
1127
+ "epoch": 10.59,
1128
+ "learning_rate": 0.00018912500000000003,
1129
+ "loss": 1.4642,
1130
+ "step": 187
1131
+ },
1132
+ {
1133
+ "epoch": 10.65,
1134
+ "learning_rate": 0.00018899999999999999,
1135
+ "loss": 1.5397,
1136
+ "step": 188
1137
+ },
1138
+ {
1139
+ "epoch": 10.7,
1140
+ "learning_rate": 0.000188875,
1141
+ "loss": 1.4003,
1142
+ "step": 189
1143
+ },
1144
+ {
1145
+ "epoch": 10.76,
1146
+ "learning_rate": 0.00018875,
1147
+ "loss": 1.5798,
1148
+ "step": 190
1149
+ },
1150
+ {
1151
+ "epoch": 10.82,
1152
+ "learning_rate": 0.00018862500000000002,
1153
+ "loss": 1.4797,
1154
+ "step": 191
1155
+ },
1156
+ {
1157
+ "epoch": 10.87,
1158
+ "learning_rate": 0.0001885,
1159
+ "loss": 1.429,
1160
+ "step": 192
1161
+ },
1162
+ {
1163
+ "epoch": 10.93,
1164
+ "learning_rate": 0.000188375,
1165
+ "loss": 1.49,
1166
+ "step": 193
1167
+ },
1168
+ {
1169
+ "epoch": 10.99,
1170
+ "learning_rate": 0.00018825000000000002,
1171
+ "loss": 1.5075,
1172
+ "step": 194
1173
+ },
1174
+ {
1175
+ "epoch": 11.04,
1176
+ "learning_rate": 0.000188125,
1177
+ "loss": 1.4662,
1178
+ "step": 195
1179
+ },
1180
+ {
1181
+ "epoch": 11.1,
1182
+ "learning_rate": 0.000188,
1183
+ "loss": 1.4103,
1184
+ "step": 196
1185
+ },
1186
+ {
1187
+ "epoch": 11.16,
1188
+ "learning_rate": 0.000187875,
1189
+ "loss": 1.4858,
1190
+ "step": 197
1191
+ },
1192
+ {
1193
+ "epoch": 11.21,
1194
+ "learning_rate": 0.00018775,
1195
+ "loss": 1.4841,
1196
+ "step": 198
1197
+ },
1198
+ {
1199
+ "epoch": 11.27,
1200
+ "learning_rate": 0.000187625,
1201
+ "loss": 1.4102,
1202
+ "step": 199
1203
+ },
1204
+ {
1205
+ "epoch": 11.33,
1206
+ "learning_rate": 0.0001875,
1207
+ "loss": 1.3401,
1208
+ "step": 200
1209
+ },
1210
+ {
1211
+ "epoch": 11.33,
1212
+ "eval_loss": 2.023402452468872,
1213
+ "eval_runtime": 59.5382,
1214
+ "eval_samples_per_second": 4.216,
1215
+ "eval_steps_per_second": 1.058,
1216
+ "step": 200
1217
+ },
1218
+ {
1219
+ "epoch": 11.38,
1220
+ "learning_rate": 0.00018737500000000001,
1221
+ "loss": 1.3293,
1222
+ "step": 201
1223
+ },
1224
+ {
1225
+ "epoch": 11.44,
1226
+ "learning_rate": 0.00018725000000000002,
1227
+ "loss": 1.5936,
1228
+ "step": 202
1229
+ },
1230
+ {
1231
+ "epoch": 11.5,
1232
+ "learning_rate": 0.000187125,
1233
+ "loss": 1.4509,
1234
+ "step": 203
1235
+ },
1236
+ {
1237
+ "epoch": 11.55,
1238
+ "learning_rate": 0.00018700000000000002,
1239
+ "loss": 1.4002,
1240
+ "step": 204
1241
+ },
1242
+ {
1243
+ "epoch": 11.61,
1244
+ "learning_rate": 0.000186875,
1245
+ "loss": 1.4654,
1246
+ "step": 205
1247
+ },
1248
+ {
1249
+ "epoch": 11.67,
1250
+ "learning_rate": 0.00018675,
1251
+ "loss": 1.456,
1252
+ "step": 206
1253
+ },
1254
+ {
1255
+ "epoch": 11.72,
1256
+ "learning_rate": 0.000186625,
1257
+ "loss": 1.5051,
1258
+ "step": 207
1259
+ },
1260
+ {
1261
+ "epoch": 11.78,
1262
+ "learning_rate": 0.0001865,
1263
+ "loss": 1.4447,
1264
+ "step": 208
1265
+ },
1266
+ {
1267
+ "epoch": 11.84,
1268
+ "learning_rate": 0.00018637500000000002,
1269
+ "loss": 1.4817,
1270
+ "step": 209
1271
+ },
1272
+ {
1273
+ "epoch": 11.89,
1274
+ "learning_rate": 0.00018625,
1275
+ "loss": 1.3649,
1276
+ "step": 210
1277
+ },
1278
+ {
1279
+ "epoch": 11.95,
1280
+ "learning_rate": 0.000186125,
1281
+ "loss": 1.4798,
1282
+ "step": 211
1283
+ },
1284
+ {
1285
+ "epoch": 12.01,
1286
+ "learning_rate": 0.00018600000000000002,
1287
+ "loss": 1.469,
1288
+ "step": 212
1289
+ },
1290
+ {
1291
+ "epoch": 12.06,
1292
+ "learning_rate": 0.000185875,
1293
+ "loss": 1.4963,
1294
+ "step": 213
1295
+ },
1296
+ {
1297
+ "epoch": 12.12,
1298
+ "learning_rate": 0.00018575,
1299
+ "loss": 1.377,
1300
+ "step": 214
1301
+ },
1302
+ {
1303
+ "epoch": 12.18,
1304
+ "learning_rate": 0.000185625,
1305
+ "loss": 1.6133,
1306
+ "step": 215
1307
+ },
1308
+ {
1309
+ "epoch": 12.23,
1310
+ "learning_rate": 0.0001855,
1311
+ "loss": 1.3288,
1312
+ "step": 216
1313
+ },
1314
+ {
1315
+ "epoch": 12.29,
1316
+ "learning_rate": 0.00018537500000000002,
1317
+ "loss": 1.514,
1318
+ "step": 217
1319
+ },
1320
+ {
1321
+ "epoch": 12.35,
1322
+ "learning_rate": 0.00018525,
1323
+ "loss": 1.4412,
1324
+ "step": 218
1325
+ },
1326
+ {
1327
+ "epoch": 12.4,
1328
+ "learning_rate": 0.00018512500000000001,
1329
+ "loss": 1.4004,
1330
+ "step": 219
1331
+ },
1332
+ {
1333
+ "epoch": 12.46,
1334
+ "learning_rate": 0.00018500000000000002,
1335
+ "loss": 1.359,
1336
+ "step": 220
1337
+ },
1338
+ {
1339
+ "epoch": 12.52,
1340
+ "learning_rate": 0.000184875,
1341
+ "loss": 1.3539,
1342
+ "step": 221
1343
+ },
1344
+ {
1345
+ "epoch": 12.57,
1346
+ "learning_rate": 0.00018475,
1347
+ "loss": 1.3255,
1348
+ "step": 222
1349
+ },
1350
+ {
1351
+ "epoch": 12.63,
1352
+ "learning_rate": 0.000184625,
1353
+ "loss": 1.4911,
1354
+ "step": 223
1355
+ },
1356
+ {
1357
+ "epoch": 12.69,
1358
+ "learning_rate": 0.0001845,
1359
+ "loss": 1.4146,
1360
+ "step": 224
1361
+ },
1362
+ {
1363
+ "epoch": 12.74,
1364
+ "learning_rate": 0.000184375,
1365
+ "loss": 1.3291,
1366
+ "step": 225
1367
+ },
1368
+ {
1369
+ "epoch": 12.8,
1370
+ "learning_rate": 0.00018425,
1371
+ "loss": 1.356,
1372
+ "step": 226
1373
+ },
1374
+ {
1375
+ "epoch": 12.86,
1376
+ "learning_rate": 0.00018412500000000002,
1377
+ "loss": 1.4216,
1378
+ "step": 227
1379
+ },
1380
+ {
1381
+ "epoch": 12.91,
1382
+ "learning_rate": 0.00018400000000000003,
1383
+ "loss": 1.4282,
1384
+ "step": 228
1385
+ },
1386
+ {
1387
+ "epoch": 12.97,
1388
+ "learning_rate": 0.000183875,
1389
+ "loss": 1.354,
1390
+ "step": 229
1391
+ },
1392
+ {
1393
+ "epoch": 13.03,
1394
+ "learning_rate": 0.00018375,
1395
+ "loss": 1.4358,
1396
+ "step": 230
1397
+ },
1398
+ {
1399
+ "epoch": 13.08,
1400
+ "learning_rate": 0.000183625,
1401
+ "loss": 1.5088,
1402
+ "step": 231
1403
+ },
1404
+ {
1405
+ "epoch": 13.14,
1406
+ "learning_rate": 0.00018350000000000002,
1407
+ "loss": 1.3588,
1408
+ "step": 232
1409
+ },
1410
+ {
1411
+ "epoch": 13.2,
1412
+ "learning_rate": 0.000183375,
1413
+ "loss": 1.3718,
1414
+ "step": 233
1415
+ },
1416
+ {
1417
+ "epoch": 13.25,
1418
+ "learning_rate": 0.00018325,
1419
+ "loss": 1.4119,
1420
+ "step": 234
1421
+ },
1422
+ {
1423
+ "epoch": 13.31,
1424
+ "learning_rate": 0.00018312500000000002,
1425
+ "loss": 1.2867,
1426
+ "step": 235
1427
+ },
1428
+ {
1429
+ "epoch": 13.37,
1430
+ "learning_rate": 0.000183,
1431
+ "loss": 1.2827,
1432
+ "step": 236
1433
+ },
1434
+ {
1435
+ "epoch": 13.42,
1436
+ "learning_rate": 0.000182875,
1437
+ "loss": 1.4267,
1438
+ "step": 237
1439
+ },
1440
+ {
1441
+ "epoch": 13.48,
1442
+ "learning_rate": 0.00018275,
1443
+ "loss": 1.3174,
1444
+ "step": 238
1445
+ },
1446
+ {
1447
+ "epoch": 13.54,
1448
+ "learning_rate": 0.000182625,
1449
+ "loss": 1.4199,
1450
+ "step": 239
1451
+ },
1452
+ {
1453
+ "epoch": 13.59,
1454
+ "learning_rate": 0.0001825,
1455
+ "loss": 1.4644,
1456
+ "step": 240
1457
+ },
1458
+ {
1459
+ "epoch": 13.65,
1460
+ "learning_rate": 0.000182375,
1461
+ "loss": 1.3819,
1462
+ "step": 241
1463
+ },
1464
+ {
1465
+ "epoch": 13.71,
1466
+ "learning_rate": 0.00018225,
1467
+ "loss": 1.3101,
1468
+ "step": 242
1469
+ },
1470
+ {
1471
+ "epoch": 13.76,
1472
+ "learning_rate": 0.00018212500000000002,
1473
+ "loss": 1.353,
1474
+ "step": 243
1475
+ },
1476
+ {
1477
+ "epoch": 13.82,
1478
+ "learning_rate": 0.000182,
1479
+ "loss": 1.3191,
1480
+ "step": 244
1481
+ },
1482
+ {
1483
+ "epoch": 13.88,
1484
+ "learning_rate": 0.00018187500000000002,
1485
+ "loss": 1.3851,
1486
+ "step": 245
1487
+ },
1488
+ {
1489
+ "epoch": 13.93,
1490
+ "learning_rate": 0.00018175,
1491
+ "loss": 1.3599,
1492
+ "step": 246
1493
+ },
1494
+ {
1495
+ "epoch": 13.99,
1496
+ "learning_rate": 0.000181625,
1497
+ "loss": 1.4739,
1498
+ "step": 247
1499
+ },
1500
+ {
1501
+ "epoch": 14.05,
1502
+ "learning_rate": 0.0001815,
1503
+ "loss": 1.3944,
1504
+ "step": 248
1505
+ },
1506
+ {
1507
+ "epoch": 14.1,
1508
+ "learning_rate": 0.000181375,
1509
+ "loss": 1.208,
1510
+ "step": 249
1511
+ },
1512
+ {
1513
+ "epoch": 14.16,
1514
+ "learning_rate": 0.00018125000000000001,
1515
+ "loss": 1.4049,
1516
+ "step": 250
1517
+ },
1518
+ {
1519
+ "epoch": 14.22,
1520
+ "learning_rate": 0.000181125,
1521
+ "loss": 1.2871,
1522
+ "step": 251
1523
+ },
1524
+ {
1525
+ "epoch": 14.27,
1526
+ "learning_rate": 0.000181,
1527
+ "loss": 1.4616,
1528
+ "step": 252
1529
+ },
1530
+ {
1531
+ "epoch": 14.33,
1532
+ "learning_rate": 0.00018087500000000002,
1533
+ "loss": 1.3318,
1534
+ "step": 253
1535
+ },
1536
+ {
1537
+ "epoch": 14.39,
1538
+ "learning_rate": 0.00018075000000000003,
1539
+ "loss": 1.328,
1540
+ "step": 254
1541
+ },
1542
+ {
1543
+ "epoch": 14.44,
1544
+ "learning_rate": 0.000180625,
1545
+ "loss": 1.2645,
1546
+ "step": 255
1547
+ },
1548
+ {
1549
+ "epoch": 14.5,
1550
+ "learning_rate": 0.0001805,
1551
+ "loss": 1.2477,
1552
+ "step": 256
1553
+ },
1554
+ {
1555
+ "epoch": 14.56,
1556
+ "learning_rate": 0.000180375,
1557
+ "loss": 1.4021,
1558
+ "step": 257
1559
+ },
1560
+ {
1561
+ "epoch": 14.61,
1562
+ "learning_rate": 0.00018025000000000002,
1563
+ "loss": 1.424,
1564
+ "step": 258
1565
+ },
1566
+ {
1567
+ "epoch": 14.67,
1568
+ "learning_rate": 0.000180125,
1569
+ "loss": 1.3124,
1570
+ "step": 259
1571
+ },
1572
+ {
1573
+ "epoch": 14.73,
1574
+ "learning_rate": 0.00018,
1575
+ "loss": 1.3239,
1576
+ "step": 260
1577
+ },
1578
+ {
1579
+ "epoch": 14.78,
1580
+ "learning_rate": 0.00017987500000000002,
1581
+ "loss": 1.3568,
1582
+ "step": 261
1583
+ },
1584
+ {
1585
+ "epoch": 14.84,
1586
+ "learning_rate": 0.00017975,
1587
+ "loss": 1.1723,
1588
+ "step": 262
1589
+ },
1590
+ {
1591
+ "epoch": 14.9,
1592
+ "learning_rate": 0.000179625,
1593
+ "loss": 1.3399,
1594
+ "step": 263
1595
+ },
1596
+ {
1597
+ "epoch": 14.95,
1598
+ "learning_rate": 0.0001795,
1599
+ "loss": 1.3437,
1600
+ "step": 264
1601
+ },
1602
+ {
1603
+ "epoch": 15.01,
1604
+ "learning_rate": 0.000179375,
1605
+ "loss": 1.3229,
1606
+ "step": 265
1607
+ },
1608
+ {
1609
+ "epoch": 15.07,
1610
+ "learning_rate": 0.00017925000000000002,
1611
+ "loss": 1.2831,
1612
+ "step": 266
1613
+ },
1614
+ {
1615
+ "epoch": 15.12,
1616
+ "learning_rate": 0.000179125,
1617
+ "loss": 1.1986,
1618
+ "step": 267
1619
+ },
1620
+ {
1621
+ "epoch": 15.18,
1622
+ "learning_rate": 0.00017900000000000001,
1623
+ "loss": 1.236,
1624
+ "step": 268
1625
+ },
1626
+ {
1627
+ "epoch": 15.24,
1628
+ "learning_rate": 0.00017887500000000002,
1629
+ "loss": 1.3338,
1630
+ "step": 269
1631
+ },
1632
+ {
1633
+ "epoch": 15.29,
1634
+ "learning_rate": 0.00017875,
1635
+ "loss": 1.2406,
1636
+ "step": 270
1637
+ },
1638
+ {
1639
+ "epoch": 15.35,
1640
+ "learning_rate": 0.000178625,
1641
+ "loss": 1.1528,
1642
+ "step": 271
1643
+ },
1644
+ {
1645
+ "epoch": 15.41,
1646
+ "learning_rate": 0.0001785,
1647
+ "loss": 1.3401,
1648
+ "step": 272
1649
+ },
1650
+ {
1651
+ "epoch": 15.46,
1652
+ "learning_rate": 0.000178375,
1653
+ "loss": 1.3493,
1654
+ "step": 273
1655
+ },
1656
+ {
1657
+ "epoch": 15.52,
1658
+ "learning_rate": 0.00017825,
1659
+ "loss": 1.2721,
1660
+ "step": 274
1661
+ },
1662
+ {
1663
+ "epoch": 15.58,
1664
+ "learning_rate": 0.000178125,
1665
+ "loss": 1.3501,
1666
+ "step": 275
1667
+ },
1668
+ {
1669
+ "epoch": 15.63,
1670
+ "learning_rate": 0.00017800000000000002,
1671
+ "loss": 1.3662,
1672
+ "step": 276
1673
+ },
1674
+ {
1675
+ "epoch": 15.69,
1676
+ "learning_rate": 0.000177875,
1677
+ "loss": 1.2117,
1678
+ "step": 277
1679
+ },
1680
+ {
1681
+ "epoch": 15.75,
1682
+ "learning_rate": 0.00017775,
1683
+ "loss": 1.3798,
1684
+ "step": 278
1685
+ },
1686
+ {
1687
+ "epoch": 15.8,
1688
+ "learning_rate": 0.00017762500000000002,
1689
+ "loss": 1.3108,
1690
+ "step": 279
1691
+ },
1692
+ {
1693
+ "epoch": 15.86,
1694
+ "learning_rate": 0.0001775,
1695
+ "loss": 1.3565,
1696
+ "step": 280
1697
+ },
1698
+ {
1699
+ "epoch": 15.92,
1700
+ "learning_rate": 0.00017737500000000002,
1701
+ "loss": 1.3961,
1702
+ "step": 281
1703
+ },
1704
+ {
1705
+ "epoch": 15.97,
1706
+ "learning_rate": 0.00017725,
1707
+ "loss": 1.216,
1708
+ "step": 282
1709
+ },
1710
+ {
1711
+ "epoch": 16.03,
1712
+ "learning_rate": 0.000177125,
1713
+ "loss": 1.3027,
1714
+ "step": 283
1715
+ },
1716
+ {
1717
+ "epoch": 16.08,
1718
+ "learning_rate": 0.00017700000000000002,
1719
+ "loss": 1.1798,
1720
+ "step": 284
1721
+ },
1722
+ {
1723
+ "epoch": 16.14,
1724
+ "learning_rate": 0.000176875,
1725
+ "loss": 1.2375,
1726
+ "step": 285
1727
+ },
1728
+ {
1729
+ "epoch": 16.2,
1730
+ "learning_rate": 0.00017675000000000001,
1731
+ "loss": 1.2179,
1732
+ "step": 286
1733
+ },
1734
+ {
1735
+ "epoch": 16.25,
1736
+ "learning_rate": 0.00017662500000000002,
1737
+ "loss": 1.1218,
1738
+ "step": 287
1739
+ },
1740
+ {
1741
+ "epoch": 16.31,
1742
+ "learning_rate": 0.0001765,
1743
+ "loss": 1.2144,
1744
+ "step": 288
1745
+ },
1746
+ {
1747
+ "epoch": 16.37,
1748
+ "learning_rate": 0.000176375,
1749
+ "loss": 1.1878,
1750
+ "step": 289
1751
+ },
1752
+ {
1753
+ "epoch": 16.42,
1754
+ "learning_rate": 0.00017625,
1755
+ "loss": 1.2809,
1756
+ "step": 290
1757
+ },
1758
+ {
1759
+ "epoch": 16.48,
1760
+ "learning_rate": 0.000176125,
1761
+ "loss": 1.1862,
1762
+ "step": 291
1763
+ },
1764
+ {
1765
+ "epoch": 16.54,
1766
+ "learning_rate": 0.00017600000000000002,
1767
+ "loss": 1.2139,
1768
+ "step": 292
1769
+ },
1770
+ {
1771
+ "epoch": 16.59,
1772
+ "learning_rate": 0.000175875,
1773
+ "loss": 1.2991,
1774
+ "step": 293
1775
+ },
1776
+ {
1777
+ "epoch": 16.65,
1778
+ "learning_rate": 0.00017575000000000002,
1779
+ "loss": 1.3304,
1780
+ "step": 294
1781
+ },
1782
+ {
1783
+ "epoch": 16.71,
1784
+ "learning_rate": 0.00017562500000000003,
1785
+ "loss": 1.4349,
1786
+ "step": 295
1787
+ },
1788
+ {
1789
+ "epoch": 16.76,
1790
+ "learning_rate": 0.0001755,
1791
+ "loss": 1.2817,
1792
+ "step": 296
1793
+ },
1794
+ {
1795
+ "epoch": 16.82,
1796
+ "learning_rate": 0.000175375,
1797
+ "loss": 1.2263,
1798
+ "step": 297
1799
+ },
1800
+ {
1801
+ "epoch": 16.88,
1802
+ "learning_rate": 0.00017525,
1803
+ "loss": 1.3584,
1804
+ "step": 298
1805
+ },
1806
+ {
1807
+ "epoch": 16.93,
1808
+ "learning_rate": 0.00017512500000000001,
1809
+ "loss": 1.4542,
1810
+ "step": 299
1811
+ },
1812
+ {
1813
+ "epoch": 16.99,
1814
+ "learning_rate": 0.000175,
1815
+ "loss": 1.392,
1816
+ "step": 300
1817
+ },
1818
+ {
1819
+ "epoch": 17.05,
1820
+ "learning_rate": 0.000174875,
1821
+ "loss": 1.2685,
1822
+ "step": 301
1823
+ },
1824
+ {
1825
+ "epoch": 17.1,
1826
+ "learning_rate": 0.00017475000000000002,
1827
+ "loss": 1.2337,
1828
+ "step": 302
1829
+ },
1830
+ {
1831
+ "epoch": 17.16,
1832
+ "learning_rate": 0.00017462500000000003,
1833
+ "loss": 1.3299,
1834
+ "step": 303
1835
+ },
1836
+ {
1837
+ "epoch": 17.22,
1838
+ "learning_rate": 0.0001745,
1839
+ "loss": 1.2579,
1840
+ "step": 304
1841
+ },
1842
+ {
1843
+ "epoch": 17.27,
1844
+ "learning_rate": 0.000174375,
1845
+ "loss": 1.234,
1846
+ "step": 305
1847
+ },
1848
+ {
1849
+ "epoch": 17.33,
1850
+ "learning_rate": 0.00017425,
1851
+ "loss": 1.2273,
1852
+ "step": 306
1853
+ },
1854
+ {
1855
+ "epoch": 17.39,
1856
+ "learning_rate": 0.00017412500000000002,
1857
+ "loss": 1.3174,
1858
+ "step": 307
1859
+ },
1860
+ {
1861
+ "epoch": 17.44,
1862
+ "learning_rate": 0.000174,
1863
+ "loss": 1.1738,
1864
+ "step": 308
1865
+ },
1866
+ {
1867
+ "epoch": 17.5,
1868
+ "learning_rate": 0.000173875,
1869
+ "loss": 1.0985,
1870
+ "step": 309
1871
+ },
1872
+ {
1873
+ "epoch": 17.56,
1874
+ "learning_rate": 0.00017375000000000002,
1875
+ "loss": 1.283,
1876
+ "step": 310
1877
+ },
1878
+ {
1879
+ "epoch": 17.61,
1880
+ "learning_rate": 0.000173625,
1881
+ "loss": 1.318,
1882
+ "step": 311
1883
+ },
1884
+ {
1885
+ "epoch": 17.67,
1886
+ "learning_rate": 0.00017350000000000002,
1887
+ "loss": 1.3006,
1888
+ "step": 312
1889
+ },
1890
+ {
1891
+ "epoch": 17.73,
1892
+ "learning_rate": 0.000173375,
1893
+ "loss": 1.3629,
1894
+ "step": 313
1895
+ },
1896
+ {
1897
+ "epoch": 17.78,
1898
+ "learning_rate": 0.00017325,
1899
+ "loss": 1.3144,
1900
+ "step": 314
1901
+ },
1902
+ {
1903
+ "epoch": 17.84,
1904
+ "learning_rate": 0.000173125,
1905
+ "loss": 1.3476,
1906
+ "step": 315
1907
+ },
1908
+ {
1909
+ "epoch": 17.9,
1910
+ "learning_rate": 0.000173,
1911
+ "loss": 1.1403,
1912
+ "step": 316
1913
+ },
1914
+ {
1915
+ "epoch": 17.95,
1916
+ "learning_rate": 0.00017287500000000001,
1917
+ "loss": 1.1434,
1918
+ "step": 317
1919
+ },
1920
+ {
1921
+ "epoch": 18.01,
1922
+ "learning_rate": 0.00017275000000000002,
1923
+ "loss": 1.1653,
1924
+ "step": 318
1925
+ },
1926
+ {
1927
+ "epoch": 18.07,
1928
+ "learning_rate": 0.000172625,
1929
+ "loss": 1.1411,
1930
+ "step": 319
1931
+ },
1932
+ {
1933
+ "epoch": 18.12,
1934
+ "learning_rate": 0.00017250000000000002,
1935
+ "loss": 1.2393,
1936
+ "step": 320
1937
+ },
1938
+ {
1939
+ "epoch": 18.18,
1940
+ "learning_rate": 0.000172375,
1941
+ "loss": 1.2092,
1942
+ "step": 321
1943
+ },
1944
+ {
1945
+ "epoch": 18.24,
1946
+ "learning_rate": 0.00017225,
1947
+ "loss": 1.1437,
1948
+ "step": 322
1949
+ },
1950
+ {
1951
+ "epoch": 18.29,
1952
+ "learning_rate": 0.000172125,
1953
+ "loss": 1.1145,
1954
+ "step": 323
1955
+ },
1956
+ {
1957
+ "epoch": 18.35,
1958
+ "learning_rate": 0.000172,
1959
+ "loss": 1.2723,
1960
+ "step": 324
1961
+ },
1962
+ {
1963
+ "epoch": 18.41,
1964
+ "learning_rate": 0.00017187500000000002,
1965
+ "loss": 1.2471,
1966
+ "step": 325
1967
+ },
1968
+ {
1969
+ "epoch": 18.46,
1970
+ "learning_rate": 0.00017175,
1971
+ "loss": 1.1047,
1972
+ "step": 326
1973
+ },
1974
+ {
1975
+ "epoch": 18.52,
1976
+ "learning_rate": 0.000171625,
1977
+ "loss": 1.0908,
1978
+ "step": 327
1979
+ },
1980
+ {
1981
+ "epoch": 18.58,
1982
+ "learning_rate": 0.00017150000000000002,
1983
+ "loss": 1.3261,
1984
+ "step": 328
1985
+ },
1986
+ {
1987
+ "epoch": 18.63,
1988
+ "learning_rate": 0.00017137500000000003,
1989
+ "loss": 1.2138,
1990
+ "step": 329
1991
+ },
1992
+ {
1993
+ "epoch": 18.69,
1994
+ "learning_rate": 0.00017125,
1995
+ "loss": 1.253,
1996
+ "step": 330
1997
+ },
1998
+ {
1999
+ "epoch": 18.75,
2000
+ "learning_rate": 0.000171125,
2001
+ "loss": 1.2831,
2002
+ "step": 331
2003
+ },
2004
+ {
2005
+ "epoch": 18.8,
2006
+ "learning_rate": 0.000171,
2007
+ "loss": 1.1795,
2008
+ "step": 332
2009
+ },
2010
+ {
2011
+ "epoch": 18.86,
2012
+ "learning_rate": 0.00017087500000000002,
2013
+ "loss": 1.2454,
2014
+ "step": 333
2015
+ },
2016
+ {
2017
+ "epoch": 18.92,
2018
+ "learning_rate": 0.00017075,
2019
+ "loss": 1.2997,
2020
+ "step": 334
2021
+ },
2022
+ {
2023
+ "epoch": 18.97,
2024
+ "learning_rate": 0.00017062500000000001,
2025
+ "loss": 1.2243,
2026
+ "step": 335
2027
+ },
2028
+ {
2029
+ "epoch": 19.03,
2030
+ "learning_rate": 0.00017050000000000002,
2031
+ "loss": 1.1705,
2032
+ "step": 336
2033
+ },
2034
+ {
2035
+ "epoch": 19.09,
2036
+ "learning_rate": 0.000170375,
2037
+ "loss": 1.133,
2038
+ "step": 337
2039
+ },
2040
+ {
2041
+ "epoch": 19.14,
2042
+ "learning_rate": 0.00017025,
2043
+ "loss": 1.1997,
2044
+ "step": 338
2045
+ },
2046
+ {
2047
+ "epoch": 19.2,
2048
+ "learning_rate": 0.000170125,
2049
+ "loss": 1.1572,
2050
+ "step": 339
2051
+ },
2052
+ {
2053
+ "epoch": 19.26,
2054
+ "learning_rate": 0.00017,
2055
+ "loss": 1.2115,
2056
+ "step": 340
2057
+ },
2058
+ {
2059
+ "epoch": 19.31,
2060
+ "learning_rate": 0.000169875,
2061
+ "loss": 1.0557,
2062
+ "step": 341
2063
+ },
2064
+ {
2065
+ "epoch": 19.37,
2066
+ "learning_rate": 0.00016975,
2067
+ "loss": 1.2455,
2068
+ "step": 342
2069
+ },
2070
+ {
2071
+ "epoch": 19.43,
2072
+ "learning_rate": 0.00016962500000000002,
2073
+ "loss": 1.1917,
2074
+ "step": 343
2075
+ },
2076
+ {
2077
+ "epoch": 19.48,
2078
+ "learning_rate": 0.00016950000000000003,
2079
+ "loss": 1.2846,
2080
+ "step": 344
2081
+ },
2082
+ {
2083
+ "epoch": 19.54,
2084
+ "learning_rate": 0.000169375,
2085
+ "loss": 1.2735,
2086
+ "step": 345
2087
+ },
2088
+ {
2089
+ "epoch": 19.6,
2090
+ "learning_rate": 0.00016925,
2091
+ "loss": 1.1168,
2092
+ "step": 346
2093
+ },
2094
+ {
2095
+ "epoch": 19.65,
2096
+ "learning_rate": 0.000169125,
2097
+ "loss": 1.2314,
2098
+ "step": 347
2099
+ },
2100
+ {
2101
+ "epoch": 19.71,
2102
+ "learning_rate": 0.00016900000000000002,
2103
+ "loss": 1.1641,
2104
+ "step": 348
2105
+ },
2106
+ {
2107
+ "epoch": 19.77,
2108
+ "learning_rate": 0.000168875,
2109
+ "loss": 1.2631,
2110
+ "step": 349
2111
+ },
2112
+ {
2113
+ "epoch": 19.82,
2114
+ "learning_rate": 0.00016875,
2115
+ "loss": 1.0431,
2116
+ "step": 350
2117
+ },
2118
+ {
2119
+ "epoch": 19.88,
2120
+ "learning_rate": 0.00016862500000000002,
2121
+ "loss": 1.0931,
2122
+ "step": 351
2123
+ },
2124
+ {
2125
+ "epoch": 19.94,
2126
+ "learning_rate": 0.0001685,
2127
+ "loss": 1.2183,
2128
+ "step": 352
2129
+ },
2130
+ {
2131
+ "epoch": 19.99,
2132
+ "learning_rate": 0.000168375,
2133
+ "loss": 1.0584,
2134
+ "step": 353
2135
+ },
2136
+ {
2137
+ "epoch": 20.05,
2138
+ "learning_rate": 0.00016825000000000002,
2139
+ "loss": 1.1146,
2140
+ "step": 354
2141
+ },
2142
+ {
2143
+ "epoch": 20.11,
2144
+ "learning_rate": 0.000168125,
2145
+ "loss": 1.0702,
2146
+ "step": 355
2147
+ },
2148
+ {
2149
+ "epoch": 20.16,
2150
+ "learning_rate": 0.000168,
2151
+ "loss": 1.0612,
2152
+ "step": 356
2153
+ },
2154
+ {
2155
+ "epoch": 20.22,
2156
+ "learning_rate": 0.000167875,
2157
+ "loss": 1.2609,
2158
+ "step": 357
2159
+ },
2160
+ {
2161
+ "epoch": 20.28,
2162
+ "learning_rate": 0.00016775,
2163
+ "loss": 0.9842,
2164
+ "step": 358
2165
+ },
2166
+ {
2167
+ "epoch": 20.33,
2168
+ "learning_rate": 0.00016762500000000002,
2169
+ "loss": 1.0986,
2170
+ "step": 359
2171
+ },
2172
+ {
2173
+ "epoch": 20.39,
2174
+ "learning_rate": 0.0001675,
2175
+ "loss": 1.2615,
2176
+ "step": 360
2177
+ },
2178
+ {
2179
+ "epoch": 20.45,
2180
+ "learning_rate": 0.00016737500000000002,
2181
+ "loss": 1.187,
2182
+ "step": 361
2183
+ },
2184
+ {
2185
+ "epoch": 20.5,
2186
+ "learning_rate": 0.00016725000000000003,
2187
+ "loss": 1.1115,
2188
+ "step": 362
2189
+ },
2190
+ {
2191
+ "epoch": 20.56,
2192
+ "learning_rate": 0.000167125,
2193
+ "loss": 1.1215,
2194
+ "step": 363
2195
+ },
2196
+ {
2197
+ "epoch": 20.62,
2198
+ "learning_rate": 0.000167,
2199
+ "loss": 1.0814,
2200
+ "step": 364
2201
+ },
2202
+ {
2203
+ "epoch": 20.67,
2204
+ "learning_rate": 0.000166875,
2205
+ "loss": 1.2158,
2206
+ "step": 365
2207
+ },
2208
+ {
2209
+ "epoch": 20.73,
2210
+ "learning_rate": 0.00016675000000000001,
2211
+ "loss": 1.1993,
2212
+ "step": 366
2213
+ },
2214
+ {
2215
+ "epoch": 20.79,
2216
+ "learning_rate": 0.000166625,
2217
+ "loss": 1.2018,
2218
+ "step": 367
2219
+ },
2220
+ {
2221
+ "epoch": 20.84,
2222
+ "learning_rate": 0.0001665,
2223
+ "loss": 1.1376,
2224
+ "step": 368
2225
+ },
2226
+ {
2227
+ "epoch": 20.9,
2228
+ "learning_rate": 0.00016637500000000002,
2229
+ "loss": 1.1453,
2230
+ "step": 369
2231
+ },
2232
+ {
2233
+ "epoch": 20.96,
2234
+ "learning_rate": 0.00016625000000000003,
2235
+ "loss": 1.2199,
2236
+ "step": 370
2237
+ },
2238
+ {
2239
+ "epoch": 21.01,
2240
+ "learning_rate": 0.00016612499999999999,
2241
+ "loss": 1.2592,
2242
+ "step": 371
2243
+ },
2244
+ {
2245
+ "epoch": 21.07,
2246
+ "learning_rate": 0.000166,
2247
+ "loss": 1.1196,
2248
+ "step": 372
2249
+ },
2250
+ {
2251
+ "epoch": 21.13,
2252
+ "learning_rate": 0.000165875,
2253
+ "loss": 1.2162,
2254
+ "step": 373
2255
+ },
2256
+ {
2257
+ "epoch": 21.18,
2258
+ "learning_rate": 0.00016575000000000002,
2259
+ "loss": 1.2218,
2260
+ "step": 374
2261
+ },
2262
+ {
2263
+ "epoch": 21.24,
2264
+ "learning_rate": 0.000165625,
2265
+ "loss": 1.2069,
2266
+ "step": 375
2267
+ },
2268
+ {
2269
+ "epoch": 21.3,
2270
+ "learning_rate": 0.0001655,
2271
+ "loss": 1.0723,
2272
+ "step": 376
2273
+ },
2274
+ {
2275
+ "epoch": 21.35,
2276
+ "learning_rate": 0.00016537500000000002,
2277
+ "loss": 1.0855,
2278
+ "step": 377
2279
+ },
2280
+ {
2281
+ "epoch": 21.41,
2282
+ "learning_rate": 0.00016525,
2283
+ "loss": 1.0705,
2284
+ "step": 378
2285
+ },
2286
+ {
2287
+ "epoch": 21.47,
2288
+ "learning_rate": 0.00016512500000000002,
2289
+ "loss": 1.1985,
2290
+ "step": 379
2291
+ },
2292
+ {
2293
+ "epoch": 21.52,
2294
+ "learning_rate": 0.000165,
2295
+ "loss": 1.1375,
2296
+ "step": 380
2297
+ },
2298
+ {
2299
+ "epoch": 21.58,
2300
+ "learning_rate": 0.000164875,
2301
+ "loss": 1.0706,
2302
+ "step": 381
2303
+ },
2304
+ {
2305
+ "epoch": 21.64,
2306
+ "learning_rate": 0.00016475,
2307
+ "loss": 1.1144,
2308
+ "step": 382
2309
+ },
2310
+ {
2311
+ "epoch": 21.69,
2312
+ "learning_rate": 0.000164625,
2313
+ "loss": 1.1604,
2314
+ "step": 383
2315
+ },
2316
+ {
2317
+ "epoch": 21.75,
2318
+ "learning_rate": 0.00016450000000000001,
2319
+ "loss": 1.0285,
2320
+ "step": 384
2321
+ },
2322
+ {
2323
+ "epoch": 21.81,
2324
+ "learning_rate": 0.00016437500000000002,
2325
+ "loss": 1.2177,
2326
+ "step": 385
2327
+ },
2328
+ {
2329
+ "epoch": 21.86,
2330
+ "learning_rate": 0.00016425,
2331
+ "loss": 1.1429,
2332
+ "step": 386
2333
+ },
2334
+ {
2335
+ "epoch": 21.92,
2336
+ "learning_rate": 0.00016412500000000002,
2337
+ "loss": 1.067,
2338
+ "step": 387
2339
+ },
2340
+ {
2341
+ "epoch": 21.98,
2342
+ "learning_rate": 0.000164,
2343
+ "loss": 1.1697,
2344
+ "step": 388
2345
+ },
2346
+ {
2347
+ "epoch": 22.03,
2348
+ "learning_rate": 0.000163875,
2349
+ "loss": 1.0991,
2350
+ "step": 389
2351
+ },
2352
+ {
2353
+ "epoch": 22.09,
2354
+ "learning_rate": 0.00016375,
2355
+ "loss": 1.0132,
2356
+ "step": 390
2357
+ },
2358
+ {
2359
+ "epoch": 22.15,
2360
+ "learning_rate": 0.000163625,
2361
+ "loss": 1.0528,
2362
+ "step": 391
2363
+ },
2364
+ {
2365
+ "epoch": 22.2,
2366
+ "learning_rate": 0.00016350000000000002,
2367
+ "loss": 1.1334,
2368
+ "step": 392
2369
+ },
2370
+ {
2371
+ "epoch": 22.26,
2372
+ "learning_rate": 0.000163375,
2373
+ "loss": 1.0935,
2374
+ "step": 393
2375
+ },
2376
+ {
2377
+ "epoch": 22.32,
2378
+ "learning_rate": 0.00016325,
2379
+ "loss": 0.9794,
2380
+ "step": 394
2381
+ },
2382
+ {
2383
+ "epoch": 22.37,
2384
+ "learning_rate": 0.00016312500000000002,
2385
+ "loss": 1.1197,
2386
+ "step": 395
2387
+ },
2388
+ {
2389
+ "epoch": 22.43,
2390
+ "learning_rate": 0.000163,
2391
+ "loss": 1.0824,
2392
+ "step": 396
2393
+ },
2394
+ {
2395
+ "epoch": 22.48,
2396
+ "learning_rate": 0.000162875,
2397
+ "loss": 1.1128,
2398
+ "step": 397
2399
+ },
2400
+ {
2401
+ "epoch": 22.54,
2402
+ "learning_rate": 0.00016275,
2403
+ "loss": 1.1092,
2404
+ "step": 398
2405
+ },
2406
+ {
2407
+ "epoch": 22.6,
2408
+ "learning_rate": 0.000162625,
2409
+ "loss": 1.1914,
2410
+ "step": 399
2411
+ },
2412
+ {
2413
+ "epoch": 22.65,
2414
+ "learning_rate": 0.00016250000000000002,
2415
+ "loss": 1.1871,
2416
+ "step": 400
2417
+ },
2418
+ {
2419
+ "epoch": 22.65,
2420
+ "eval_loss": 2.3080484867095947,
2421
+ "eval_runtime": 67.9996,
2422
+ "eval_samples_per_second": 3.691,
2423
+ "eval_steps_per_second": 0.926,
2424
+ "step": 400
2425
+ },
2426
+ {
2427
+ "epoch": 22.71,
2428
+ "learning_rate": 0.000162375,
2429
+ "loss": 1.1742,
2430
+ "step": 401
2431
+ },
2432
+ {
2433
+ "epoch": 22.77,
2434
+ "learning_rate": 0.00016225000000000001,
2435
+ "loss": 1.3617,
2436
+ "step": 402
2437
+ },
2438
+ {
2439
+ "epoch": 22.82,
2440
+ "learning_rate": 0.00016212500000000002,
2441
+ "loss": 1.0186,
2442
+ "step": 403
2443
+ },
2444
+ {
2445
+ "epoch": 22.88,
2446
+ "learning_rate": 0.000162,
2447
+ "loss": 1.1819,
2448
+ "step": 404
2449
+ },
2450
+ {
2451
+ "epoch": 22.94,
2452
+ "learning_rate": 0.000161875,
2453
+ "loss": 1.0858,
2454
+ "step": 405
2455
+ },
2456
+ {
2457
+ "epoch": 22.99,
2458
+ "learning_rate": 0.00016175,
2459
+ "loss": 1.084,
2460
+ "step": 406
2461
+ },
2462
+ {
2463
+ "epoch": 23.05,
2464
+ "learning_rate": 0.000161625,
2465
+ "loss": 1.0896,
2466
+ "step": 407
2467
+ },
2468
+ {
2469
+ "epoch": 23.11,
2470
+ "learning_rate": 0.0001615,
2471
+ "loss": 0.9239,
2472
+ "step": 408
2473
+ },
2474
+ {
2475
+ "epoch": 23.16,
2476
+ "learning_rate": 0.000161375,
2477
+ "loss": 1.16,
2478
+ "step": 409
2479
+ },
2480
+ {
2481
+ "epoch": 23.22,
2482
+ "learning_rate": 0.00016125000000000002,
2483
+ "loss": 1.0874,
2484
+ "step": 410
2485
+ },
2486
+ {
2487
+ "epoch": 23.28,
2488
+ "learning_rate": 0.00016112500000000003,
2489
+ "loss": 1.0956,
2490
+ "step": 411
2491
+ },
2492
+ {
2493
+ "epoch": 23.33,
2494
+ "learning_rate": 0.000161,
2495
+ "loss": 1.1366,
2496
+ "step": 412
2497
+ },
2498
+ {
2499
+ "epoch": 23.39,
2500
+ "learning_rate": 0.000160875,
2501
+ "loss": 1.1507,
2502
+ "step": 413
2503
+ },
2504
+ {
2505
+ "epoch": 23.45,
2506
+ "learning_rate": 0.00016075,
2507
+ "loss": 1.0451,
2508
+ "step": 414
2509
+ },
2510
+ {
2511
+ "epoch": 23.5,
2512
+ "learning_rate": 0.00016062500000000001,
2513
+ "loss": 1.1593,
2514
+ "step": 415
2515
+ },
2516
+ {
2517
+ "epoch": 23.56,
2518
+ "learning_rate": 0.0001605,
2519
+ "loss": 1.0692,
2520
+ "step": 416
2521
+ },
2522
+ {
2523
+ "epoch": 23.62,
2524
+ "learning_rate": 0.000160375,
2525
+ "loss": 1.0252,
2526
+ "step": 417
2527
+ },
2528
+ {
2529
+ "epoch": 23.67,
2530
+ "learning_rate": 0.00016025000000000002,
2531
+ "loss": 1.0877,
2532
+ "step": 418
2533
+ },
2534
+ {
2535
+ "epoch": 23.73,
2536
+ "learning_rate": 0.000160125,
2537
+ "loss": 1.2211,
2538
+ "step": 419
2539
+ },
2540
+ {
2541
+ "epoch": 23.79,
2542
+ "learning_rate": 0.00016,
2543
+ "loss": 1.1834,
2544
+ "step": 420
2545
+ },
2546
+ {
2547
+ "epoch": 23.84,
2548
+ "learning_rate": 0.000159875,
2549
+ "loss": 1.0965,
2550
+ "step": 421
2551
+ },
2552
+ {
2553
+ "epoch": 23.9,
2554
+ "learning_rate": 0.00015975,
2555
+ "loss": 1.0839,
2556
+ "step": 422
2557
+ },
2558
+ {
2559
+ "epoch": 23.96,
2560
+ "learning_rate": 0.000159625,
2561
+ "loss": 1.0622,
2562
+ "step": 423
2563
+ },
2564
+ {
2565
+ "epoch": 24.01,
2566
+ "learning_rate": 0.0001595,
2567
+ "loss": 1.1252,
2568
+ "step": 424
2569
+ },
2570
+ {
2571
+ "epoch": 24.07,
2572
+ "learning_rate": 0.000159375,
2573
+ "loss": 1.1057,
2574
+ "step": 425
2575
+ },
2576
+ {
2577
+ "epoch": 24.13,
2578
+ "learning_rate": 0.00015925000000000002,
2579
+ "loss": 1.0499,
2580
+ "step": 426
2581
+ },
2582
+ {
2583
+ "epoch": 24.18,
2584
+ "learning_rate": 0.000159125,
2585
+ "loss": 1.2239,
2586
+ "step": 427
2587
+ },
2588
+ {
2589
+ "epoch": 24.24,
2590
+ "learning_rate": 0.00015900000000000002,
2591
+ "loss": 1.0456,
2592
+ "step": 428
2593
+ },
2594
+ {
2595
+ "epoch": 24.3,
2596
+ "learning_rate": 0.00015887500000000003,
2597
+ "loss": 1.1966,
2598
+ "step": 429
2599
+ },
2600
+ {
2601
+ "epoch": 24.35,
2602
+ "learning_rate": 0.00015875,
2603
+ "loss": 1.0116,
2604
+ "step": 430
2605
+ },
2606
+ {
2607
+ "epoch": 24.41,
2608
+ "learning_rate": 0.000158625,
2609
+ "loss": 1.0544,
2610
+ "step": 431
2611
+ },
2612
+ {
2613
+ "epoch": 24.47,
2614
+ "learning_rate": 0.0001585,
2615
+ "loss": 0.9815,
2616
+ "step": 432
2617
+ },
2618
+ {
2619
+ "epoch": 24.52,
2620
+ "learning_rate": 0.00015837500000000001,
2621
+ "loss": 1.0492,
2622
+ "step": 433
2623
+ },
2624
+ {
2625
+ "epoch": 24.58,
2626
+ "learning_rate": 0.00015825,
2627
+ "loss": 1.067,
2628
+ "step": 434
2629
+ },
2630
+ {
2631
+ "epoch": 24.64,
2632
+ "learning_rate": 0.000158125,
2633
+ "loss": 1.0854,
2634
+ "step": 435
2635
+ },
2636
+ {
2637
+ "epoch": 24.69,
2638
+ "learning_rate": 0.00015800000000000002,
2639
+ "loss": 1.1032,
2640
+ "step": 436
2641
+ },
2642
+ {
2643
+ "epoch": 24.75,
2644
+ "learning_rate": 0.00015787500000000003,
2645
+ "loss": 1.0008,
2646
+ "step": 437
2647
+ },
2648
+ {
2649
+ "epoch": 24.81,
2650
+ "learning_rate": 0.00015774999999999999,
2651
+ "loss": 1.0489,
2652
+ "step": 438
2653
+ },
2654
+ {
2655
+ "epoch": 24.86,
2656
+ "learning_rate": 0.000157625,
2657
+ "loss": 1.0176,
2658
+ "step": 439
2659
+ },
2660
+ {
2661
+ "epoch": 24.92,
2662
+ "learning_rate": 0.0001575,
2663
+ "loss": 0.9086,
2664
+ "step": 440
2665
+ },
2666
+ {
2667
+ "epoch": 24.98,
2668
+ "learning_rate": 0.00015737500000000002,
2669
+ "loss": 1.105,
2670
+ "step": 441
2671
+ },
2672
+ {
2673
+ "epoch": 25.03,
2674
+ "learning_rate": 0.00015725,
2675
+ "loss": 1.163,
2676
+ "step": 442
2677
+ },
2678
+ {
2679
+ "epoch": 25.09,
2680
+ "learning_rate": 0.000157125,
2681
+ "loss": 1.0763,
2682
+ "step": 443
2683
+ },
2684
+ {
2685
+ "epoch": 25.15,
2686
+ "learning_rate": 0.00015700000000000002,
2687
+ "loss": 0.9489,
2688
+ "step": 444
2689
+ },
2690
+ {
2691
+ "epoch": 25.2,
2692
+ "learning_rate": 0.000156875,
2693
+ "loss": 1.0961,
2694
+ "step": 445
2695
+ },
2696
+ {
2697
+ "epoch": 25.26,
2698
+ "learning_rate": 0.00015675,
2699
+ "loss": 1.1935,
2700
+ "step": 446
2701
+ },
2702
+ {
2703
+ "epoch": 25.32,
2704
+ "learning_rate": 0.000156625,
2705
+ "loss": 1.0047,
2706
+ "step": 447
2707
+ },
2708
+ {
2709
+ "epoch": 25.37,
2710
+ "learning_rate": 0.0001565,
2711
+ "loss": 1.0807,
2712
+ "step": 448
2713
+ },
2714
+ {
2715
+ "epoch": 25.43,
2716
+ "learning_rate": 0.000156375,
2717
+ "loss": 1.0766,
2718
+ "step": 449
2719
+ },
2720
+ {
2721
+ "epoch": 25.49,
2722
+ "learning_rate": 0.00015625,
2723
+ "loss": 0.9833,
2724
+ "step": 450
2725
+ },
2726
+ {
2727
+ "epoch": 25.54,
2728
+ "learning_rate": 0.00015612500000000001,
2729
+ "loss": 1.032,
2730
+ "step": 451
2731
+ },
2732
+ {
2733
+ "epoch": 25.6,
2734
+ "learning_rate": 0.00015600000000000002,
2735
+ "loss": 1.0264,
2736
+ "step": 452
2737
+ },
2738
+ {
2739
+ "epoch": 25.66,
2740
+ "learning_rate": 0.000155875,
2741
+ "loss": 1.1064,
2742
+ "step": 453
2743
+ },
2744
+ {
2745
+ "epoch": 25.71,
2746
+ "learning_rate": 0.00015575000000000002,
2747
+ "loss": 0.9957,
2748
+ "step": 454
2749
+ },
2750
+ {
2751
+ "epoch": 25.77,
2752
+ "learning_rate": 0.000155625,
2753
+ "loss": 1.1988,
2754
+ "step": 455
2755
+ },
2756
+ {
2757
+ "epoch": 25.83,
2758
+ "learning_rate": 0.0001555,
2759
+ "loss": 1.0918,
2760
+ "step": 456
2761
+ },
2762
+ {
2763
+ "epoch": 25.88,
2764
+ "learning_rate": 0.000155375,
2765
+ "loss": 1.0202,
2766
+ "step": 457
2767
+ },
2768
+ {
2769
+ "epoch": 25.94,
2770
+ "learning_rate": 0.00015525,
2771
+ "loss": 1.0904,
2772
+ "step": 458
2773
+ },
2774
+ {
2775
+ "epoch": 26.0,
2776
+ "learning_rate": 0.00015512500000000002,
2777
+ "loss": 1.0169,
2778
+ "step": 459
2779
+ },
2780
+ {
2781
+ "epoch": 26.05,
2782
+ "learning_rate": 0.000155,
2783
+ "loss": 1.0173,
2784
+ "step": 460
2785
+ },
2786
+ {
2787
+ "epoch": 26.11,
2788
+ "learning_rate": 0.000154875,
2789
+ "loss": 1.1198,
2790
+ "step": 461
2791
+ },
2792
+ {
2793
+ "epoch": 26.17,
2794
+ "learning_rate": 0.00015475000000000002,
2795
+ "loss": 1.1392,
2796
+ "step": 462
2797
+ },
2798
+ {
2799
+ "epoch": 26.22,
2800
+ "learning_rate": 0.000154625,
2801
+ "loss": 1.014,
2802
+ "step": 463
2803
+ },
2804
+ {
2805
+ "epoch": 26.28,
2806
+ "learning_rate": 0.0001545,
2807
+ "loss": 0.9519,
2808
+ "step": 464
2809
+ },
2810
+ {
2811
+ "epoch": 26.34,
2812
+ "learning_rate": 0.000154375,
2813
+ "loss": 0.9909,
2814
+ "step": 465
2815
+ },
2816
+ {
2817
+ "epoch": 26.39,
2818
+ "learning_rate": 0.00015425,
2819
+ "loss": 0.9521,
2820
+ "step": 466
2821
+ },
2822
+ {
2823
+ "epoch": 26.45,
2824
+ "learning_rate": 0.00015412500000000002,
2825
+ "loss": 1.1342,
2826
+ "step": 467
2827
+ },
2828
+ {
2829
+ "epoch": 26.51,
2830
+ "learning_rate": 0.000154,
2831
+ "loss": 1.0784,
2832
+ "step": 468
2833
+ },
2834
+ {
2835
+ "epoch": 26.56,
2836
+ "learning_rate": 0.000153875,
2837
+ "loss": 1.0328,
2838
+ "step": 469
2839
+ },
2840
+ {
2841
+ "epoch": 26.62,
2842
+ "learning_rate": 0.00015375000000000002,
2843
+ "loss": 1.0087,
2844
+ "step": 470
2845
+ },
2846
+ {
2847
+ "epoch": 26.68,
2848
+ "learning_rate": 0.000153625,
2849
+ "loss": 1.0398,
2850
+ "step": 471
2851
+ },
2852
+ {
2853
+ "epoch": 26.73,
2854
+ "learning_rate": 0.0001535,
2855
+ "loss": 0.9858,
2856
+ "step": 472
2857
+ },
2858
+ {
2859
+ "epoch": 26.79,
2860
+ "learning_rate": 0.000153375,
2861
+ "loss": 1.2076,
2862
+ "step": 473
2863
+ },
2864
+ {
2865
+ "epoch": 26.85,
2866
+ "learning_rate": 0.00015325,
2867
+ "loss": 1.1085,
2868
+ "step": 474
2869
+ },
2870
+ {
2871
+ "epoch": 26.9,
2872
+ "learning_rate": 0.000153125,
2873
+ "loss": 1.1035,
2874
+ "step": 475
2875
+ },
2876
+ {
2877
+ "epoch": 26.96,
2878
+ "learning_rate": 0.000153,
2879
+ "loss": 0.9399,
2880
+ "step": 476
2881
+ },
2882
+ {
2883
+ "epoch": 27.02,
2884
+ "learning_rate": 0.00015287500000000002,
2885
+ "loss": 1.0872,
2886
+ "step": 477
2887
+ },
2888
+ {
2889
+ "epoch": 27.07,
2890
+ "learning_rate": 0.00015275000000000003,
2891
+ "loss": 1.0331,
2892
+ "step": 478
2893
+ },
2894
+ {
2895
+ "epoch": 27.13,
2896
+ "learning_rate": 0.000152625,
2897
+ "loss": 0.9109,
2898
+ "step": 479
2899
+ },
2900
+ {
2901
+ "epoch": 27.19,
2902
+ "learning_rate": 0.0001525,
2903
+ "loss": 0.9697,
2904
+ "step": 480
2905
+ },
2906
+ {
2907
+ "epoch": 27.24,
2908
+ "learning_rate": 0.000152375,
2909
+ "loss": 1.0549,
2910
+ "step": 481
2911
+ },
2912
+ {
2913
+ "epoch": 27.3,
2914
+ "learning_rate": 0.00015225000000000001,
2915
+ "loss": 0.998,
2916
+ "step": 482
2917
+ },
2918
+ {
2919
+ "epoch": 27.36,
2920
+ "learning_rate": 0.000152125,
2921
+ "loss": 1.0075,
2922
+ "step": 483
2923
+ },
2924
+ {
2925
+ "epoch": 27.41,
2926
+ "learning_rate": 0.000152,
2927
+ "loss": 1.0581,
2928
+ "step": 484
2929
+ },
2930
+ {
2931
+ "epoch": 27.47,
2932
+ "learning_rate": 0.00015187500000000002,
2933
+ "loss": 1.0142,
2934
+ "step": 485
2935
+ },
2936
+ {
2937
+ "epoch": 27.53,
2938
+ "learning_rate": 0.00015175,
2939
+ "loss": 0.9666,
2940
+ "step": 486
2941
+ },
2942
+ {
2943
+ "epoch": 27.58,
2944
+ "learning_rate": 0.000151625,
2945
+ "loss": 1.0657,
2946
+ "step": 487
2947
+ },
2948
+ {
2949
+ "epoch": 27.64,
2950
+ "learning_rate": 0.0001515,
2951
+ "loss": 1.0518,
2952
+ "step": 488
2953
+ },
2954
+ {
2955
+ "epoch": 27.7,
2956
+ "learning_rate": 0.000151375,
2957
+ "loss": 1.0108,
2958
+ "step": 489
2959
+ },
2960
+ {
2961
+ "epoch": 27.75,
2962
+ "learning_rate": 0.00015125,
2963
+ "loss": 1.0609,
2964
+ "step": 490
2965
+ },
2966
+ {
2967
+ "epoch": 27.81,
2968
+ "learning_rate": 0.000151125,
2969
+ "loss": 0.9527,
2970
+ "step": 491
2971
+ },
2972
+ {
2973
+ "epoch": 27.87,
2974
+ "learning_rate": 0.000151,
2975
+ "loss": 1.0935,
2976
+ "step": 492
2977
+ },
2978
+ {
2979
+ "epoch": 27.92,
2980
+ "learning_rate": 0.00015087500000000002,
2981
+ "loss": 1.0846,
2982
+ "step": 493
2983
+ },
2984
+ {
2985
+ "epoch": 27.98,
2986
+ "learning_rate": 0.00015075,
2987
+ "loss": 1.155,
2988
+ "step": 494
2989
+ },
2990
+ {
2991
+ "epoch": 28.04,
2992
+ "learning_rate": 0.00015062500000000002,
2993
+ "loss": 1.0016,
2994
+ "step": 495
2995
+ },
2996
+ {
2997
+ "epoch": 28.09,
2998
+ "learning_rate": 0.0001505,
2999
+ "loss": 0.9649,
3000
+ "step": 496
3001
+ },
3002
+ {
3003
+ "epoch": 28.15,
3004
+ "learning_rate": 0.000150375,
3005
+ "loss": 0.8216,
3006
+ "step": 497
3007
+ },
3008
+ {
3009
+ "epoch": 28.21,
3010
+ "learning_rate": 0.00015025,
3011
+ "loss": 0.8122,
3012
+ "step": 498
3013
+ },
3014
+ {
3015
+ "epoch": 28.26,
3016
+ "learning_rate": 0.000150125,
3017
+ "loss": 0.9406,
3018
+ "step": 499
3019
+ },
3020
+ {
3021
+ "epoch": 28.32,
3022
+ "learning_rate": 0.00015000000000000001,
3023
+ "loss": 0.9773,
3024
+ "step": 500
3025
+ },
3026
+ {
3027
+ "epoch": 28.38,
3028
+ "learning_rate": 0.000149875,
3029
+ "loss": 1.0644,
3030
+ "step": 501
3031
+ },
3032
+ {
3033
+ "epoch": 28.43,
3034
+ "learning_rate": 0.00014975,
3035
+ "loss": 0.94,
3036
+ "step": 502
3037
+ },
3038
+ {
3039
+ "epoch": 28.49,
3040
+ "learning_rate": 0.00014962500000000002,
3041
+ "loss": 0.9985,
3042
+ "step": 503
3043
+ },
3044
+ {
3045
+ "epoch": 28.55,
3046
+ "learning_rate": 0.00014950000000000003,
3047
+ "loss": 0.9291,
3048
+ "step": 504
3049
+ },
3050
+ {
3051
+ "epoch": 28.6,
3052
+ "learning_rate": 0.00014937499999999999,
3053
+ "loss": 1.105,
3054
+ "step": 505
3055
+ },
3056
+ {
3057
+ "epoch": 28.66,
3058
+ "learning_rate": 0.00014925,
3059
+ "loss": 1.0298,
3060
+ "step": 506
3061
+ },
3062
+ {
3063
+ "epoch": 28.72,
3064
+ "learning_rate": 0.000149125,
3065
+ "loss": 0.8352,
3066
+ "step": 507
3067
+ },
3068
+ {
3069
+ "epoch": 28.77,
3070
+ "learning_rate": 0.00014900000000000002,
3071
+ "loss": 0.9752,
3072
+ "step": 508
3073
+ },
3074
+ {
3075
+ "epoch": 28.83,
3076
+ "learning_rate": 0.000148875,
3077
+ "loss": 1.0227,
3078
+ "step": 509
3079
+ },
3080
+ {
3081
+ "epoch": 28.88,
3082
+ "learning_rate": 0.00014875,
3083
+ "loss": 1.0159,
3084
+ "step": 510
3085
+ },
3086
+ {
3087
+ "epoch": 28.94,
3088
+ "learning_rate": 0.00014862500000000002,
3089
+ "loss": 0.9519,
3090
+ "step": 511
3091
+ },
3092
+ {
3093
+ "epoch": 29.0,
3094
+ "learning_rate": 0.0001485,
3095
+ "loss": 1.0509,
3096
+ "step": 512
3097
+ },
3098
+ {
3099
+ "epoch": 29.05,
3100
+ "learning_rate": 0.000148375,
3101
+ "loss": 0.9518,
3102
+ "step": 513
3103
+ },
3104
+ {
3105
+ "epoch": 29.11,
3106
+ "learning_rate": 0.00014825,
3107
+ "loss": 0.916,
3108
+ "step": 514
3109
+ },
3110
+ {
3111
+ "epoch": 29.17,
3112
+ "learning_rate": 0.000148125,
3113
+ "loss": 0.9861,
3114
+ "step": 515
3115
+ },
3116
+ {
3117
+ "epoch": 29.22,
3118
+ "learning_rate": 0.000148,
3119
+ "loss": 0.8708,
3120
+ "step": 516
3121
+ },
3122
+ {
3123
+ "epoch": 29.28,
3124
+ "learning_rate": 0.000147875,
3125
+ "loss": 0.9761,
3126
+ "step": 517
3127
+ },
3128
+ {
3129
+ "epoch": 29.34,
3130
+ "learning_rate": 0.00014775,
3131
+ "loss": 0.9873,
3132
+ "step": 518
3133
+ },
3134
+ {
3135
+ "epoch": 29.39,
3136
+ "learning_rate": 0.00014762500000000002,
3137
+ "loss": 0.9831,
3138
+ "step": 519
3139
+ },
3140
+ {
3141
+ "epoch": 29.45,
3142
+ "learning_rate": 0.0001475,
3143
+ "loss": 0.9699,
3144
+ "step": 520
3145
+ },
3146
+ {
3147
+ "epoch": 29.51,
3148
+ "learning_rate": 0.000147375,
3149
+ "loss": 1.0476,
3150
+ "step": 521
3151
+ },
3152
+ {
3153
+ "epoch": 29.56,
3154
+ "learning_rate": 0.00014725,
3155
+ "loss": 0.9194,
3156
+ "step": 522
3157
+ },
3158
+ {
3159
+ "epoch": 29.62,
3160
+ "learning_rate": 0.000147125,
3161
+ "loss": 0.8511,
3162
+ "step": 523
3163
+ },
3164
+ {
3165
+ "epoch": 29.68,
3166
+ "learning_rate": 0.000147,
3167
+ "loss": 0.9279,
3168
+ "step": 524
3169
+ },
3170
+ {
3171
+ "epoch": 29.73,
3172
+ "learning_rate": 0.000146875,
3173
+ "loss": 0.9263,
3174
+ "step": 525
3175
+ },
3176
+ {
3177
+ "epoch": 29.79,
3178
+ "learning_rate": 0.00014675000000000002,
3179
+ "loss": 1.0193,
3180
+ "step": 526
3181
+ },
3182
+ {
3183
+ "epoch": 29.85,
3184
+ "learning_rate": 0.000146625,
3185
+ "loss": 0.855,
3186
+ "step": 527
3187
+ },
3188
+ {
3189
+ "epoch": 29.9,
3190
+ "learning_rate": 0.0001465,
3191
+ "loss": 1.0262,
3192
+ "step": 528
3193
+ },
3194
+ {
3195
+ "epoch": 29.96,
3196
+ "learning_rate": 0.00014637500000000002,
3197
+ "loss": 0.9289,
3198
+ "step": 529
3199
+ },
3200
+ {
3201
+ "epoch": 30.02,
3202
+ "learning_rate": 0.00014625,
3203
+ "loss": 0.9726,
3204
+ "step": 530
3205
+ },
3206
+ {
3207
+ "epoch": 30.07,
3208
+ "learning_rate": 0.000146125,
3209
+ "loss": 0.9372,
3210
+ "step": 531
3211
+ },
3212
+ {
3213
+ "epoch": 30.13,
3214
+ "learning_rate": 0.000146,
3215
+ "loss": 0.9264,
3216
+ "step": 532
3217
+ },
3218
+ {
3219
+ "epoch": 30.19,
3220
+ "learning_rate": 0.000145875,
3221
+ "loss": 1.082,
3222
+ "step": 533
3223
+ },
3224
+ {
3225
+ "epoch": 30.24,
3226
+ "learning_rate": 0.00014575000000000002,
3227
+ "loss": 0.8726,
3228
+ "step": 534
3229
+ },
3230
+ {
3231
+ "epoch": 30.3,
3232
+ "learning_rate": 0.000145625,
3233
+ "loss": 0.946,
3234
+ "step": 535
3235
+ },
3236
+ {
3237
+ "epoch": 30.36,
3238
+ "learning_rate": 0.0001455,
3239
+ "loss": 0.9745,
3240
+ "step": 536
3241
+ },
3242
+ {
3243
+ "epoch": 30.41,
3244
+ "learning_rate": 0.00014537500000000002,
3245
+ "loss": 0.9029,
3246
+ "step": 537
3247
+ },
3248
+ {
3249
+ "epoch": 30.47,
3250
+ "learning_rate": 0.00014525,
3251
+ "loss": 0.9419,
3252
+ "step": 538
3253
+ },
3254
+ {
3255
+ "epoch": 30.53,
3256
+ "learning_rate": 0.000145125,
3257
+ "loss": 0.9483,
3258
+ "step": 539
3259
+ },
3260
+ {
3261
+ "epoch": 30.58,
3262
+ "learning_rate": 0.000145,
3263
+ "loss": 0.931,
3264
+ "step": 540
3265
+ },
3266
+ {
3267
+ "epoch": 30.64,
3268
+ "learning_rate": 0.000144875,
3269
+ "loss": 0.8506,
3270
+ "step": 541
3271
+ },
3272
+ {
3273
+ "epoch": 30.7,
3274
+ "learning_rate": 0.00014475,
3275
+ "loss": 0.9794,
3276
+ "step": 542
3277
+ },
3278
+ {
3279
+ "epoch": 30.75,
3280
+ "learning_rate": 0.000144625,
3281
+ "loss": 0.9518,
3282
+ "step": 543
3283
+ },
3284
+ {
3285
+ "epoch": 30.81,
3286
+ "learning_rate": 0.00014450000000000002,
3287
+ "loss": 1.0354,
3288
+ "step": 544
3289
+ },
3290
+ {
3291
+ "epoch": 30.87,
3292
+ "learning_rate": 0.00014437500000000003,
3293
+ "loss": 0.942,
3294
+ "step": 545
3295
+ },
3296
+ {
3297
+ "epoch": 30.92,
3298
+ "learning_rate": 0.00014425,
3299
+ "loss": 1.0193,
3300
+ "step": 546
3301
+ },
3302
+ {
3303
+ "epoch": 30.98,
3304
+ "learning_rate": 0.000144125,
3305
+ "loss": 1.0498,
3306
+ "step": 547
3307
+ },
3308
+ {
3309
+ "epoch": 31.04,
3310
+ "learning_rate": 0.000144,
3311
+ "loss": 0.8734,
3312
+ "step": 548
3313
+ },
3314
+ {
3315
+ "epoch": 31.09,
3316
+ "learning_rate": 0.00014387500000000001,
3317
+ "loss": 0.8525,
3318
+ "step": 549
3319
+ },
3320
+ {
3321
+ "epoch": 31.15,
3322
+ "learning_rate": 0.00014375,
3323
+ "loss": 0.8724,
3324
+ "step": 550
3325
+ },
3326
+ {
3327
+ "epoch": 31.21,
3328
+ "learning_rate": 0.000143625,
3329
+ "loss": 0.8856,
3330
+ "step": 551
3331
+ },
3332
+ {
3333
+ "epoch": 31.26,
3334
+ "learning_rate": 0.00014350000000000002,
3335
+ "loss": 0.9098,
3336
+ "step": 552
3337
+ },
3338
+ {
3339
+ "epoch": 31.32,
3340
+ "learning_rate": 0.000143375,
3341
+ "loss": 0.8638,
3342
+ "step": 553
3343
+ },
3344
+ {
3345
+ "epoch": 31.38,
3346
+ "learning_rate": 0.00014325,
3347
+ "loss": 0.869,
3348
+ "step": 554
3349
+ },
3350
+ {
3351
+ "epoch": 31.43,
3352
+ "learning_rate": 0.000143125,
3353
+ "loss": 0.9197,
3354
+ "step": 555
3355
+ },
3356
+ {
3357
+ "epoch": 31.49,
3358
+ "learning_rate": 0.000143,
3359
+ "loss": 0.9562,
3360
+ "step": 556
3361
+ },
3362
+ {
3363
+ "epoch": 31.55,
3364
+ "learning_rate": 0.000142875,
3365
+ "loss": 0.9259,
3366
+ "step": 557
3367
+ },
3368
+ {
3369
+ "epoch": 31.6,
3370
+ "learning_rate": 0.00014275,
3371
+ "loss": 0.9913,
3372
+ "step": 558
3373
+ },
3374
+ {
3375
+ "epoch": 31.66,
3376
+ "learning_rate": 0.000142625,
3377
+ "loss": 0.8817,
3378
+ "step": 559
3379
+ },
3380
+ {
3381
+ "epoch": 31.72,
3382
+ "learning_rate": 0.00014250000000000002,
3383
+ "loss": 0.9828,
3384
+ "step": 560
3385
+ },
3386
+ {
3387
+ "epoch": 31.77,
3388
+ "learning_rate": 0.000142375,
3389
+ "loss": 1.019,
3390
+ "step": 561
3391
+ },
3392
+ {
3393
+ "epoch": 31.83,
3394
+ "learning_rate": 0.00014225000000000002,
3395
+ "loss": 1.0316,
3396
+ "step": 562
3397
+ },
3398
+ {
3399
+ "epoch": 31.89,
3400
+ "learning_rate": 0.000142125,
3401
+ "loss": 0.8991,
3402
+ "step": 563
3403
+ },
3404
+ {
3405
+ "epoch": 31.94,
3406
+ "learning_rate": 0.000142,
3407
+ "loss": 0.9081,
3408
+ "step": 564
3409
+ },
3410
+ {
3411
+ "epoch": 32.0,
3412
+ "learning_rate": 0.000141875,
3413
+ "loss": 0.9126,
3414
+ "step": 565
3415
+ },
3416
+ {
3417
+ "epoch": 32.06,
3418
+ "learning_rate": 0.00014175,
3419
+ "loss": 0.8007,
3420
+ "step": 566
3421
+ },
3422
+ {
3423
+ "epoch": 32.11,
3424
+ "learning_rate": 0.00014162500000000001,
3425
+ "loss": 0.9289,
3426
+ "step": 567
3427
+ },
3428
+ {
3429
+ "epoch": 32.17,
3430
+ "learning_rate": 0.0001415,
3431
+ "loss": 1.001,
3432
+ "step": 568
3433
+ },
3434
+ {
3435
+ "epoch": 32.23,
3436
+ "learning_rate": 0.000141375,
3437
+ "loss": 0.896,
3438
+ "step": 569
3439
+ },
3440
+ {
3441
+ "epoch": 32.28,
3442
+ "learning_rate": 0.00014125000000000002,
3443
+ "loss": 0.9181,
3444
+ "step": 570
3445
+ },
3446
+ {
3447
+ "epoch": 32.34,
3448
+ "learning_rate": 0.000141125,
3449
+ "loss": 0.9645,
3450
+ "step": 571
3451
+ },
3452
+ {
3453
+ "epoch": 32.4,
3454
+ "learning_rate": 0.000141,
3455
+ "loss": 0.8713,
3456
+ "step": 572
3457
+ },
3458
+ {
3459
+ "epoch": 32.45,
3460
+ "learning_rate": 0.000140875,
3461
+ "loss": 0.8495,
3462
+ "step": 573
3463
+ },
3464
+ {
3465
+ "epoch": 32.51,
3466
+ "learning_rate": 0.00014075,
3467
+ "loss": 0.8962,
3468
+ "step": 574
3469
+ },
3470
+ {
3471
+ "epoch": 32.57,
3472
+ "learning_rate": 0.00014062500000000002,
3473
+ "loss": 0.8542,
3474
+ "step": 575
3475
+ },
3476
+ {
3477
+ "epoch": 32.62,
3478
+ "learning_rate": 0.0001405,
3479
+ "loss": 0.8794,
3480
+ "step": 576
3481
+ },
3482
+ {
3483
+ "epoch": 32.68,
3484
+ "learning_rate": 0.000140375,
3485
+ "loss": 0.9115,
3486
+ "step": 577
3487
+ },
3488
+ {
3489
+ "epoch": 32.74,
3490
+ "learning_rate": 0.00014025000000000002,
3491
+ "loss": 0.815,
3492
+ "step": 578
3493
+ },
3494
+ {
3495
+ "epoch": 32.79,
3496
+ "learning_rate": 0.000140125,
3497
+ "loss": 0.8095,
3498
+ "step": 579
3499
+ },
3500
+ {
3501
+ "epoch": 32.85,
3502
+ "learning_rate": 0.00014,
3503
+ "loss": 0.8478,
3504
+ "step": 580
3505
+ },
3506
+ {
3507
+ "epoch": 32.91,
3508
+ "learning_rate": 0.000139875,
3509
+ "loss": 0.8056,
3510
+ "step": 581
3511
+ },
3512
+ {
3513
+ "epoch": 32.96,
3514
+ "learning_rate": 0.00013975,
3515
+ "loss": 0.9301,
3516
+ "step": 582
3517
+ },
3518
+ {
3519
+ "epoch": 33.02,
3520
+ "learning_rate": 0.00013962500000000002,
3521
+ "loss": 0.8053,
3522
+ "step": 583
3523
+ },
3524
+ {
3525
+ "epoch": 33.08,
3526
+ "learning_rate": 0.0001395,
3527
+ "loss": 0.8094,
3528
+ "step": 584
3529
+ },
3530
+ {
3531
+ "epoch": 33.13,
3532
+ "learning_rate": 0.000139375,
3533
+ "loss": 0.8191,
3534
+ "step": 585
3535
+ },
3536
+ {
3537
+ "epoch": 33.19,
3538
+ "learning_rate": 0.00013925000000000002,
3539
+ "loss": 0.8934,
3540
+ "step": 586
3541
+ },
3542
+ {
3543
+ "epoch": 33.25,
3544
+ "learning_rate": 0.000139125,
3545
+ "loss": 0.88,
3546
+ "step": 587
3547
+ },
3548
+ {
3549
+ "epoch": 33.3,
3550
+ "learning_rate": 0.000139,
3551
+ "loss": 0.9598,
3552
+ "step": 588
3553
+ },
3554
+ {
3555
+ "epoch": 33.36,
3556
+ "learning_rate": 0.000138875,
3557
+ "loss": 0.8177,
3558
+ "step": 589
3559
+ },
3560
+ {
3561
+ "epoch": 33.42,
3562
+ "learning_rate": 0.00013875,
3563
+ "loss": 0.7988,
3564
+ "step": 590
3565
+ },
3566
+ {
3567
+ "epoch": 33.47,
3568
+ "learning_rate": 0.000138625,
3569
+ "loss": 0.976,
3570
+ "step": 591
3571
+ },
3572
+ {
3573
+ "epoch": 33.53,
3574
+ "learning_rate": 0.0001385,
3575
+ "loss": 0.7895,
3576
+ "step": 592
3577
+ },
3578
+ {
3579
+ "epoch": 33.59,
3580
+ "learning_rate": 0.00013837500000000002,
3581
+ "loss": 0.8736,
3582
+ "step": 593
3583
+ },
3584
+ {
3585
+ "epoch": 33.64,
3586
+ "learning_rate": 0.00013825,
3587
+ "loss": 0.821,
3588
+ "step": 594
3589
+ },
3590
+ {
3591
+ "epoch": 33.7,
3592
+ "learning_rate": 0.000138125,
3593
+ "loss": 0.8448,
3594
+ "step": 595
3595
+ },
3596
+ {
3597
+ "epoch": 33.76,
3598
+ "learning_rate": 0.000138,
3599
+ "loss": 0.8763,
3600
+ "step": 596
3601
+ },
3602
+ {
3603
+ "epoch": 33.81,
3604
+ "learning_rate": 0.000137875,
3605
+ "loss": 0.9005,
3606
+ "step": 597
3607
+ },
3608
+ {
3609
+ "epoch": 33.87,
3610
+ "learning_rate": 0.00013775000000000001,
3611
+ "loss": 0.7962,
3612
+ "step": 598
3613
+ },
3614
+ {
3615
+ "epoch": 33.93,
3616
+ "learning_rate": 0.000137625,
3617
+ "loss": 0.9145,
3618
+ "step": 599
3619
+ },
3620
+ {
3621
+ "epoch": 33.98,
3622
+ "learning_rate": 0.0001375,
3623
+ "loss": 0.8612,
3624
+ "step": 600
3625
+ },
3626
+ {
3627
+ "epoch": 33.98,
3628
+ "eval_loss": 2.6225430965423584,
3629
+ "eval_runtime": 90.4931,
3630
+ "eval_samples_per_second": 2.774,
3631
+ "eval_steps_per_second": 0.696,
3632
+ "step": 600
3633
+ }
3634
+ ],
3635
+ "max_steps": 1700,
3636
+ "num_train_epochs": 100,
3637
+ "total_flos": 4.163924540462285e+17,
3638
+ "trial_name": null,
3639
+ "trial_params": null
3640
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:185ee5a8381bbeaa47b1b1b01858891652bda537aa6fa9c983bb9b86683b0d83
3
+ size 3899
vocab.json ADDED
The diff for this file is too large to render. See raw diff