bradmin commited on
Commit
6f1526a
1 Parent(s): 27aae48

Training in progress, step 700, checkpoint

Browse files
checkpoint-700/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "EleutherAI/polyglot-ko-1.3b",
3
+ "architectures": [
4
+ "GPTNeoXForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": 0.1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout": 0.0,
12
+ "hidden_size": 2048,
13
+ "id2label": {
14
+ "0": "LABEL_0"
15
+ },
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 8192,
18
+ "label2id": {
19
+ "LABEL_0": 0
20
+ },
21
+ "layer_norm_eps": 1e-05,
22
+ "max_position_embeddings": 2048,
23
+ "model_type": "gpt_neox",
24
+ "num_attention_heads": 16,
25
+ "num_hidden_layers": 24,
26
+ "pad_token_id": 2,
27
+ "rope_scaling": null,
28
+ "rotary_emb_base": 10000,
29
+ "rotary_pct": 0.5,
30
+ "tie_word_embeddings": false,
31
+ "torch_dtype": "float32",
32
+ "transformers_version": "4.35.0",
33
+ "use_cache": true,
34
+ "use_parallel_residual": true,
35
+ "vocab_size": 30080
36
+ }
checkpoint-700/model-00001-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56090e5c9cd0c1fd8b7e986deb435cf20df8f75cda13c13a145e74fdc0fc0a11
3
+ size 4946584368
checkpoint-700/model-00002-of-00002.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d4666516b5b9bc8f3e2db53d337d5cd315351987dbfe8bc52c66fa808a1c117
3
+ size 134284024
checkpoint-700/model.safetensors.index.json ADDED
@@ -0,0 +1,299 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 5080834048
4
+ },
5
+ "weight_map": {
6
+ "gpt_neox.embed_in.weight": "model-00001-of-00002.safetensors",
7
+ "gpt_neox.final_layer_norm.bias": "model-00002-of-00002.safetensors",
8
+ "gpt_neox.final_layer_norm.weight": "model-00002-of-00002.safetensors",
9
+ "gpt_neox.layers.0.attention.dense.bias": "model-00001-of-00002.safetensors",
10
+ "gpt_neox.layers.0.attention.dense.weight": "model-00001-of-00002.safetensors",
11
+ "gpt_neox.layers.0.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
12
+ "gpt_neox.layers.0.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
13
+ "gpt_neox.layers.0.input_layernorm.bias": "model-00001-of-00002.safetensors",
14
+ "gpt_neox.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
15
+ "gpt_neox.layers.0.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
16
+ "gpt_neox.layers.0.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
17
+ "gpt_neox.layers.0.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
18
+ "gpt_neox.layers.0.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
19
+ "gpt_neox.layers.0.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
20
+ "gpt_neox.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
21
+ "gpt_neox.layers.1.attention.dense.bias": "model-00001-of-00002.safetensors",
22
+ "gpt_neox.layers.1.attention.dense.weight": "model-00001-of-00002.safetensors",
23
+ "gpt_neox.layers.1.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
24
+ "gpt_neox.layers.1.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
25
+ "gpt_neox.layers.1.input_layernorm.bias": "model-00001-of-00002.safetensors",
26
+ "gpt_neox.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
27
+ "gpt_neox.layers.1.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
28
+ "gpt_neox.layers.1.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
29
+ "gpt_neox.layers.1.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
30
+ "gpt_neox.layers.1.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
31
+ "gpt_neox.layers.1.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
32
+ "gpt_neox.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
33
+ "gpt_neox.layers.10.attention.dense.bias": "model-00001-of-00002.safetensors",
34
+ "gpt_neox.layers.10.attention.dense.weight": "model-00001-of-00002.safetensors",
35
+ "gpt_neox.layers.10.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
36
+ "gpt_neox.layers.10.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
37
+ "gpt_neox.layers.10.input_layernorm.bias": "model-00001-of-00002.safetensors",
38
+ "gpt_neox.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
39
+ "gpt_neox.layers.10.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
40
+ "gpt_neox.layers.10.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
41
+ "gpt_neox.layers.10.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
42
+ "gpt_neox.layers.10.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
43
+ "gpt_neox.layers.10.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
44
+ "gpt_neox.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
45
+ "gpt_neox.layers.11.attention.dense.bias": "model-00001-of-00002.safetensors",
46
+ "gpt_neox.layers.11.attention.dense.weight": "model-00001-of-00002.safetensors",
47
+ "gpt_neox.layers.11.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
48
+ "gpt_neox.layers.11.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
49
+ "gpt_neox.layers.11.input_layernorm.bias": "model-00001-of-00002.safetensors",
50
+ "gpt_neox.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
51
+ "gpt_neox.layers.11.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
52
+ "gpt_neox.layers.11.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
53
+ "gpt_neox.layers.11.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
54
+ "gpt_neox.layers.11.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
55
+ "gpt_neox.layers.11.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
56
+ "gpt_neox.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
57
+ "gpt_neox.layers.12.attention.dense.bias": "model-00001-of-00002.safetensors",
58
+ "gpt_neox.layers.12.attention.dense.weight": "model-00001-of-00002.safetensors",
59
+ "gpt_neox.layers.12.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
60
+ "gpt_neox.layers.12.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
61
+ "gpt_neox.layers.12.input_layernorm.bias": "model-00001-of-00002.safetensors",
62
+ "gpt_neox.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
63
+ "gpt_neox.layers.12.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
64
+ "gpt_neox.layers.12.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
65
+ "gpt_neox.layers.12.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
66
+ "gpt_neox.layers.12.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
67
+ "gpt_neox.layers.12.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
68
+ "gpt_neox.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
69
+ "gpt_neox.layers.13.attention.dense.bias": "model-00001-of-00002.safetensors",
70
+ "gpt_neox.layers.13.attention.dense.weight": "model-00001-of-00002.safetensors",
71
+ "gpt_neox.layers.13.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
72
+ "gpt_neox.layers.13.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
73
+ "gpt_neox.layers.13.input_layernorm.bias": "model-00001-of-00002.safetensors",
74
+ "gpt_neox.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
75
+ "gpt_neox.layers.13.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
76
+ "gpt_neox.layers.13.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
77
+ "gpt_neox.layers.13.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
78
+ "gpt_neox.layers.13.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
79
+ "gpt_neox.layers.13.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
80
+ "gpt_neox.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
81
+ "gpt_neox.layers.14.attention.dense.bias": "model-00001-of-00002.safetensors",
82
+ "gpt_neox.layers.14.attention.dense.weight": "model-00001-of-00002.safetensors",
83
+ "gpt_neox.layers.14.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
84
+ "gpt_neox.layers.14.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
85
+ "gpt_neox.layers.14.input_layernorm.bias": "model-00001-of-00002.safetensors",
86
+ "gpt_neox.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
87
+ "gpt_neox.layers.14.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
88
+ "gpt_neox.layers.14.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
89
+ "gpt_neox.layers.14.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
90
+ "gpt_neox.layers.14.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
91
+ "gpt_neox.layers.14.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
92
+ "gpt_neox.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
93
+ "gpt_neox.layers.15.attention.dense.bias": "model-00001-of-00002.safetensors",
94
+ "gpt_neox.layers.15.attention.dense.weight": "model-00001-of-00002.safetensors",
95
+ "gpt_neox.layers.15.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
96
+ "gpt_neox.layers.15.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
97
+ "gpt_neox.layers.15.input_layernorm.bias": "model-00001-of-00002.safetensors",
98
+ "gpt_neox.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
99
+ "gpt_neox.layers.15.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
100
+ "gpt_neox.layers.15.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
101
+ "gpt_neox.layers.15.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
102
+ "gpt_neox.layers.15.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
103
+ "gpt_neox.layers.15.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
104
+ "gpt_neox.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
105
+ "gpt_neox.layers.16.attention.dense.bias": "model-00001-of-00002.safetensors",
106
+ "gpt_neox.layers.16.attention.dense.weight": "model-00001-of-00002.safetensors",
107
+ "gpt_neox.layers.16.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
108
+ "gpt_neox.layers.16.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
109
+ "gpt_neox.layers.16.input_layernorm.bias": "model-00001-of-00002.safetensors",
110
+ "gpt_neox.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
111
+ "gpt_neox.layers.16.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
112
+ "gpt_neox.layers.16.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
113
+ "gpt_neox.layers.16.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
114
+ "gpt_neox.layers.16.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
115
+ "gpt_neox.layers.16.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
116
+ "gpt_neox.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
117
+ "gpt_neox.layers.17.attention.dense.bias": "model-00001-of-00002.safetensors",
118
+ "gpt_neox.layers.17.attention.dense.weight": "model-00001-of-00002.safetensors",
119
+ "gpt_neox.layers.17.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
120
+ "gpt_neox.layers.17.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
121
+ "gpt_neox.layers.17.input_layernorm.bias": "model-00001-of-00002.safetensors",
122
+ "gpt_neox.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
123
+ "gpt_neox.layers.17.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
124
+ "gpt_neox.layers.17.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
125
+ "gpt_neox.layers.17.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
126
+ "gpt_neox.layers.17.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
127
+ "gpt_neox.layers.17.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
128
+ "gpt_neox.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
129
+ "gpt_neox.layers.18.attention.dense.bias": "model-00001-of-00002.safetensors",
130
+ "gpt_neox.layers.18.attention.dense.weight": "model-00001-of-00002.safetensors",
131
+ "gpt_neox.layers.18.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
132
+ "gpt_neox.layers.18.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
133
+ "gpt_neox.layers.18.input_layernorm.bias": "model-00001-of-00002.safetensors",
134
+ "gpt_neox.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
135
+ "gpt_neox.layers.18.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
136
+ "gpt_neox.layers.18.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
137
+ "gpt_neox.layers.18.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
138
+ "gpt_neox.layers.18.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
139
+ "gpt_neox.layers.18.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
140
+ "gpt_neox.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
141
+ "gpt_neox.layers.19.attention.dense.bias": "model-00001-of-00002.safetensors",
142
+ "gpt_neox.layers.19.attention.dense.weight": "model-00001-of-00002.safetensors",
143
+ "gpt_neox.layers.19.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
144
+ "gpt_neox.layers.19.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
145
+ "gpt_neox.layers.19.input_layernorm.bias": "model-00001-of-00002.safetensors",
146
+ "gpt_neox.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
147
+ "gpt_neox.layers.19.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
148
+ "gpt_neox.layers.19.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
149
+ "gpt_neox.layers.19.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
150
+ "gpt_neox.layers.19.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
151
+ "gpt_neox.layers.19.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
152
+ "gpt_neox.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
153
+ "gpt_neox.layers.2.attention.dense.bias": "model-00001-of-00002.safetensors",
154
+ "gpt_neox.layers.2.attention.dense.weight": "model-00001-of-00002.safetensors",
155
+ "gpt_neox.layers.2.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
156
+ "gpt_neox.layers.2.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
157
+ "gpt_neox.layers.2.input_layernorm.bias": "model-00001-of-00002.safetensors",
158
+ "gpt_neox.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
159
+ "gpt_neox.layers.2.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
160
+ "gpt_neox.layers.2.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
161
+ "gpt_neox.layers.2.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
162
+ "gpt_neox.layers.2.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
163
+ "gpt_neox.layers.2.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
164
+ "gpt_neox.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
165
+ "gpt_neox.layers.20.attention.dense.bias": "model-00001-of-00002.safetensors",
166
+ "gpt_neox.layers.20.attention.dense.weight": "model-00001-of-00002.safetensors",
167
+ "gpt_neox.layers.20.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
168
+ "gpt_neox.layers.20.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
169
+ "gpt_neox.layers.20.input_layernorm.bias": "model-00001-of-00002.safetensors",
170
+ "gpt_neox.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
171
+ "gpt_neox.layers.20.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
172
+ "gpt_neox.layers.20.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
173
+ "gpt_neox.layers.20.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
174
+ "gpt_neox.layers.20.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
175
+ "gpt_neox.layers.20.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
176
+ "gpt_neox.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
177
+ "gpt_neox.layers.21.attention.dense.bias": "model-00001-of-00002.safetensors",
178
+ "gpt_neox.layers.21.attention.dense.weight": "model-00001-of-00002.safetensors",
179
+ "gpt_neox.layers.21.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
180
+ "gpt_neox.layers.21.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
181
+ "gpt_neox.layers.21.input_layernorm.bias": "model-00001-of-00002.safetensors",
182
+ "gpt_neox.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
183
+ "gpt_neox.layers.21.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
184
+ "gpt_neox.layers.21.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
185
+ "gpt_neox.layers.21.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
186
+ "gpt_neox.layers.21.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
187
+ "gpt_neox.layers.21.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
188
+ "gpt_neox.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
189
+ "gpt_neox.layers.22.attention.dense.bias": "model-00001-of-00002.safetensors",
190
+ "gpt_neox.layers.22.attention.dense.weight": "model-00001-of-00002.safetensors",
191
+ "gpt_neox.layers.22.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
192
+ "gpt_neox.layers.22.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
193
+ "gpt_neox.layers.22.input_layernorm.bias": "model-00001-of-00002.safetensors",
194
+ "gpt_neox.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
195
+ "gpt_neox.layers.22.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
196
+ "gpt_neox.layers.22.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
197
+ "gpt_neox.layers.22.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
198
+ "gpt_neox.layers.22.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
199
+ "gpt_neox.layers.22.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
200
+ "gpt_neox.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
201
+ "gpt_neox.layers.23.attention.dense.bias": "model-00001-of-00002.safetensors",
202
+ "gpt_neox.layers.23.attention.dense.weight": "model-00001-of-00002.safetensors",
203
+ "gpt_neox.layers.23.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
204
+ "gpt_neox.layers.23.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
205
+ "gpt_neox.layers.23.input_layernorm.bias": "model-00001-of-00002.safetensors",
206
+ "gpt_neox.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
207
+ "gpt_neox.layers.23.mlp.dense_4h_to_h.bias": "model-00002-of-00002.safetensors",
208
+ "gpt_neox.layers.23.mlp.dense_4h_to_h.weight": "model-00002-of-00002.safetensors",
209
+ "gpt_neox.layers.23.mlp.dense_h_to_4h.bias": "model-00002-of-00002.safetensors",
210
+ "gpt_neox.layers.23.mlp.dense_h_to_4h.weight": "model-00002-of-00002.safetensors",
211
+ "gpt_neox.layers.23.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
212
+ "gpt_neox.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
213
+ "gpt_neox.layers.3.attention.dense.bias": "model-00001-of-00002.safetensors",
214
+ "gpt_neox.layers.3.attention.dense.weight": "model-00001-of-00002.safetensors",
215
+ "gpt_neox.layers.3.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
216
+ "gpt_neox.layers.3.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
217
+ "gpt_neox.layers.3.input_layernorm.bias": "model-00001-of-00002.safetensors",
218
+ "gpt_neox.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
219
+ "gpt_neox.layers.3.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
220
+ "gpt_neox.layers.3.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
221
+ "gpt_neox.layers.3.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
222
+ "gpt_neox.layers.3.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
223
+ "gpt_neox.layers.3.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
224
+ "gpt_neox.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
225
+ "gpt_neox.layers.4.attention.dense.bias": "model-00001-of-00002.safetensors",
226
+ "gpt_neox.layers.4.attention.dense.weight": "model-00001-of-00002.safetensors",
227
+ "gpt_neox.layers.4.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
228
+ "gpt_neox.layers.4.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
229
+ "gpt_neox.layers.4.input_layernorm.bias": "model-00001-of-00002.safetensors",
230
+ "gpt_neox.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
231
+ "gpt_neox.layers.4.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
232
+ "gpt_neox.layers.4.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
233
+ "gpt_neox.layers.4.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
234
+ "gpt_neox.layers.4.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
235
+ "gpt_neox.layers.4.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
236
+ "gpt_neox.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
237
+ "gpt_neox.layers.5.attention.dense.bias": "model-00001-of-00002.safetensors",
238
+ "gpt_neox.layers.5.attention.dense.weight": "model-00001-of-00002.safetensors",
239
+ "gpt_neox.layers.5.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
240
+ "gpt_neox.layers.5.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
241
+ "gpt_neox.layers.5.input_layernorm.bias": "model-00001-of-00002.safetensors",
242
+ "gpt_neox.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
243
+ "gpt_neox.layers.5.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
244
+ "gpt_neox.layers.5.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
245
+ "gpt_neox.layers.5.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
246
+ "gpt_neox.layers.5.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
247
+ "gpt_neox.layers.5.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
248
+ "gpt_neox.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
249
+ "gpt_neox.layers.6.attention.dense.bias": "model-00001-of-00002.safetensors",
250
+ "gpt_neox.layers.6.attention.dense.weight": "model-00001-of-00002.safetensors",
251
+ "gpt_neox.layers.6.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
252
+ "gpt_neox.layers.6.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
253
+ "gpt_neox.layers.6.input_layernorm.bias": "model-00001-of-00002.safetensors",
254
+ "gpt_neox.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
255
+ "gpt_neox.layers.6.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
256
+ "gpt_neox.layers.6.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
257
+ "gpt_neox.layers.6.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
258
+ "gpt_neox.layers.6.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
259
+ "gpt_neox.layers.6.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
260
+ "gpt_neox.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
261
+ "gpt_neox.layers.7.attention.dense.bias": "model-00001-of-00002.safetensors",
262
+ "gpt_neox.layers.7.attention.dense.weight": "model-00001-of-00002.safetensors",
263
+ "gpt_neox.layers.7.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
264
+ "gpt_neox.layers.7.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
265
+ "gpt_neox.layers.7.input_layernorm.bias": "model-00001-of-00002.safetensors",
266
+ "gpt_neox.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
267
+ "gpt_neox.layers.7.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
268
+ "gpt_neox.layers.7.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
269
+ "gpt_neox.layers.7.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
270
+ "gpt_neox.layers.7.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
271
+ "gpt_neox.layers.7.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
272
+ "gpt_neox.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
273
+ "gpt_neox.layers.8.attention.dense.bias": "model-00001-of-00002.safetensors",
274
+ "gpt_neox.layers.8.attention.dense.weight": "model-00001-of-00002.safetensors",
275
+ "gpt_neox.layers.8.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
276
+ "gpt_neox.layers.8.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
277
+ "gpt_neox.layers.8.input_layernorm.bias": "model-00001-of-00002.safetensors",
278
+ "gpt_neox.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
279
+ "gpt_neox.layers.8.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
280
+ "gpt_neox.layers.8.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
281
+ "gpt_neox.layers.8.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
282
+ "gpt_neox.layers.8.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
283
+ "gpt_neox.layers.8.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
284
+ "gpt_neox.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
285
+ "gpt_neox.layers.9.attention.dense.bias": "model-00001-of-00002.safetensors",
286
+ "gpt_neox.layers.9.attention.dense.weight": "model-00001-of-00002.safetensors",
287
+ "gpt_neox.layers.9.attention.query_key_value.bias": "model-00001-of-00002.safetensors",
288
+ "gpt_neox.layers.9.attention.query_key_value.weight": "model-00001-of-00002.safetensors",
289
+ "gpt_neox.layers.9.input_layernorm.bias": "model-00001-of-00002.safetensors",
290
+ "gpt_neox.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
291
+ "gpt_neox.layers.9.mlp.dense_4h_to_h.bias": "model-00001-of-00002.safetensors",
292
+ "gpt_neox.layers.9.mlp.dense_4h_to_h.weight": "model-00001-of-00002.safetensors",
293
+ "gpt_neox.layers.9.mlp.dense_h_to_4h.bias": "model-00001-of-00002.safetensors",
294
+ "gpt_neox.layers.9.mlp.dense_h_to_4h.weight": "model-00001-of-00002.safetensors",
295
+ "gpt_neox.layers.9.post_attention_layernorm.bias": "model-00001-of-00002.safetensors",
296
+ "gpt_neox.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
297
+ "score.weight": "model-00002-of-00002.safetensors"
298
+ }
299
+ }
checkpoint-700/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:611d6f64e8ffca433242517a8a840aca0a80b4ad8c2194e9166868af9fdcb59a
3
+ size 10161845830
checkpoint-700/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df01d99e83c90666c61d1ba4877079bcde444031932b1cd7ab18003d475e35d5
3
+ size 14308
checkpoint-700/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a690ddf619e2bedc0f644439875bb3745f4010acfa6aab3153f7f6936fb55ec5
3
+ size 1064
checkpoint-700/special_tokens_map.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|endoftext|>",
4
+ "<|sep|>",
5
+ "<|acc|>",
6
+ "<|tel|>",
7
+ "<|rrn|>"
8
+ ],
9
+ "eos_token": "<|endoftext|>",
10
+ "pad_token": "<|endoftext|>"
11
+ }
checkpoint-700/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-700/tokenizer_config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<|unused0|>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<|unused1|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<|endoftext|>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<|sep|>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "30000": {
36
+ "content": "<|acc|>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "30001": {
44
+ "content": "<|tel|>",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "30002": {
52
+ "content": "<|rrn|>",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ }
59
+ },
60
+ "additional_special_tokens": [
61
+ "<|endoftext|>",
62
+ "<|sep|>",
63
+ "<|acc|>",
64
+ "<|tel|>",
65
+ "<|rrn|>"
66
+ ],
67
+ "clean_up_tokenization_spaces": true,
68
+ "eos_token": "<|endoftext|>",
69
+ "model_max_length": 1000000000000000019884624838656,
70
+ "pad_token": "<|endoftext|>",
71
+ "tokenizer_class": "PreTrainedTokenizerFast"
72
+ }
checkpoint-700/trainer_state.json ADDED
@@ -0,0 +1,502 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.8800880088008801,
5
+ "eval_steps": 100,
6
+ "global_step": 700,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 8.996486892091073e-06,
14
+ "loss": 0.5074,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.03,
19
+ "learning_rate": 8.985953053665263e-06,
20
+ "loss": 0.3259,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.04,
25
+ "learning_rate": 8.968414932060827e-06,
26
+ "loss": 0.2229,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.05,
31
+ "learning_rate": 8.94389991097275e-06,
32
+ "loss": 0.2081,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.06,
37
+ "learning_rate": 8.912446267696357e-06,
38
+ "loss": 0.0982,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.08,
43
+ "learning_rate": 8.874103113361848e-06,
44
+ "loss": 0.1889,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.09,
49
+ "learning_rate": 8.828930316253112e-06,
50
+ "loss": 0.1204,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.1,
55
+ "learning_rate": 8.7769984083305e-06,
56
+ "loss": 0.1826,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.11,
61
+ "learning_rate": 8.718388475103545e-06,
62
+ "loss": 0.1188,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.13,
67
+ "learning_rate": 8.653192029025561e-06,
68
+ "loss": 0.1975,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.13,
73
+ "eval_accuracy": 0.0,
74
+ "eval_loss": 0.047521572560071945,
75
+ "eval_runtime": 42.0503,
76
+ "eval_samples_per_second": 67.253,
77
+ "eval_steps_per_second": 16.813,
78
+ "step": 100
79
+ },
80
+ {
81
+ "epoch": 0.14,
82
+ "learning_rate": 8.581510866607817e-06,
83
+ "loss": 0.1713,
84
+ "step": 110
85
+ },
86
+ {
87
+ "epoch": 0.15,
88
+ "learning_rate": 8.503456909476364e-06,
89
+ "loss": 0.1711,
90
+ "step": 120
91
+ },
92
+ {
93
+ "epoch": 0.16,
94
+ "learning_rate": 8.419152029619696e-06,
95
+ "loss": 0.1162,
96
+ "step": 130
97
+ },
98
+ {
99
+ "epoch": 0.18,
100
+ "learning_rate": 8.32872785910012e-06,
101
+ "loss": 0.1478,
102
+ "step": 140
103
+ },
104
+ {
105
+ "epoch": 0.19,
106
+ "learning_rate": 8.232325584525908e-06,
107
+ "loss": 0.0978,
108
+ "step": 150
109
+ },
110
+ {
111
+ "epoch": 0.2,
112
+ "learning_rate": 8.130095726605169e-06,
113
+ "loss": 0.1324,
114
+ "step": 160
115
+ },
116
+ {
117
+ "epoch": 0.21,
118
+ "learning_rate": 8.022197905125629e-06,
119
+ "loss": 0.186,
120
+ "step": 170
121
+ },
122
+ {
123
+ "epoch": 0.23,
124
+ "learning_rate": 7.90880058972729e-06,
125
+ "loss": 0.1056,
126
+ "step": 180
127
+ },
128
+ {
129
+ "epoch": 0.24,
130
+ "learning_rate": 7.790080836857073e-06,
131
+ "loss": 0.1531,
132
+ "step": 190
133
+ },
134
+ {
135
+ "epoch": 0.25,
136
+ "learning_rate": 7.666224013316203e-06,
137
+ "loss": 0.0945,
138
+ "step": 200
139
+ },
140
+ {
141
+ "epoch": 0.25,
142
+ "eval_accuracy": 0.0,
143
+ "eval_loss": 0.03287160396575928,
144
+ "eval_runtime": 44.0534,
145
+ "eval_samples_per_second": 64.195,
146
+ "eval_steps_per_second": 16.049,
147
+ "step": 200
148
+ },
149
+ {
150
+ "epoch": 0.26,
151
+ "learning_rate": 7.537423506831948e-06,
152
+ "loss": 0.123,
153
+ "step": 210
154
+ },
155
+ {
156
+ "epoch": 0.28,
157
+ "learning_rate": 7.403880424105644e-06,
158
+ "loss": 0.1193,
159
+ "step": 220
160
+ },
161
+ {
162
+ "epoch": 0.29,
163
+ "learning_rate": 7.2658032768084485e-06,
164
+ "loss": 0.1133,
165
+ "step": 230
166
+ },
167
+ {
168
+ "epoch": 0.3,
169
+ "learning_rate": 7.123407656015124e-06,
170
+ "loss": 0.1132,
171
+ "step": 240
172
+ },
173
+ {
174
+ "epoch": 0.31,
175
+ "learning_rate": 6.976915895584163e-06,
176
+ "loss": 0.0595,
177
+ "step": 250
178
+ },
179
+ {
180
+ "epoch": 0.33,
181
+ "learning_rate": 6.826556725009857e-06,
182
+ "loss": 0.0863,
183
+ "step": 260
184
+ },
185
+ {
186
+ "epoch": 0.34,
187
+ "learning_rate": 6.672564912288353e-06,
188
+ "loss": 0.1418,
189
+ "step": 270
190
+ },
191
+ {
192
+ "epoch": 0.35,
193
+ "learning_rate": 6.515180897355285e-06,
194
+ "loss": 0.0834,
195
+ "step": 280
196
+ },
197
+ {
198
+ "epoch": 0.36,
199
+ "learning_rate": 6.354650416667365e-06,
200
+ "loss": 0.0952,
201
+ "step": 290
202
+ },
203
+ {
204
+ "epoch": 0.38,
205
+ "learning_rate": 6.191224119514071e-06,
206
+ "loss": 0.0739,
207
+ "step": 300
208
+ },
209
+ {
210
+ "epoch": 0.38,
211
+ "eval_accuracy": 0.0,
212
+ "eval_loss": 0.015648044645786285,
213
+ "eval_runtime": 43.1054,
214
+ "eval_samples_per_second": 65.607,
215
+ "eval_steps_per_second": 16.402,
216
+ "step": 300
217
+ },
218
+ {
219
+ "epoch": 0.39,
220
+ "learning_rate": 6.0251571766585436e-06,
221
+ "loss": 0.1315,
222
+ "step": 310
223
+ },
224
+ {
225
+ "epoch": 0.4,
226
+ "learning_rate": 5.856708881918714e-06,
227
+ "loss": 0.0622,
228
+ "step": 320
229
+ },
230
+ {
231
+ "epoch": 0.41,
232
+ "learning_rate": 5.6861422473108045e-06,
233
+ "loss": 0.1584,
234
+ "step": 330
235
+ },
236
+ {
237
+ "epoch": 0.43,
238
+ "learning_rate": 5.513723592387277e-06,
239
+ "loss": 0.0692,
240
+ "step": 340
241
+ },
242
+ {
243
+ "epoch": 0.44,
244
+ "learning_rate": 5.339722128410471e-06,
245
+ "loss": 0.1092,
246
+ "step": 350
247
+ },
248
+ {
249
+ "epoch": 0.45,
250
+ "learning_rate": 5.164409538011166e-06,
251
+ "loss": 0.1042,
252
+ "step": 360
253
+ },
254
+ {
255
+ "epoch": 0.47,
256
+ "learning_rate": 4.988059550988415e-06,
257
+ "loss": 0.0705,
258
+ "step": 370
259
+ },
260
+ {
261
+ "epoch": 0.48,
262
+ "learning_rate": 4.810947516912951e-06,
263
+ "loss": 0.103,
264
+ "step": 380
265
+ },
266
+ {
267
+ "epoch": 0.49,
268
+ "learning_rate": 4.633349975201519e-06,
269
+ "loss": 0.0836,
270
+ "step": 390
271
+ },
272
+ {
273
+ "epoch": 0.5,
274
+ "learning_rate": 4.455544223333405e-06,
275
+ "loss": 0.1354,
276
+ "step": 400
277
+ },
278
+ {
279
+ "epoch": 0.5,
280
+ "eval_accuracy": 0.5,
281
+ "eval_loss": 0.022996455430984497,
282
+ "eval_runtime": 42.1328,
283
+ "eval_samples_per_second": 67.121,
284
+ "eval_steps_per_second": 16.78,
285
+ "step": 400
286
+ },
287
+ {
288
+ "epoch": 0.52,
289
+ "learning_rate": 4.277807883883337e-06,
290
+ "loss": 0.0535,
291
+ "step": 410
292
+ },
293
+ {
294
+ "epoch": 0.53,
295
+ "learning_rate": 4.100418471046794e-06,
296
+ "loss": 0.0905,
297
+ "step": 420
298
+ },
299
+ {
300
+ "epoch": 0.54,
301
+ "learning_rate": 3.9236529573345295e-06,
302
+ "loss": 0.0816,
303
+ "step": 430
304
+ },
305
+ {
306
+ "epoch": 0.55,
307
+ "learning_rate": 3.7477873411128757e-06,
308
+ "loss": 0.0506,
309
+ "step": 440
310
+ },
311
+ {
312
+ "epoch": 0.57,
313
+ "learning_rate": 3.573096215665058e-06,
314
+ "loss": 0.0645,
315
+ "step": 450
316
+ },
317
+ {
318
+ "epoch": 0.58,
319
+ "learning_rate": 3.399852340446378e-06,
320
+ "loss": 0.1117,
321
+ "step": 460
322
+ },
323
+ {
324
+ "epoch": 0.59,
325
+ "learning_rate": 3.2283262152027056e-06,
326
+ "loss": 0.0811,
327
+ "step": 470
328
+ },
329
+ {
330
+ "epoch": 0.6,
331
+ "learning_rate": 3.0587856576172314e-06,
332
+ "loss": 0.0565,
333
+ "step": 480
334
+ },
335
+ {
336
+ "epoch": 0.62,
337
+ "learning_rate": 2.89149538514495e-06,
338
+ "loss": 0.0982,
339
+ "step": 490
340
+ },
341
+ {
342
+ "epoch": 0.63,
343
+ "learning_rate": 2.726716601687775e-06,
344
+ "loss": 0.1114,
345
+ "step": 500
346
+ },
347
+ {
348
+ "epoch": 0.63,
349
+ "eval_accuracy": 0.5,
350
+ "eval_loss": 0.0259599220007658,
351
+ "eval_runtime": 44.1798,
352
+ "eval_samples_per_second": 64.011,
353
+ "eval_steps_per_second": 16.003,
354
+ "step": 500
355
+ },
356
+ {
357
+ "epoch": 0.64,
358
+ "learning_rate": 2.5647065897556595e-06,
359
+ "loss": 0.0602,
360
+ "step": 510
361
+ },
362
+ {
363
+ "epoch": 0.65,
364
+ "learning_rate": 2.405718308750485e-06,
365
+ "loss": 0.0819,
366
+ "step": 520
367
+ },
368
+ {
369
+ "epoch": 0.67,
370
+ "learning_rate": 2.250000000000001e-06,
371
+ "loss": 0.0783,
372
+ "step": 530
373
+ },
374
+ {
375
+ "epoch": 0.68,
376
+ "learning_rate": 2.097794799158444e-06,
377
+ "loss": 0.0366,
378
+ "step": 540
379
+ },
380
+ {
381
+ "epoch": 0.69,
382
+ "learning_rate": 1.9493403565790804e-06,
383
+ "loss": 0.0948,
384
+ "step": 550
385
+ },
386
+ {
387
+ "epoch": 0.7,
388
+ "learning_rate": 1.8048684662514e-06,
389
+ "loss": 0.1191,
390
+ "step": 560
391
+ },
392
+ {
393
+ "epoch": 0.72,
394
+ "learning_rate": 1.6646047038823006e-06,
395
+ "loss": 0.0714,
396
+ "step": 570
397
+ },
398
+ {
399
+ "epoch": 0.73,
400
+ "learning_rate": 1.5287680746864134e-06,
401
+ "loss": 0.1644,
402
+ "step": 580
403
+ },
404
+ {
405
+ "epoch": 0.74,
406
+ "learning_rate": 1.3975706714354478e-06,
407
+ "loss": 0.0743,
408
+ "step": 590
409
+ },
410
+ {
411
+ "epoch": 0.75,
412
+ "learning_rate": 1.2712173433005088e-06,
413
+ "loss": 0.0688,
414
+ "step": 600
415
+ },
416
+ {
417
+ "epoch": 0.75,
418
+ "eval_accuracy": 0.5,
419
+ "eval_loss": 0.011897406540811062,
420
+ "eval_runtime": 44.0698,
421
+ "eval_samples_per_second": 64.171,
422
+ "eval_steps_per_second": 16.043,
423
+ "step": 600
424
+ },
425
+ {
426
+ "epoch": 0.77,
427
+ "learning_rate": 1.1499053760044425e-06,
428
+ "loss": 0.0558,
429
+ "step": 610
430
+ },
431
+ {
432
+ "epoch": 0.78,
433
+ "learning_rate": 1.0338241837835812e-06,
434
+ "loss": 0.1069,
435
+ "step": 620
436
+ },
437
+ {
438
+ "epoch": 0.79,
439
+ "learning_rate": 9.231550136399099e-07,
440
+ "loss": 0.1019,
441
+ "step": 630
442
+ },
443
+ {
444
+ "epoch": 0.8,
445
+ "learning_rate": 8.180706623453892e-07,
446
+ "loss": 0.0985,
447
+ "step": 640
448
+ },
449
+ {
450
+ "epoch": 0.82,
451
+ "learning_rate": 7.187352066403012e-07,
452
+ "loss": 0.1097,
453
+ "step": 650
454
+ },
455
+ {
456
+ "epoch": 0.83,
457
+ "learning_rate": 6.253037470469e-07,
458
+ "loss": 0.0917,
459
+ "step": 660
460
+ },
461
+ {
462
+ "epoch": 0.84,
463
+ "learning_rate": 5.379221656983615e-07,
464
+ "loss": 0.037,
465
+ "step": 670
466
+ },
467
+ {
468
+ "epoch": 0.85,
469
+ "learning_rate": 4.567268985611408e-07,
470
+ "loss": 0.0333,
471
+ "step": 680
472
+ },
473
+ {
474
+ "epoch": 0.87,
475
+ "learning_rate": 3.8184472240641696e-07,
476
+ "loss": 0.0371,
477
+ "step": 690
478
+ },
479
+ {
480
+ "epoch": 0.88,
481
+ "learning_rate": 3.133925568632049e-07,
482
+ "loss": 0.0744,
483
+ "step": 700
484
+ },
485
+ {
486
+ "epoch": 0.88,
487
+ "eval_accuracy": 0.5,
488
+ "eval_loss": 0.011535878293216228,
489
+ "eval_runtime": 42.519,
490
+ "eval_samples_per_second": 66.511,
491
+ "eval_steps_per_second": 16.628,
492
+ "step": 700
493
+ }
494
+ ],
495
+ "logging_steps": 10,
496
+ "max_steps": 795,
497
+ "num_train_epochs": 1,
498
+ "save_steps": 100,
499
+ "total_flos": 0.0,
500
+ "trial_name": null,
501
+ "trial_params": null
502
+ }
checkpoint-700/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2434b2859f8752a8923164e66ca17d46083a046a373e000ea815eabbe3c1d156
3
+ size 4664