kiranr commited on
Commit
581dcae
1 Parent(s): 68380df

Upload model

Browse files
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "./camel-20b",
3
  "activation_function": "gelu",
4
  "architectures": [
5
  "GPT2LMHeadModel"
@@ -25,8 +25,8 @@
25
  "summary_proj_to_labels": true,
26
  "summary_type": "cls_index",
27
  "summary_use_proj": true,
28
- "torch_dtype": "float32",
29
- "transformers_version": "4.28.0.dev0",
30
  "use_cache": true,
31
  "vocab_size": 50258
32
  }
 
1
  {
2
+ "_name_or_path": "/home/ubuntu/server/InstructPalmyra-20b",
3
  "activation_function": "gelu",
4
  "architectures": [
5
  "GPT2LMHeadModel"
 
25
  "summary_proj_to_labels": true,
26
  "summary_type": "cls_index",
27
  "summary_use_proj": true,
28
+ "torch_dtype": "float16",
29
+ "transformers_version": "4.32.0.dev0",
30
  "use_cache": true,
31
  "vocab_size": 50258
32
  }
generation_config.json CHANGED
@@ -2,5 +2,5 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 50256,
4
  "eos_token_id": 50256,
5
- "transformers_version": "4.28.0.dev0"
6
  }
 
2
  "_from_model_config": true,
3
  "bos_token_id": 50256,
4
  "eos_token_id": 50256,
5
+ "transformers_version": "4.32.0.dev0"
6
  }
pytorch_model-00001-of-00005.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:016eb83728cc4f3ea3ad677700c287bf584d0011411efcb216caea4ddccfda31
3
+ size 9930627841
pytorch_model-00002-of-00005.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9be3e150665aa6b28567d6141258fa9fc148d5a54e820330673c5b5ad51a72eb
3
+ size 9967469263
pytorch_model-00003-of-00005.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a20b7b36a1e686f76ae5f3a98421d2ee9d6299ba8d0686181a34817fc1a60edd
3
+ size 9967469263
pytorch_model-00004-of-00005.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77278e4e765edd31cc085e4ef699443f610950a3668d9c51e1055bd063238f12
3
+ size 9967469263
pytorch_model-00005-of-00005.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f72cd3d8a5fc5b20d4b6dffb2288671a854c8dae19a5c57de227ab6871ae7c3
3
+ size 679603843
pytorch_model.bin.index.json CHANGED
@@ -1,628 +1,540 @@
1
  {
2
  "metadata": {
3
- "total_size": 82283118768.0
4
  },
5
  "weight_map": {
6
- "lm_head.weight": "pytorch_model-00009-of-00009.bin",
7
- "transformer.h.0.attn.bias": "pytorch_model-00001-of-00009.bin",
8
- "transformer.h.0.attn.c_attn.bias": "pytorch_model-00001-of-00009.bin",
9
- "transformer.h.0.attn.c_attn.weight": "pytorch_model-00001-of-00009.bin",
10
- "transformer.h.0.attn.c_proj.bias": "pytorch_model-00001-of-00009.bin",
11
- "transformer.h.0.attn.c_proj.weight": "pytorch_model-00001-of-00009.bin",
12
- "transformer.h.0.attn.masked_bias": "pytorch_model-00001-of-00009.bin",
13
- "transformer.h.0.ln_1.bias": "pytorch_model-00001-of-00009.bin",
14
- "transformer.h.0.ln_1.weight": "pytorch_model-00001-of-00009.bin",
15
- "transformer.h.0.ln_2.bias": "pytorch_model-00001-of-00009.bin",
16
- "transformer.h.0.ln_2.weight": "pytorch_model-00001-of-00009.bin",
17
- "transformer.h.0.mlp.c_fc.bias": "pytorch_model-00001-of-00009.bin",
18
- "transformer.h.0.mlp.c_fc.weight": "pytorch_model-00001-of-00009.bin",
19
- "transformer.h.0.mlp.c_proj.bias": "pytorch_model-00001-of-00009.bin",
20
- "transformer.h.0.mlp.c_proj.weight": "pytorch_model-00001-of-00009.bin",
21
- "transformer.h.1.attn.bias": "pytorch_model-00001-of-00009.bin",
22
- "transformer.h.1.attn.c_attn.bias": "pytorch_model-00001-of-00009.bin",
23
- "transformer.h.1.attn.c_attn.weight": "pytorch_model-00001-of-00009.bin",
24
- "transformer.h.1.attn.c_proj.bias": "pytorch_model-00001-of-00009.bin",
25
- "transformer.h.1.attn.c_proj.weight": "pytorch_model-00001-of-00009.bin",
26
- "transformer.h.1.attn.masked_bias": "pytorch_model-00001-of-00009.bin",
27
- "transformer.h.1.ln_1.bias": "pytorch_model-00001-of-00009.bin",
28
- "transformer.h.1.ln_1.weight": "pytorch_model-00001-of-00009.bin",
29
- "transformer.h.1.ln_2.bias": "pytorch_model-00001-of-00009.bin",
30
- "transformer.h.1.ln_2.weight": "pytorch_model-00001-of-00009.bin",
31
- "transformer.h.1.mlp.c_fc.bias": "pytorch_model-00001-of-00009.bin",
32
- "transformer.h.1.mlp.c_fc.weight": "pytorch_model-00001-of-00009.bin",
33
- "transformer.h.1.mlp.c_proj.bias": "pytorch_model-00001-of-00009.bin",
34
- "transformer.h.1.mlp.c_proj.weight": "pytorch_model-00001-of-00009.bin",
35
- "transformer.h.10.attn.bias": "pytorch_model-00002-of-00009.bin",
36
- "transformer.h.10.attn.c_attn.bias": "pytorch_model-00003-of-00009.bin",
37
- "transformer.h.10.attn.c_attn.weight": "pytorch_model-00003-of-00009.bin",
38
- "transformer.h.10.attn.c_proj.bias": "pytorch_model-00003-of-00009.bin",
39
- "transformer.h.10.attn.c_proj.weight": "pytorch_model-00003-of-00009.bin",
40
- "transformer.h.10.attn.masked_bias": "pytorch_model-00002-of-00009.bin",
41
- "transformer.h.10.ln_1.bias": "pytorch_model-00002-of-00009.bin",
42
- "transformer.h.10.ln_1.weight": "pytorch_model-00002-of-00009.bin",
43
- "transformer.h.10.ln_2.bias": "pytorch_model-00003-of-00009.bin",
44
- "transformer.h.10.ln_2.weight": "pytorch_model-00003-of-00009.bin",
45
- "transformer.h.10.mlp.c_fc.bias": "pytorch_model-00003-of-00009.bin",
46
- "transformer.h.10.mlp.c_fc.weight": "pytorch_model-00003-of-00009.bin",
47
- "transformer.h.10.mlp.c_proj.bias": "pytorch_model-00003-of-00009.bin",
48
- "transformer.h.10.mlp.c_proj.weight": "pytorch_model-00003-of-00009.bin",
49
- "transformer.h.11.attn.bias": "pytorch_model-00003-of-00009.bin",
50
- "transformer.h.11.attn.c_attn.bias": "pytorch_model-00003-of-00009.bin",
51
- "transformer.h.11.attn.c_attn.weight": "pytorch_model-00003-of-00009.bin",
52
- "transformer.h.11.attn.c_proj.bias": "pytorch_model-00003-of-00009.bin",
53
- "transformer.h.11.attn.c_proj.weight": "pytorch_model-00003-of-00009.bin",
54
- "transformer.h.11.attn.masked_bias": "pytorch_model-00003-of-00009.bin",
55
- "transformer.h.11.ln_1.bias": "pytorch_model-00003-of-00009.bin",
56
- "transformer.h.11.ln_1.weight": "pytorch_model-00003-of-00009.bin",
57
- "transformer.h.11.ln_2.bias": "pytorch_model-00003-of-00009.bin",
58
- "transformer.h.11.ln_2.weight": "pytorch_model-00003-of-00009.bin",
59
- "transformer.h.11.mlp.c_fc.bias": "pytorch_model-00003-of-00009.bin",
60
- "transformer.h.11.mlp.c_fc.weight": "pytorch_model-00003-of-00009.bin",
61
- "transformer.h.11.mlp.c_proj.bias": "pytorch_model-00003-of-00009.bin",
62
- "transformer.h.11.mlp.c_proj.weight": "pytorch_model-00003-of-00009.bin",
63
- "transformer.h.12.attn.bias": "pytorch_model-00003-of-00009.bin",
64
- "transformer.h.12.attn.c_attn.bias": "pytorch_model-00003-of-00009.bin",
65
- "transformer.h.12.attn.c_attn.weight": "pytorch_model-00003-of-00009.bin",
66
- "transformer.h.12.attn.c_proj.bias": "pytorch_model-00003-of-00009.bin",
67
- "transformer.h.12.attn.c_proj.weight": "pytorch_model-00003-of-00009.bin",
68
- "transformer.h.12.attn.masked_bias": "pytorch_model-00003-of-00009.bin",
69
- "transformer.h.12.ln_1.bias": "pytorch_model-00003-of-00009.bin",
70
- "transformer.h.12.ln_1.weight": "pytorch_model-00003-of-00009.bin",
71
- "transformer.h.12.ln_2.bias": "pytorch_model-00003-of-00009.bin",
72
- "transformer.h.12.ln_2.weight": "pytorch_model-00003-of-00009.bin",
73
- "transformer.h.12.mlp.c_fc.bias": "pytorch_model-00003-of-00009.bin",
74
- "transformer.h.12.mlp.c_fc.weight": "pytorch_model-00003-of-00009.bin",
75
- "transformer.h.12.mlp.c_proj.bias": "pytorch_model-00003-of-00009.bin",
76
- "transformer.h.12.mlp.c_proj.weight": "pytorch_model-00003-of-00009.bin",
77
- "transformer.h.13.attn.bias": "pytorch_model-00003-of-00009.bin",
78
- "transformer.h.13.attn.c_attn.bias": "pytorch_model-00003-of-00009.bin",
79
- "transformer.h.13.attn.c_attn.weight": "pytorch_model-00003-of-00009.bin",
80
- "transformer.h.13.attn.c_proj.bias": "pytorch_model-00003-of-00009.bin",
81
- "transformer.h.13.attn.c_proj.weight": "pytorch_model-00003-of-00009.bin",
82
- "transformer.h.13.attn.masked_bias": "pytorch_model-00003-of-00009.bin",
83
- "transformer.h.13.ln_1.bias": "pytorch_model-00003-of-00009.bin",
84
- "transformer.h.13.ln_1.weight": "pytorch_model-00003-of-00009.bin",
85
- "transformer.h.13.ln_2.bias": "pytorch_model-00003-of-00009.bin",
86
- "transformer.h.13.ln_2.weight": "pytorch_model-00003-of-00009.bin",
87
- "transformer.h.13.mlp.c_fc.bias": "pytorch_model-00003-of-00009.bin",
88
- "transformer.h.13.mlp.c_fc.weight": "pytorch_model-00003-of-00009.bin",
89
- "transformer.h.13.mlp.c_proj.bias": "pytorch_model-00003-of-00009.bin",
90
- "transformer.h.13.mlp.c_proj.weight": "pytorch_model-00003-of-00009.bin",
91
- "transformer.h.14.attn.bias": "pytorch_model-00003-of-00009.bin",
92
- "transformer.h.14.attn.c_attn.bias": "pytorch_model-00003-of-00009.bin",
93
- "transformer.h.14.attn.c_attn.weight": "pytorch_model-00003-of-00009.bin",
94
- "transformer.h.14.attn.c_proj.bias": "pytorch_model-00003-of-00009.bin",
95
- "transformer.h.14.attn.c_proj.weight": "pytorch_model-00003-of-00009.bin",
96
- "transformer.h.14.attn.masked_bias": "pytorch_model-00003-of-00009.bin",
97
- "transformer.h.14.ln_1.bias": "pytorch_model-00003-of-00009.bin",
98
- "transformer.h.14.ln_1.weight": "pytorch_model-00003-of-00009.bin",
99
- "transformer.h.14.ln_2.bias": "pytorch_model-00003-of-00009.bin",
100
- "transformer.h.14.ln_2.weight": "pytorch_model-00003-of-00009.bin",
101
- "transformer.h.14.mlp.c_fc.bias": "pytorch_model-00003-of-00009.bin",
102
- "transformer.h.14.mlp.c_fc.weight": "pytorch_model-00003-of-00009.bin",
103
- "transformer.h.14.mlp.c_proj.bias": "pytorch_model-00003-of-00009.bin",
104
- "transformer.h.14.mlp.c_proj.weight": "pytorch_model-00003-of-00009.bin",
105
- "transformer.h.15.attn.bias": "pytorch_model-00003-of-00009.bin",
106
- "transformer.h.15.attn.c_attn.bias": "pytorch_model-00003-of-00009.bin",
107
- "transformer.h.15.attn.c_attn.weight": "pytorch_model-00003-of-00009.bin",
108
- "transformer.h.15.attn.c_proj.bias": "pytorch_model-00003-of-00009.bin",
109
- "transformer.h.15.attn.c_proj.weight": "pytorch_model-00003-of-00009.bin",
110
- "transformer.h.15.attn.masked_bias": "pytorch_model-00003-of-00009.bin",
111
- "transformer.h.15.ln_1.bias": "pytorch_model-00003-of-00009.bin",
112
- "transformer.h.15.ln_1.weight": "pytorch_model-00003-of-00009.bin",
113
- "transformer.h.15.ln_2.bias": "pytorch_model-00003-of-00009.bin",
114
- "transformer.h.15.ln_2.weight": "pytorch_model-00003-of-00009.bin",
115
- "transformer.h.15.mlp.c_fc.bias": "pytorch_model-00004-of-00009.bin",
116
- "transformer.h.15.mlp.c_fc.weight": "pytorch_model-00004-of-00009.bin",
117
- "transformer.h.15.mlp.c_proj.bias": "pytorch_model-00004-of-00009.bin",
118
- "transformer.h.15.mlp.c_proj.weight": "pytorch_model-00004-of-00009.bin",
119
- "transformer.h.16.attn.bias": "pytorch_model-00004-of-00009.bin",
120
- "transformer.h.16.attn.c_attn.bias": "pytorch_model-00004-of-00009.bin",
121
- "transformer.h.16.attn.c_attn.weight": "pytorch_model-00004-of-00009.bin",
122
- "transformer.h.16.attn.c_proj.bias": "pytorch_model-00004-of-00009.bin",
123
- "transformer.h.16.attn.c_proj.weight": "pytorch_model-00004-of-00009.bin",
124
- "transformer.h.16.attn.masked_bias": "pytorch_model-00004-of-00009.bin",
125
- "transformer.h.16.ln_1.bias": "pytorch_model-00004-of-00009.bin",
126
- "transformer.h.16.ln_1.weight": "pytorch_model-00004-of-00009.bin",
127
- "transformer.h.16.ln_2.bias": "pytorch_model-00004-of-00009.bin",
128
- "transformer.h.16.ln_2.weight": "pytorch_model-00004-of-00009.bin",
129
- "transformer.h.16.mlp.c_fc.bias": "pytorch_model-00004-of-00009.bin",
130
- "transformer.h.16.mlp.c_fc.weight": "pytorch_model-00004-of-00009.bin",
131
- "transformer.h.16.mlp.c_proj.bias": "pytorch_model-00004-of-00009.bin",
132
- "transformer.h.16.mlp.c_proj.weight": "pytorch_model-00004-of-00009.bin",
133
- "transformer.h.17.attn.bias": "pytorch_model-00004-of-00009.bin",
134
- "transformer.h.17.attn.c_attn.bias": "pytorch_model-00004-of-00009.bin",
135
- "transformer.h.17.attn.c_attn.weight": "pytorch_model-00004-of-00009.bin",
136
- "transformer.h.17.attn.c_proj.bias": "pytorch_model-00004-of-00009.bin",
137
- "transformer.h.17.attn.c_proj.weight": "pytorch_model-00004-of-00009.bin",
138
- "transformer.h.17.attn.masked_bias": "pytorch_model-00004-of-00009.bin",
139
- "transformer.h.17.ln_1.bias": "pytorch_model-00004-of-00009.bin",
140
- "transformer.h.17.ln_1.weight": "pytorch_model-00004-of-00009.bin",
141
- "transformer.h.17.ln_2.bias": "pytorch_model-00004-of-00009.bin",
142
- "transformer.h.17.ln_2.weight": "pytorch_model-00004-of-00009.bin",
143
- "transformer.h.17.mlp.c_fc.bias": "pytorch_model-00004-of-00009.bin",
144
- "transformer.h.17.mlp.c_fc.weight": "pytorch_model-00004-of-00009.bin",
145
- "transformer.h.17.mlp.c_proj.bias": "pytorch_model-00004-of-00009.bin",
146
- "transformer.h.17.mlp.c_proj.weight": "pytorch_model-00004-of-00009.bin",
147
- "transformer.h.18.attn.bias": "pytorch_model-00004-of-00009.bin",
148
- "transformer.h.18.attn.c_attn.bias": "pytorch_model-00004-of-00009.bin",
149
- "transformer.h.18.attn.c_attn.weight": "pytorch_model-00004-of-00009.bin",
150
- "transformer.h.18.attn.c_proj.bias": "pytorch_model-00004-of-00009.bin",
151
- "transformer.h.18.attn.c_proj.weight": "pytorch_model-00004-of-00009.bin",
152
- "transformer.h.18.attn.masked_bias": "pytorch_model-00004-of-00009.bin",
153
- "transformer.h.18.ln_1.bias": "pytorch_model-00004-of-00009.bin",
154
- "transformer.h.18.ln_1.weight": "pytorch_model-00004-of-00009.bin",
155
- "transformer.h.18.ln_2.bias": "pytorch_model-00004-of-00009.bin",
156
- "transformer.h.18.ln_2.weight": "pytorch_model-00004-of-00009.bin",
157
- "transformer.h.18.mlp.c_fc.bias": "pytorch_model-00004-of-00009.bin",
158
- "transformer.h.18.mlp.c_fc.weight": "pytorch_model-00004-of-00009.bin",
159
- "transformer.h.18.mlp.c_proj.bias": "pytorch_model-00004-of-00009.bin",
160
- "transformer.h.18.mlp.c_proj.weight": "pytorch_model-00004-of-00009.bin",
161
- "transformer.h.19.attn.bias": "pytorch_model-00004-of-00009.bin",
162
- "transformer.h.19.attn.c_attn.bias": "pytorch_model-00004-of-00009.bin",
163
- "transformer.h.19.attn.c_attn.weight": "pytorch_model-00004-of-00009.bin",
164
- "transformer.h.19.attn.c_proj.bias": "pytorch_model-00004-of-00009.bin",
165
- "transformer.h.19.attn.c_proj.weight": "pytorch_model-00004-of-00009.bin",
166
- "transformer.h.19.attn.masked_bias": "pytorch_model-00004-of-00009.bin",
167
- "transformer.h.19.ln_1.bias": "pytorch_model-00004-of-00009.bin",
168
- "transformer.h.19.ln_1.weight": "pytorch_model-00004-of-00009.bin",
169
- "transformer.h.19.ln_2.bias": "pytorch_model-00004-of-00009.bin",
170
- "transformer.h.19.ln_2.weight": "pytorch_model-00004-of-00009.bin",
171
- "transformer.h.19.mlp.c_fc.bias": "pytorch_model-00004-of-00009.bin",
172
- "transformer.h.19.mlp.c_fc.weight": "pytorch_model-00004-of-00009.bin",
173
- "transformer.h.19.mlp.c_proj.bias": "pytorch_model-00004-of-00009.bin",
174
- "transformer.h.19.mlp.c_proj.weight": "pytorch_model-00004-of-00009.bin",
175
- "transformer.h.2.attn.bias": "pytorch_model-00001-of-00009.bin",
176
- "transformer.h.2.attn.c_attn.bias": "pytorch_model-00001-of-00009.bin",
177
- "transformer.h.2.attn.c_attn.weight": "pytorch_model-00001-of-00009.bin",
178
- "transformer.h.2.attn.c_proj.bias": "pytorch_model-00001-of-00009.bin",
179
- "transformer.h.2.attn.c_proj.weight": "pytorch_model-00001-of-00009.bin",
180
- "transformer.h.2.attn.masked_bias": "pytorch_model-00001-of-00009.bin",
181
- "transformer.h.2.ln_1.bias": "pytorch_model-00001-of-00009.bin",
182
- "transformer.h.2.ln_1.weight": "pytorch_model-00001-of-00009.bin",
183
- "transformer.h.2.ln_2.bias": "pytorch_model-00001-of-00009.bin",
184
- "transformer.h.2.ln_2.weight": "pytorch_model-00001-of-00009.bin",
185
- "transformer.h.2.mlp.c_fc.bias": "pytorch_model-00001-of-00009.bin",
186
- "transformer.h.2.mlp.c_fc.weight": "pytorch_model-00001-of-00009.bin",
187
- "transformer.h.2.mlp.c_proj.bias": "pytorch_model-00001-of-00009.bin",
188
- "transformer.h.2.mlp.c_proj.weight": "pytorch_model-00001-of-00009.bin",
189
- "transformer.h.20.attn.bias": "pytorch_model-00004-of-00009.bin",
190
- "transformer.h.20.attn.c_attn.bias": "pytorch_model-00004-of-00009.bin",
191
- "transformer.h.20.attn.c_attn.weight": "pytorch_model-00004-of-00009.bin",
192
- "transformer.h.20.attn.c_proj.bias": "pytorch_model-00004-of-00009.bin",
193
- "transformer.h.20.attn.c_proj.weight": "pytorch_model-00004-of-00009.bin",
194
- "transformer.h.20.attn.masked_bias": "pytorch_model-00004-of-00009.bin",
195
- "transformer.h.20.ln_1.bias": "pytorch_model-00004-of-00009.bin",
196
- "transformer.h.20.ln_1.weight": "pytorch_model-00004-of-00009.bin",
197
- "transformer.h.20.ln_2.bias": "pytorch_model-00004-of-00009.bin",
198
- "transformer.h.20.ln_2.weight": "pytorch_model-00004-of-00009.bin",
199
- "transformer.h.20.mlp.c_fc.bias": "pytorch_model-00004-of-00009.bin",
200
- "transformer.h.20.mlp.c_fc.weight": "pytorch_model-00004-of-00009.bin",
201
- "transformer.h.20.mlp.c_proj.bias": "pytorch_model-00005-of-00009.bin",
202
- "transformer.h.20.mlp.c_proj.weight": "pytorch_model-00005-of-00009.bin",
203
- "transformer.h.21.attn.bias": "pytorch_model-00005-of-00009.bin",
204
- "transformer.h.21.attn.c_attn.bias": "pytorch_model-00005-of-00009.bin",
205
- "transformer.h.21.attn.c_attn.weight": "pytorch_model-00005-of-00009.bin",
206
- "transformer.h.21.attn.c_proj.bias": "pytorch_model-00005-of-00009.bin",
207
- "transformer.h.21.attn.c_proj.weight": "pytorch_model-00005-of-00009.bin",
208
- "transformer.h.21.attn.masked_bias": "pytorch_model-00005-of-00009.bin",
209
- "transformer.h.21.ln_1.bias": "pytorch_model-00005-of-00009.bin",
210
- "transformer.h.21.ln_1.weight": "pytorch_model-00005-of-00009.bin",
211
- "transformer.h.21.ln_2.bias": "pytorch_model-00005-of-00009.bin",
212
- "transformer.h.21.ln_2.weight": "pytorch_model-00005-of-00009.bin",
213
- "transformer.h.21.mlp.c_fc.bias": "pytorch_model-00005-of-00009.bin",
214
- "transformer.h.21.mlp.c_fc.weight": "pytorch_model-00005-of-00009.bin",
215
- "transformer.h.21.mlp.c_proj.bias": "pytorch_model-00005-of-00009.bin",
216
- "transformer.h.21.mlp.c_proj.weight": "pytorch_model-00005-of-00009.bin",
217
- "transformer.h.22.attn.bias": "pytorch_model-00005-of-00009.bin",
218
- "transformer.h.22.attn.c_attn.bias": "pytorch_model-00005-of-00009.bin",
219
- "transformer.h.22.attn.c_attn.weight": "pytorch_model-00005-of-00009.bin",
220
- "transformer.h.22.attn.c_proj.bias": "pytorch_model-00005-of-00009.bin",
221
- "transformer.h.22.attn.c_proj.weight": "pytorch_model-00005-of-00009.bin",
222
- "transformer.h.22.attn.masked_bias": "pytorch_model-00005-of-00009.bin",
223
- "transformer.h.22.ln_1.bias": "pytorch_model-00005-of-00009.bin",
224
- "transformer.h.22.ln_1.weight": "pytorch_model-00005-of-00009.bin",
225
- "transformer.h.22.ln_2.bias": "pytorch_model-00005-of-00009.bin",
226
- "transformer.h.22.ln_2.weight": "pytorch_model-00005-of-00009.bin",
227
- "transformer.h.22.mlp.c_fc.bias": "pytorch_model-00005-of-00009.bin",
228
- "transformer.h.22.mlp.c_fc.weight": "pytorch_model-00005-of-00009.bin",
229
- "transformer.h.22.mlp.c_proj.bias": "pytorch_model-00005-of-00009.bin",
230
- "transformer.h.22.mlp.c_proj.weight": "pytorch_model-00005-of-00009.bin",
231
- "transformer.h.23.attn.bias": "pytorch_model-00005-of-00009.bin",
232
- "transformer.h.23.attn.c_attn.bias": "pytorch_model-00005-of-00009.bin",
233
- "transformer.h.23.attn.c_attn.weight": "pytorch_model-00005-of-00009.bin",
234
- "transformer.h.23.attn.c_proj.bias": "pytorch_model-00005-of-00009.bin",
235
- "transformer.h.23.attn.c_proj.weight": "pytorch_model-00005-of-00009.bin",
236
- "transformer.h.23.attn.masked_bias": "pytorch_model-00005-of-00009.bin",
237
- "transformer.h.23.ln_1.bias": "pytorch_model-00005-of-00009.bin",
238
- "transformer.h.23.ln_1.weight": "pytorch_model-00005-of-00009.bin",
239
- "transformer.h.23.ln_2.bias": "pytorch_model-00005-of-00009.bin",
240
- "transformer.h.23.ln_2.weight": "pytorch_model-00005-of-00009.bin",
241
- "transformer.h.23.mlp.c_fc.bias": "pytorch_model-00005-of-00009.bin",
242
- "transformer.h.23.mlp.c_fc.weight": "pytorch_model-00005-of-00009.bin",
243
- "transformer.h.23.mlp.c_proj.bias": "pytorch_model-00005-of-00009.bin",
244
- "transformer.h.23.mlp.c_proj.weight": "pytorch_model-00005-of-00009.bin",
245
- "transformer.h.24.attn.bias": "pytorch_model-00005-of-00009.bin",
246
- "transformer.h.24.attn.c_attn.bias": "pytorch_model-00005-of-00009.bin",
247
- "transformer.h.24.attn.c_attn.weight": "pytorch_model-00005-of-00009.bin",
248
- "transformer.h.24.attn.c_proj.bias": "pytorch_model-00005-of-00009.bin",
249
- "transformer.h.24.attn.c_proj.weight": "pytorch_model-00005-of-00009.bin",
250
- "transformer.h.24.attn.masked_bias": "pytorch_model-00005-of-00009.bin",
251
- "transformer.h.24.ln_1.bias": "pytorch_model-00005-of-00009.bin",
252
- "transformer.h.24.ln_1.weight": "pytorch_model-00005-of-00009.bin",
253
- "transformer.h.24.ln_2.bias": "pytorch_model-00005-of-00009.bin",
254
- "transformer.h.24.ln_2.weight": "pytorch_model-00005-of-00009.bin",
255
- "transformer.h.24.mlp.c_fc.bias": "pytorch_model-00005-of-00009.bin",
256
- "transformer.h.24.mlp.c_fc.weight": "pytorch_model-00005-of-00009.bin",
257
- "transformer.h.24.mlp.c_proj.bias": "pytorch_model-00005-of-00009.bin",
258
- "transformer.h.24.mlp.c_proj.weight": "pytorch_model-00005-of-00009.bin",
259
- "transformer.h.25.attn.bias": "pytorch_model-00005-of-00009.bin",
260
- "transformer.h.25.attn.c_attn.bias": "pytorch_model-00005-of-00009.bin",
261
- "transformer.h.25.attn.c_attn.weight": "pytorch_model-00005-of-00009.bin",
262
- "transformer.h.25.attn.c_proj.bias": "pytorch_model-00005-of-00009.bin",
263
- "transformer.h.25.attn.c_proj.weight": "pytorch_model-00005-of-00009.bin",
264
- "transformer.h.25.attn.masked_bias": "pytorch_model-00005-of-00009.bin",
265
- "transformer.h.25.ln_1.bias": "pytorch_model-00005-of-00009.bin",
266
- "transformer.h.25.ln_1.weight": "pytorch_model-00005-of-00009.bin",
267
- "transformer.h.25.ln_2.bias": "pytorch_model-00005-of-00009.bin",
268
- "transformer.h.25.ln_2.weight": "pytorch_model-00005-of-00009.bin",
269
- "transformer.h.25.mlp.c_fc.bias": "pytorch_model-00005-of-00009.bin",
270
- "transformer.h.25.mlp.c_fc.weight": "pytorch_model-00005-of-00009.bin",
271
- "transformer.h.25.mlp.c_proj.bias": "pytorch_model-00005-of-00009.bin",
272
- "transformer.h.25.mlp.c_proj.weight": "pytorch_model-00005-of-00009.bin",
273
- "transformer.h.26.attn.bias": "pytorch_model-00005-of-00009.bin",
274
- "transformer.h.26.attn.c_attn.bias": "pytorch_model-00006-of-00009.bin",
275
- "transformer.h.26.attn.c_attn.weight": "pytorch_model-00006-of-00009.bin",
276
- "transformer.h.26.attn.c_proj.bias": "pytorch_model-00006-of-00009.bin",
277
- "transformer.h.26.attn.c_proj.weight": "pytorch_model-00006-of-00009.bin",
278
- "transformer.h.26.attn.masked_bias": "pytorch_model-00005-of-00009.bin",
279
- "transformer.h.26.ln_1.bias": "pytorch_model-00005-of-00009.bin",
280
- "transformer.h.26.ln_1.weight": "pytorch_model-00005-of-00009.bin",
281
- "transformer.h.26.ln_2.bias": "pytorch_model-00006-of-00009.bin",
282
- "transformer.h.26.ln_2.weight": "pytorch_model-00006-of-00009.bin",
283
- "transformer.h.26.mlp.c_fc.bias": "pytorch_model-00006-of-00009.bin",
284
- "transformer.h.26.mlp.c_fc.weight": "pytorch_model-00006-of-00009.bin",
285
- "transformer.h.26.mlp.c_proj.bias": "pytorch_model-00006-of-00009.bin",
286
- "transformer.h.26.mlp.c_proj.weight": "pytorch_model-00006-of-00009.bin",
287
- "transformer.h.27.attn.bias": "pytorch_model-00006-of-00009.bin",
288
- "transformer.h.27.attn.c_attn.bias": "pytorch_model-00006-of-00009.bin",
289
- "transformer.h.27.attn.c_attn.weight": "pytorch_model-00006-of-00009.bin",
290
- "transformer.h.27.attn.c_proj.bias": "pytorch_model-00006-of-00009.bin",
291
- "transformer.h.27.attn.c_proj.weight": "pytorch_model-00006-of-00009.bin",
292
- "transformer.h.27.attn.masked_bias": "pytorch_model-00006-of-00009.bin",
293
- "transformer.h.27.ln_1.bias": "pytorch_model-00006-of-00009.bin",
294
- "transformer.h.27.ln_1.weight": "pytorch_model-00006-of-00009.bin",
295
- "transformer.h.27.ln_2.bias": "pytorch_model-00006-of-00009.bin",
296
- "transformer.h.27.ln_2.weight": "pytorch_model-00006-of-00009.bin",
297
- "transformer.h.27.mlp.c_fc.bias": "pytorch_model-00006-of-00009.bin",
298
- "transformer.h.27.mlp.c_fc.weight": "pytorch_model-00006-of-00009.bin",
299
- "transformer.h.27.mlp.c_proj.bias": "pytorch_model-00006-of-00009.bin",
300
- "transformer.h.27.mlp.c_proj.weight": "pytorch_model-00006-of-00009.bin",
301
- "transformer.h.28.attn.bias": "pytorch_model-00006-of-00009.bin",
302
- "transformer.h.28.attn.c_attn.bias": "pytorch_model-00006-of-00009.bin",
303
- "transformer.h.28.attn.c_attn.weight": "pytorch_model-00006-of-00009.bin",
304
- "transformer.h.28.attn.c_proj.bias": "pytorch_model-00006-of-00009.bin",
305
- "transformer.h.28.attn.c_proj.weight": "pytorch_model-00006-of-00009.bin",
306
- "transformer.h.28.attn.masked_bias": "pytorch_model-00006-of-00009.bin",
307
- "transformer.h.28.ln_1.bias": "pytorch_model-00006-of-00009.bin",
308
- "transformer.h.28.ln_1.weight": "pytorch_model-00006-of-00009.bin",
309
- "transformer.h.28.ln_2.bias": "pytorch_model-00006-of-00009.bin",
310
- "transformer.h.28.ln_2.weight": "pytorch_model-00006-of-00009.bin",
311
- "transformer.h.28.mlp.c_fc.bias": "pytorch_model-00006-of-00009.bin",
312
- "transformer.h.28.mlp.c_fc.weight": "pytorch_model-00006-of-00009.bin",
313
- "transformer.h.28.mlp.c_proj.bias": "pytorch_model-00006-of-00009.bin",
314
- "transformer.h.28.mlp.c_proj.weight": "pytorch_model-00006-of-00009.bin",
315
- "transformer.h.29.attn.bias": "pytorch_model-00006-of-00009.bin",
316
- "transformer.h.29.attn.c_attn.bias": "pytorch_model-00006-of-00009.bin",
317
- "transformer.h.29.attn.c_attn.weight": "pytorch_model-00006-of-00009.bin",
318
- "transformer.h.29.attn.c_proj.bias": "pytorch_model-00006-of-00009.bin",
319
- "transformer.h.29.attn.c_proj.weight": "pytorch_model-00006-of-00009.bin",
320
- "transformer.h.29.attn.masked_bias": "pytorch_model-00006-of-00009.bin",
321
- "transformer.h.29.ln_1.bias": "pytorch_model-00006-of-00009.bin",
322
- "transformer.h.29.ln_1.weight": "pytorch_model-00006-of-00009.bin",
323
- "transformer.h.29.ln_2.bias": "pytorch_model-00006-of-00009.bin",
324
- "transformer.h.29.ln_2.weight": "pytorch_model-00006-of-00009.bin",
325
- "transformer.h.29.mlp.c_fc.bias": "pytorch_model-00006-of-00009.bin",
326
- "transformer.h.29.mlp.c_fc.weight": "pytorch_model-00006-of-00009.bin",
327
- "transformer.h.29.mlp.c_proj.bias": "pytorch_model-00006-of-00009.bin",
328
- "transformer.h.29.mlp.c_proj.weight": "pytorch_model-00006-of-00009.bin",
329
- "transformer.h.3.attn.bias": "pytorch_model-00001-of-00009.bin",
330
- "transformer.h.3.attn.c_attn.bias": "pytorch_model-00001-of-00009.bin",
331
- "transformer.h.3.attn.c_attn.weight": "pytorch_model-00001-of-00009.bin",
332
- "transformer.h.3.attn.c_proj.bias": "pytorch_model-00001-of-00009.bin",
333
- "transformer.h.3.attn.c_proj.weight": "pytorch_model-00001-of-00009.bin",
334
- "transformer.h.3.attn.masked_bias": "pytorch_model-00001-of-00009.bin",
335
- "transformer.h.3.ln_1.bias": "pytorch_model-00001-of-00009.bin",
336
- "transformer.h.3.ln_1.weight": "pytorch_model-00001-of-00009.bin",
337
- "transformer.h.3.ln_2.bias": "pytorch_model-00001-of-00009.bin",
338
- "transformer.h.3.ln_2.weight": "pytorch_model-00001-of-00009.bin",
339
- "transformer.h.3.mlp.c_fc.bias": "pytorch_model-00001-of-00009.bin",
340
- "transformer.h.3.mlp.c_fc.weight": "pytorch_model-00001-of-00009.bin",
341
- "transformer.h.3.mlp.c_proj.bias": "pytorch_model-00001-of-00009.bin",
342
- "transformer.h.3.mlp.c_proj.weight": "pytorch_model-00001-of-00009.bin",
343
- "transformer.h.30.attn.bias": "pytorch_model-00006-of-00009.bin",
344
- "transformer.h.30.attn.c_attn.bias": "pytorch_model-00006-of-00009.bin",
345
- "transformer.h.30.attn.c_attn.weight": "pytorch_model-00006-of-00009.bin",
346
- "transformer.h.30.attn.c_proj.bias": "pytorch_model-00006-of-00009.bin",
347
- "transformer.h.30.attn.c_proj.weight": "pytorch_model-00006-of-00009.bin",
348
- "transformer.h.30.attn.masked_bias": "pytorch_model-00006-of-00009.bin",
349
- "transformer.h.30.ln_1.bias": "pytorch_model-00006-of-00009.bin",
350
- "transformer.h.30.ln_1.weight": "pytorch_model-00006-of-00009.bin",
351
- "transformer.h.30.ln_2.bias": "pytorch_model-00006-of-00009.bin",
352
- "transformer.h.30.ln_2.weight": "pytorch_model-00006-of-00009.bin",
353
- "transformer.h.30.mlp.c_fc.bias": "pytorch_model-00006-of-00009.bin",
354
- "transformer.h.30.mlp.c_fc.weight": "pytorch_model-00006-of-00009.bin",
355
- "transformer.h.30.mlp.c_proj.bias": "pytorch_model-00006-of-00009.bin",
356
- "transformer.h.30.mlp.c_proj.weight": "pytorch_model-00006-of-00009.bin",
357
- "transformer.h.31.attn.bias": "pytorch_model-00006-of-00009.bin",
358
- "transformer.h.31.attn.c_attn.bias": "pytorch_model-00006-of-00009.bin",
359
- "transformer.h.31.attn.c_attn.weight": "pytorch_model-00006-of-00009.bin",
360
- "transformer.h.31.attn.c_proj.bias": "pytorch_model-00006-of-00009.bin",
361
- "transformer.h.31.attn.c_proj.weight": "pytorch_model-00006-of-00009.bin",
362
- "transformer.h.31.attn.masked_bias": "pytorch_model-00006-of-00009.bin",
363
- "transformer.h.31.ln_1.bias": "pytorch_model-00006-of-00009.bin",
364
- "transformer.h.31.ln_1.weight": "pytorch_model-00006-of-00009.bin",
365
- "transformer.h.31.ln_2.bias": "pytorch_model-00006-of-00009.bin",
366
- "transformer.h.31.ln_2.weight": "pytorch_model-00006-of-00009.bin",
367
- "transformer.h.31.mlp.c_fc.bias": "pytorch_model-00007-of-00009.bin",
368
- "transformer.h.31.mlp.c_fc.weight": "pytorch_model-00007-of-00009.bin",
369
- "transformer.h.31.mlp.c_proj.bias": "pytorch_model-00007-of-00009.bin",
370
- "transformer.h.31.mlp.c_proj.weight": "pytorch_model-00007-of-00009.bin",
371
- "transformer.h.32.attn.bias": "pytorch_model-00007-of-00009.bin",
372
- "transformer.h.32.attn.c_attn.bias": "pytorch_model-00007-of-00009.bin",
373
- "transformer.h.32.attn.c_attn.weight": "pytorch_model-00007-of-00009.bin",
374
- "transformer.h.32.attn.c_proj.bias": "pytorch_model-00007-of-00009.bin",
375
- "transformer.h.32.attn.c_proj.weight": "pytorch_model-00007-of-00009.bin",
376
- "transformer.h.32.attn.masked_bias": "pytorch_model-00007-of-00009.bin",
377
- "transformer.h.32.ln_1.bias": "pytorch_model-00007-of-00009.bin",
378
- "transformer.h.32.ln_1.weight": "pytorch_model-00007-of-00009.bin",
379
- "transformer.h.32.ln_2.bias": "pytorch_model-00007-of-00009.bin",
380
- "transformer.h.32.ln_2.weight": "pytorch_model-00007-of-00009.bin",
381
- "transformer.h.32.mlp.c_fc.bias": "pytorch_model-00007-of-00009.bin",
382
- "transformer.h.32.mlp.c_fc.weight": "pytorch_model-00007-of-00009.bin",
383
- "transformer.h.32.mlp.c_proj.bias": "pytorch_model-00007-of-00009.bin",
384
- "transformer.h.32.mlp.c_proj.weight": "pytorch_model-00007-of-00009.bin",
385
- "transformer.h.33.attn.bias": "pytorch_model-00007-of-00009.bin",
386
- "transformer.h.33.attn.c_attn.bias": "pytorch_model-00007-of-00009.bin",
387
- "transformer.h.33.attn.c_attn.weight": "pytorch_model-00007-of-00009.bin",
388
- "transformer.h.33.attn.c_proj.bias": "pytorch_model-00007-of-00009.bin",
389
- "transformer.h.33.attn.c_proj.weight": "pytorch_model-00007-of-00009.bin",
390
- "transformer.h.33.attn.masked_bias": "pytorch_model-00007-of-00009.bin",
391
- "transformer.h.33.ln_1.bias": "pytorch_model-00007-of-00009.bin",
392
- "transformer.h.33.ln_1.weight": "pytorch_model-00007-of-00009.bin",
393
- "transformer.h.33.ln_2.bias": "pytorch_model-00007-of-00009.bin",
394
- "transformer.h.33.ln_2.weight": "pytorch_model-00007-of-00009.bin",
395
- "transformer.h.33.mlp.c_fc.bias": "pytorch_model-00007-of-00009.bin",
396
- "transformer.h.33.mlp.c_fc.weight": "pytorch_model-00007-of-00009.bin",
397
- "transformer.h.33.mlp.c_proj.bias": "pytorch_model-00007-of-00009.bin",
398
- "transformer.h.33.mlp.c_proj.weight": "pytorch_model-00007-of-00009.bin",
399
- "transformer.h.34.attn.bias": "pytorch_model-00007-of-00009.bin",
400
- "transformer.h.34.attn.c_attn.bias": "pytorch_model-00007-of-00009.bin",
401
- "transformer.h.34.attn.c_attn.weight": "pytorch_model-00007-of-00009.bin",
402
- "transformer.h.34.attn.c_proj.bias": "pytorch_model-00007-of-00009.bin",
403
- "transformer.h.34.attn.c_proj.weight": "pytorch_model-00007-of-00009.bin",
404
- "transformer.h.34.attn.masked_bias": "pytorch_model-00007-of-00009.bin",
405
- "transformer.h.34.ln_1.bias": "pytorch_model-00007-of-00009.bin",
406
- "transformer.h.34.ln_1.weight": "pytorch_model-00007-of-00009.bin",
407
- "transformer.h.34.ln_2.bias": "pytorch_model-00007-of-00009.bin",
408
- "transformer.h.34.ln_2.weight": "pytorch_model-00007-of-00009.bin",
409
- "transformer.h.34.mlp.c_fc.bias": "pytorch_model-00007-of-00009.bin",
410
- "transformer.h.34.mlp.c_fc.weight": "pytorch_model-00007-of-00009.bin",
411
- "transformer.h.34.mlp.c_proj.bias": "pytorch_model-00007-of-00009.bin",
412
- "transformer.h.34.mlp.c_proj.weight": "pytorch_model-00007-of-00009.bin",
413
- "transformer.h.35.attn.bias": "pytorch_model-00007-of-00009.bin",
414
- "transformer.h.35.attn.c_attn.bias": "pytorch_model-00007-of-00009.bin",
415
- "transformer.h.35.attn.c_attn.weight": "pytorch_model-00007-of-00009.bin",
416
- "transformer.h.35.attn.c_proj.bias": "pytorch_model-00007-of-00009.bin",
417
- "transformer.h.35.attn.c_proj.weight": "pytorch_model-00007-of-00009.bin",
418
- "transformer.h.35.attn.masked_bias": "pytorch_model-00007-of-00009.bin",
419
- "transformer.h.35.ln_1.bias": "pytorch_model-00007-of-00009.bin",
420
- "transformer.h.35.ln_1.weight": "pytorch_model-00007-of-00009.bin",
421
- "transformer.h.35.ln_2.bias": "pytorch_model-00007-of-00009.bin",
422
- "transformer.h.35.ln_2.weight": "pytorch_model-00007-of-00009.bin",
423
- "transformer.h.35.mlp.c_fc.bias": "pytorch_model-00007-of-00009.bin",
424
- "transformer.h.35.mlp.c_fc.weight": "pytorch_model-00007-of-00009.bin",
425
- "transformer.h.35.mlp.c_proj.bias": "pytorch_model-00007-of-00009.bin",
426
- "transformer.h.35.mlp.c_proj.weight": "pytorch_model-00007-of-00009.bin",
427
- "transformer.h.36.attn.bias": "pytorch_model-00007-of-00009.bin",
428
- "transformer.h.36.attn.c_attn.bias": "pytorch_model-00007-of-00009.bin",
429
- "transformer.h.36.attn.c_attn.weight": "pytorch_model-00007-of-00009.bin",
430
- "transformer.h.36.attn.c_proj.bias": "pytorch_model-00007-of-00009.bin",
431
- "transformer.h.36.attn.c_proj.weight": "pytorch_model-00007-of-00009.bin",
432
- "transformer.h.36.attn.masked_bias": "pytorch_model-00007-of-00009.bin",
433
- "transformer.h.36.ln_1.bias": "pytorch_model-00007-of-00009.bin",
434
- "transformer.h.36.ln_1.weight": "pytorch_model-00007-of-00009.bin",
435
- "transformer.h.36.ln_2.bias": "pytorch_model-00007-of-00009.bin",
436
- "transformer.h.36.ln_2.weight": "pytorch_model-00007-of-00009.bin",
437
- "transformer.h.36.mlp.c_fc.bias": "pytorch_model-00007-of-00009.bin",
438
- "transformer.h.36.mlp.c_fc.weight": "pytorch_model-00007-of-00009.bin",
439
- "transformer.h.36.mlp.c_proj.bias": "pytorch_model-00008-of-00009.bin",
440
- "transformer.h.36.mlp.c_proj.weight": "pytorch_model-00008-of-00009.bin",
441
- "transformer.h.37.attn.bias": "pytorch_model-00008-of-00009.bin",
442
- "transformer.h.37.attn.c_attn.bias": "pytorch_model-00008-of-00009.bin",
443
- "transformer.h.37.attn.c_attn.weight": "pytorch_model-00008-of-00009.bin",
444
- "transformer.h.37.attn.c_proj.bias": "pytorch_model-00008-of-00009.bin",
445
- "transformer.h.37.attn.c_proj.weight": "pytorch_model-00008-of-00009.bin",
446
- "transformer.h.37.attn.masked_bias": "pytorch_model-00008-of-00009.bin",
447
- "transformer.h.37.ln_1.bias": "pytorch_model-00008-of-00009.bin",
448
- "transformer.h.37.ln_1.weight": "pytorch_model-00008-of-00009.bin",
449
- "transformer.h.37.ln_2.bias": "pytorch_model-00008-of-00009.bin",
450
- "transformer.h.37.ln_2.weight": "pytorch_model-00008-of-00009.bin",
451
- "transformer.h.37.mlp.c_fc.bias": "pytorch_model-00008-of-00009.bin",
452
- "transformer.h.37.mlp.c_fc.weight": "pytorch_model-00008-of-00009.bin",
453
- "transformer.h.37.mlp.c_proj.bias": "pytorch_model-00008-of-00009.bin",
454
- "transformer.h.37.mlp.c_proj.weight": "pytorch_model-00008-of-00009.bin",
455
- "transformer.h.38.attn.bias": "pytorch_model-00008-of-00009.bin",
456
- "transformer.h.38.attn.c_attn.bias": "pytorch_model-00008-of-00009.bin",
457
- "transformer.h.38.attn.c_attn.weight": "pytorch_model-00008-of-00009.bin",
458
- "transformer.h.38.attn.c_proj.bias": "pytorch_model-00008-of-00009.bin",
459
- "transformer.h.38.attn.c_proj.weight": "pytorch_model-00008-of-00009.bin",
460
- "transformer.h.38.attn.masked_bias": "pytorch_model-00008-of-00009.bin",
461
- "transformer.h.38.ln_1.bias": "pytorch_model-00008-of-00009.bin",
462
- "transformer.h.38.ln_1.weight": "pytorch_model-00008-of-00009.bin",
463
- "transformer.h.38.ln_2.bias": "pytorch_model-00008-of-00009.bin",
464
- "transformer.h.38.ln_2.weight": "pytorch_model-00008-of-00009.bin",
465
- "transformer.h.38.mlp.c_fc.bias": "pytorch_model-00008-of-00009.bin",
466
- "transformer.h.38.mlp.c_fc.weight": "pytorch_model-00008-of-00009.bin",
467
- "transformer.h.38.mlp.c_proj.bias": "pytorch_model-00008-of-00009.bin",
468
- "transformer.h.38.mlp.c_proj.weight": "pytorch_model-00008-of-00009.bin",
469
- "transformer.h.39.attn.bias": "pytorch_model-00008-of-00009.bin",
470
- "transformer.h.39.attn.c_attn.bias": "pytorch_model-00008-of-00009.bin",
471
- "transformer.h.39.attn.c_attn.weight": "pytorch_model-00008-of-00009.bin",
472
- "transformer.h.39.attn.c_proj.bias": "pytorch_model-00008-of-00009.bin",
473
- "transformer.h.39.attn.c_proj.weight": "pytorch_model-00008-of-00009.bin",
474
- "transformer.h.39.attn.masked_bias": "pytorch_model-00008-of-00009.bin",
475
- "transformer.h.39.ln_1.bias": "pytorch_model-00008-of-00009.bin",
476
- "transformer.h.39.ln_1.weight": "pytorch_model-00008-of-00009.bin",
477
- "transformer.h.39.ln_2.bias": "pytorch_model-00008-of-00009.bin",
478
- "transformer.h.39.ln_2.weight": "pytorch_model-00008-of-00009.bin",
479
- "transformer.h.39.mlp.c_fc.bias": "pytorch_model-00008-of-00009.bin",
480
- "transformer.h.39.mlp.c_fc.weight": "pytorch_model-00008-of-00009.bin",
481
- "transformer.h.39.mlp.c_proj.bias": "pytorch_model-00008-of-00009.bin",
482
- "transformer.h.39.mlp.c_proj.weight": "pytorch_model-00008-of-00009.bin",
483
- "transformer.h.4.attn.bias": "pytorch_model-00001-of-00009.bin",
484
- "transformer.h.4.attn.c_attn.bias": "pytorch_model-00001-of-00009.bin",
485
- "transformer.h.4.attn.c_attn.weight": "pytorch_model-00001-of-00009.bin",
486
- "transformer.h.4.attn.c_proj.bias": "pytorch_model-00001-of-00009.bin",
487
- "transformer.h.4.attn.c_proj.weight": "pytorch_model-00001-of-00009.bin",
488
- "transformer.h.4.attn.masked_bias": "pytorch_model-00001-of-00009.bin",
489
- "transformer.h.4.ln_1.bias": "pytorch_model-00001-of-00009.bin",
490
- "transformer.h.4.ln_1.weight": "pytorch_model-00001-of-00009.bin",
491
- "transformer.h.4.ln_2.bias": "pytorch_model-00001-of-00009.bin",
492
- "transformer.h.4.ln_2.weight": "pytorch_model-00001-of-00009.bin",
493
- "transformer.h.4.mlp.c_fc.bias": "pytorch_model-00001-of-00009.bin",
494
- "transformer.h.4.mlp.c_fc.weight": "pytorch_model-00001-of-00009.bin",
495
- "transformer.h.4.mlp.c_proj.bias": "pytorch_model-00002-of-00009.bin",
496
- "transformer.h.4.mlp.c_proj.weight": "pytorch_model-00002-of-00009.bin",
497
- "transformer.h.40.attn.bias": "pytorch_model-00008-of-00009.bin",
498
- "transformer.h.40.attn.c_attn.bias": "pytorch_model-00008-of-00009.bin",
499
- "transformer.h.40.attn.c_attn.weight": "pytorch_model-00008-of-00009.bin",
500
- "transformer.h.40.attn.c_proj.bias": "pytorch_model-00008-of-00009.bin",
501
- "transformer.h.40.attn.c_proj.weight": "pytorch_model-00008-of-00009.bin",
502
- "transformer.h.40.attn.masked_bias": "pytorch_model-00008-of-00009.bin",
503
- "transformer.h.40.ln_1.bias": "pytorch_model-00008-of-00009.bin",
504
- "transformer.h.40.ln_1.weight": "pytorch_model-00008-of-00009.bin",
505
- "transformer.h.40.ln_2.bias": "pytorch_model-00008-of-00009.bin",
506
- "transformer.h.40.ln_2.weight": "pytorch_model-00008-of-00009.bin",
507
- "transformer.h.40.mlp.c_fc.bias": "pytorch_model-00008-of-00009.bin",
508
- "transformer.h.40.mlp.c_fc.weight": "pytorch_model-00008-of-00009.bin",
509
- "transformer.h.40.mlp.c_proj.bias": "pytorch_model-00008-of-00009.bin",
510
- "transformer.h.40.mlp.c_proj.weight": "pytorch_model-00008-of-00009.bin",
511
- "transformer.h.41.attn.bias": "pytorch_model-00008-of-00009.bin",
512
- "transformer.h.41.attn.c_attn.bias": "pytorch_model-00008-of-00009.bin",
513
- "transformer.h.41.attn.c_attn.weight": "pytorch_model-00008-of-00009.bin",
514
- "transformer.h.41.attn.c_proj.bias": "pytorch_model-00008-of-00009.bin",
515
- "transformer.h.41.attn.c_proj.weight": "pytorch_model-00008-of-00009.bin",
516
- "transformer.h.41.attn.masked_bias": "pytorch_model-00008-of-00009.bin",
517
- "transformer.h.41.ln_1.bias": "pytorch_model-00008-of-00009.bin",
518
- "transformer.h.41.ln_1.weight": "pytorch_model-00008-of-00009.bin",
519
- "transformer.h.41.ln_2.bias": "pytorch_model-00008-of-00009.bin",
520
- "transformer.h.41.ln_2.weight": "pytorch_model-00008-of-00009.bin",
521
- "transformer.h.41.mlp.c_fc.bias": "pytorch_model-00008-of-00009.bin",
522
- "transformer.h.41.mlp.c_fc.weight": "pytorch_model-00008-of-00009.bin",
523
- "transformer.h.41.mlp.c_proj.bias": "pytorch_model-00008-of-00009.bin",
524
- "transformer.h.41.mlp.c_proj.weight": "pytorch_model-00008-of-00009.bin",
525
- "transformer.h.42.attn.bias": "pytorch_model-00008-of-00009.bin",
526
- "transformer.h.42.attn.c_attn.bias": "pytorch_model-00009-of-00009.bin",
527
- "transformer.h.42.attn.c_attn.weight": "pytorch_model-00009-of-00009.bin",
528
- "transformer.h.42.attn.c_proj.bias": "pytorch_model-00009-of-00009.bin",
529
- "transformer.h.42.attn.c_proj.weight": "pytorch_model-00009-of-00009.bin",
530
- "transformer.h.42.attn.masked_bias": "pytorch_model-00008-of-00009.bin",
531
- "transformer.h.42.ln_1.bias": "pytorch_model-00008-of-00009.bin",
532
- "transformer.h.42.ln_1.weight": "pytorch_model-00008-of-00009.bin",
533
- "transformer.h.42.ln_2.bias": "pytorch_model-00009-of-00009.bin",
534
- "transformer.h.42.ln_2.weight": "pytorch_model-00009-of-00009.bin",
535
- "transformer.h.42.mlp.c_fc.bias": "pytorch_model-00009-of-00009.bin",
536
- "transformer.h.42.mlp.c_fc.weight": "pytorch_model-00009-of-00009.bin",
537
- "transformer.h.42.mlp.c_proj.bias": "pytorch_model-00009-of-00009.bin",
538
- "transformer.h.42.mlp.c_proj.weight": "pytorch_model-00009-of-00009.bin",
539
- "transformer.h.43.attn.bias": "pytorch_model-00009-of-00009.bin",
540
- "transformer.h.43.attn.c_attn.bias": "pytorch_model-00009-of-00009.bin",
541
- "transformer.h.43.attn.c_attn.weight": "pytorch_model-00009-of-00009.bin",
542
- "transformer.h.43.attn.c_proj.bias": "pytorch_model-00009-of-00009.bin",
543
- "transformer.h.43.attn.c_proj.weight": "pytorch_model-00009-of-00009.bin",
544
- "transformer.h.43.attn.masked_bias": "pytorch_model-00009-of-00009.bin",
545
- "transformer.h.43.ln_1.bias": "pytorch_model-00009-of-00009.bin",
546
- "transformer.h.43.ln_1.weight": "pytorch_model-00009-of-00009.bin",
547
- "transformer.h.43.ln_2.bias": "pytorch_model-00009-of-00009.bin",
548
- "transformer.h.43.ln_2.weight": "pytorch_model-00009-of-00009.bin",
549
- "transformer.h.43.mlp.c_fc.bias": "pytorch_model-00009-of-00009.bin",
550
- "transformer.h.43.mlp.c_fc.weight": "pytorch_model-00009-of-00009.bin",
551
- "transformer.h.43.mlp.c_proj.bias": "pytorch_model-00009-of-00009.bin",
552
- "transformer.h.43.mlp.c_proj.weight": "pytorch_model-00009-of-00009.bin",
553
- "transformer.h.5.attn.bias": "pytorch_model-00002-of-00009.bin",
554
- "transformer.h.5.attn.c_attn.bias": "pytorch_model-00002-of-00009.bin",
555
- "transformer.h.5.attn.c_attn.weight": "pytorch_model-00002-of-00009.bin",
556
- "transformer.h.5.attn.c_proj.bias": "pytorch_model-00002-of-00009.bin",
557
- "transformer.h.5.attn.c_proj.weight": "pytorch_model-00002-of-00009.bin",
558
- "transformer.h.5.attn.masked_bias": "pytorch_model-00002-of-00009.bin",
559
- "transformer.h.5.ln_1.bias": "pytorch_model-00002-of-00009.bin",
560
- "transformer.h.5.ln_1.weight": "pytorch_model-00002-of-00009.bin",
561
- "transformer.h.5.ln_2.bias": "pytorch_model-00002-of-00009.bin",
562
- "transformer.h.5.ln_2.weight": "pytorch_model-00002-of-00009.bin",
563
- "transformer.h.5.mlp.c_fc.bias": "pytorch_model-00002-of-00009.bin",
564
- "transformer.h.5.mlp.c_fc.weight": "pytorch_model-00002-of-00009.bin",
565
- "transformer.h.5.mlp.c_proj.bias": "pytorch_model-00002-of-00009.bin",
566
- "transformer.h.5.mlp.c_proj.weight": "pytorch_model-00002-of-00009.bin",
567
- "transformer.h.6.attn.bias": "pytorch_model-00002-of-00009.bin",
568
- "transformer.h.6.attn.c_attn.bias": "pytorch_model-00002-of-00009.bin",
569
- "transformer.h.6.attn.c_attn.weight": "pytorch_model-00002-of-00009.bin",
570
- "transformer.h.6.attn.c_proj.bias": "pytorch_model-00002-of-00009.bin",
571
- "transformer.h.6.attn.c_proj.weight": "pytorch_model-00002-of-00009.bin",
572
- "transformer.h.6.attn.masked_bias": "pytorch_model-00002-of-00009.bin",
573
- "transformer.h.6.ln_1.bias": "pytorch_model-00002-of-00009.bin",
574
- "transformer.h.6.ln_1.weight": "pytorch_model-00002-of-00009.bin",
575
- "transformer.h.6.ln_2.bias": "pytorch_model-00002-of-00009.bin",
576
- "transformer.h.6.ln_2.weight": "pytorch_model-00002-of-00009.bin",
577
- "transformer.h.6.mlp.c_fc.bias": "pytorch_model-00002-of-00009.bin",
578
- "transformer.h.6.mlp.c_fc.weight": "pytorch_model-00002-of-00009.bin",
579
- "transformer.h.6.mlp.c_proj.bias": "pytorch_model-00002-of-00009.bin",
580
- "transformer.h.6.mlp.c_proj.weight": "pytorch_model-00002-of-00009.bin",
581
- "transformer.h.7.attn.bias": "pytorch_model-00002-of-00009.bin",
582
- "transformer.h.7.attn.c_attn.bias": "pytorch_model-00002-of-00009.bin",
583
- "transformer.h.7.attn.c_attn.weight": "pytorch_model-00002-of-00009.bin",
584
- "transformer.h.7.attn.c_proj.bias": "pytorch_model-00002-of-00009.bin",
585
- "transformer.h.7.attn.c_proj.weight": "pytorch_model-00002-of-00009.bin",
586
- "transformer.h.7.attn.masked_bias": "pytorch_model-00002-of-00009.bin",
587
- "transformer.h.7.ln_1.bias": "pytorch_model-00002-of-00009.bin",
588
- "transformer.h.7.ln_1.weight": "pytorch_model-00002-of-00009.bin",
589
- "transformer.h.7.ln_2.bias": "pytorch_model-00002-of-00009.bin",
590
- "transformer.h.7.ln_2.weight": "pytorch_model-00002-of-00009.bin",
591
- "transformer.h.7.mlp.c_fc.bias": "pytorch_model-00002-of-00009.bin",
592
- "transformer.h.7.mlp.c_fc.weight": "pytorch_model-00002-of-00009.bin",
593
- "transformer.h.7.mlp.c_proj.bias": "pytorch_model-00002-of-00009.bin",
594
- "transformer.h.7.mlp.c_proj.weight": "pytorch_model-00002-of-00009.bin",
595
- "transformer.h.8.attn.bias": "pytorch_model-00002-of-00009.bin",
596
- "transformer.h.8.attn.c_attn.bias": "pytorch_model-00002-of-00009.bin",
597
- "transformer.h.8.attn.c_attn.weight": "pytorch_model-00002-of-00009.bin",
598
- "transformer.h.8.attn.c_proj.bias": "pytorch_model-00002-of-00009.bin",
599
- "transformer.h.8.attn.c_proj.weight": "pytorch_model-00002-of-00009.bin",
600
- "transformer.h.8.attn.masked_bias": "pytorch_model-00002-of-00009.bin",
601
- "transformer.h.8.ln_1.bias": "pytorch_model-00002-of-00009.bin",
602
- "transformer.h.8.ln_1.weight": "pytorch_model-00002-of-00009.bin",
603
- "transformer.h.8.ln_2.bias": "pytorch_model-00002-of-00009.bin",
604
- "transformer.h.8.ln_2.weight": "pytorch_model-00002-of-00009.bin",
605
- "transformer.h.8.mlp.c_fc.bias": "pytorch_model-00002-of-00009.bin",
606
- "transformer.h.8.mlp.c_fc.weight": "pytorch_model-00002-of-00009.bin",
607
- "transformer.h.8.mlp.c_proj.bias": "pytorch_model-00002-of-00009.bin",
608
- "transformer.h.8.mlp.c_proj.weight": "pytorch_model-00002-of-00009.bin",
609
- "transformer.h.9.attn.bias": "pytorch_model-00002-of-00009.bin",
610
- "transformer.h.9.attn.c_attn.bias": "pytorch_model-00002-of-00009.bin",
611
- "transformer.h.9.attn.c_attn.weight": "pytorch_model-00002-of-00009.bin",
612
- "transformer.h.9.attn.c_proj.bias": "pytorch_model-00002-of-00009.bin",
613
- "transformer.h.9.attn.c_proj.weight": "pytorch_model-00002-of-00009.bin",
614
- "transformer.h.9.attn.masked_bias": "pytorch_model-00002-of-00009.bin",
615
- "transformer.h.9.ln_1.bias": "pytorch_model-00002-of-00009.bin",
616
- "transformer.h.9.ln_1.weight": "pytorch_model-00002-of-00009.bin",
617
- "transformer.h.9.ln_2.bias": "pytorch_model-00002-of-00009.bin",
618
- "transformer.h.9.ln_2.weight": "pytorch_model-00002-of-00009.bin",
619
- "transformer.h.9.mlp.c_fc.bias": "pytorch_model-00002-of-00009.bin",
620
- "transformer.h.9.mlp.c_fc.weight": "pytorch_model-00002-of-00009.bin",
621
- "transformer.h.9.mlp.c_proj.bias": "pytorch_model-00002-of-00009.bin",
622
- "transformer.h.9.mlp.c_proj.weight": "pytorch_model-00002-of-00009.bin",
623
- "transformer.ln_f.bias": "pytorch_model-00009-of-00009.bin",
624
- "transformer.ln_f.weight": "pytorch_model-00009-of-00009.bin",
625
- "transformer.wpe.weight": "pytorch_model-00001-of-00009.bin",
626
- "transformer.wte.weight": "pytorch_model-00001-of-00009.bin"
627
  }
628
  }
 
1
  {
2
  "metadata": {
3
+ "total_size": 40512454656
4
  },
5
  "weight_map": {
6
+ "lm_head.weight": "pytorch_model-00001-of-00005.bin",
7
+ "transformer.h.0.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
8
+ "transformer.h.0.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
9
+ "transformer.h.0.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
10
+ "transformer.h.0.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
11
+ "transformer.h.0.ln_1.bias": "pytorch_model-00001-of-00005.bin",
12
+ "transformer.h.0.ln_1.weight": "pytorch_model-00001-of-00005.bin",
13
+ "transformer.h.0.ln_2.bias": "pytorch_model-00001-of-00005.bin",
14
+ "transformer.h.0.ln_2.weight": "pytorch_model-00001-of-00005.bin",
15
+ "transformer.h.0.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
16
+ "transformer.h.0.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
17
+ "transformer.h.0.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
18
+ "transformer.h.0.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
19
+ "transformer.h.1.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
20
+ "transformer.h.1.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
21
+ "transformer.h.1.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
22
+ "transformer.h.1.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
23
+ "transformer.h.1.ln_1.bias": "pytorch_model-00001-of-00005.bin",
24
+ "transformer.h.1.ln_1.weight": "pytorch_model-00001-of-00005.bin",
25
+ "transformer.h.1.ln_2.bias": "pytorch_model-00001-of-00005.bin",
26
+ "transformer.h.1.ln_2.weight": "pytorch_model-00001-of-00005.bin",
27
+ "transformer.h.1.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
28
+ "transformer.h.1.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
29
+ "transformer.h.1.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
30
+ "transformer.h.1.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
31
+ "transformer.h.10.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
32
+ "transformer.h.10.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
33
+ "transformer.h.10.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
34
+ "transformer.h.10.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
35
+ "transformer.h.10.ln_1.bias": "pytorch_model-00001-of-00005.bin",
36
+ "transformer.h.10.ln_1.weight": "pytorch_model-00001-of-00005.bin",
37
+ "transformer.h.10.ln_2.bias": "pytorch_model-00002-of-00005.bin",
38
+ "transformer.h.10.ln_2.weight": "pytorch_model-00002-of-00005.bin",
39
+ "transformer.h.10.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
40
+ "transformer.h.10.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
41
+ "transformer.h.10.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
42
+ "transformer.h.10.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
43
+ "transformer.h.11.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
44
+ "transformer.h.11.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
45
+ "transformer.h.11.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
46
+ "transformer.h.11.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
47
+ "transformer.h.11.ln_1.bias": "pytorch_model-00002-of-00005.bin",
48
+ "transformer.h.11.ln_1.weight": "pytorch_model-00002-of-00005.bin",
49
+ "transformer.h.11.ln_2.bias": "pytorch_model-00002-of-00005.bin",
50
+ "transformer.h.11.ln_2.weight": "pytorch_model-00002-of-00005.bin",
51
+ "transformer.h.11.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
52
+ "transformer.h.11.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
53
+ "transformer.h.11.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
54
+ "transformer.h.11.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
55
+ "transformer.h.12.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
56
+ "transformer.h.12.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
57
+ "transformer.h.12.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
58
+ "transformer.h.12.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
59
+ "transformer.h.12.ln_1.bias": "pytorch_model-00002-of-00005.bin",
60
+ "transformer.h.12.ln_1.weight": "pytorch_model-00002-of-00005.bin",
61
+ "transformer.h.12.ln_2.bias": "pytorch_model-00002-of-00005.bin",
62
+ "transformer.h.12.ln_2.weight": "pytorch_model-00002-of-00005.bin",
63
+ "transformer.h.12.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
64
+ "transformer.h.12.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
65
+ "transformer.h.12.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
66
+ "transformer.h.12.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
67
+ "transformer.h.13.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
68
+ "transformer.h.13.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
69
+ "transformer.h.13.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
70
+ "transformer.h.13.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
71
+ "transformer.h.13.ln_1.bias": "pytorch_model-00002-of-00005.bin",
72
+ "transformer.h.13.ln_1.weight": "pytorch_model-00002-of-00005.bin",
73
+ "transformer.h.13.ln_2.bias": "pytorch_model-00002-of-00005.bin",
74
+ "transformer.h.13.ln_2.weight": "pytorch_model-00002-of-00005.bin",
75
+ "transformer.h.13.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
76
+ "transformer.h.13.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
77
+ "transformer.h.13.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
78
+ "transformer.h.13.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
79
+ "transformer.h.14.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
80
+ "transformer.h.14.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
81
+ "transformer.h.14.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
82
+ "transformer.h.14.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
83
+ "transformer.h.14.ln_1.bias": "pytorch_model-00002-of-00005.bin",
84
+ "transformer.h.14.ln_1.weight": "pytorch_model-00002-of-00005.bin",
85
+ "transformer.h.14.ln_2.bias": "pytorch_model-00002-of-00005.bin",
86
+ "transformer.h.14.ln_2.weight": "pytorch_model-00002-of-00005.bin",
87
+ "transformer.h.14.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
88
+ "transformer.h.14.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
89
+ "transformer.h.14.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
90
+ "transformer.h.14.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
91
+ "transformer.h.15.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
92
+ "transformer.h.15.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
93
+ "transformer.h.15.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
94
+ "transformer.h.15.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
95
+ "transformer.h.15.ln_1.bias": "pytorch_model-00002-of-00005.bin",
96
+ "transformer.h.15.ln_1.weight": "pytorch_model-00002-of-00005.bin",
97
+ "transformer.h.15.ln_2.bias": "pytorch_model-00002-of-00005.bin",
98
+ "transformer.h.15.ln_2.weight": "pytorch_model-00002-of-00005.bin",
99
+ "transformer.h.15.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
100
+ "transformer.h.15.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
101
+ "transformer.h.15.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
102
+ "transformer.h.15.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
103
+ "transformer.h.16.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
104
+ "transformer.h.16.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
105
+ "transformer.h.16.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
106
+ "transformer.h.16.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
107
+ "transformer.h.16.ln_1.bias": "pytorch_model-00002-of-00005.bin",
108
+ "transformer.h.16.ln_1.weight": "pytorch_model-00002-of-00005.bin",
109
+ "transformer.h.16.ln_2.bias": "pytorch_model-00002-of-00005.bin",
110
+ "transformer.h.16.ln_2.weight": "pytorch_model-00002-of-00005.bin",
111
+ "transformer.h.16.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
112
+ "transformer.h.16.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
113
+ "transformer.h.16.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
114
+ "transformer.h.16.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
115
+ "transformer.h.17.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
116
+ "transformer.h.17.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
117
+ "transformer.h.17.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
118
+ "transformer.h.17.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
119
+ "transformer.h.17.ln_1.bias": "pytorch_model-00002-of-00005.bin",
120
+ "transformer.h.17.ln_1.weight": "pytorch_model-00002-of-00005.bin",
121
+ "transformer.h.17.ln_2.bias": "pytorch_model-00002-of-00005.bin",
122
+ "transformer.h.17.ln_2.weight": "pytorch_model-00002-of-00005.bin",
123
+ "transformer.h.17.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
124
+ "transformer.h.17.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
125
+ "transformer.h.17.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
126
+ "transformer.h.17.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
127
+ "transformer.h.18.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
128
+ "transformer.h.18.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
129
+ "transformer.h.18.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
130
+ "transformer.h.18.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
131
+ "transformer.h.18.ln_1.bias": "pytorch_model-00002-of-00005.bin",
132
+ "transformer.h.18.ln_1.weight": "pytorch_model-00002-of-00005.bin",
133
+ "transformer.h.18.ln_2.bias": "pytorch_model-00002-of-00005.bin",
134
+ "transformer.h.18.ln_2.weight": "pytorch_model-00002-of-00005.bin",
135
+ "transformer.h.18.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
136
+ "transformer.h.18.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
137
+ "transformer.h.18.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
138
+ "transformer.h.18.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
139
+ "transformer.h.19.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
140
+ "transformer.h.19.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
141
+ "transformer.h.19.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
142
+ "transformer.h.19.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
143
+ "transformer.h.19.ln_1.bias": "pytorch_model-00002-of-00005.bin",
144
+ "transformer.h.19.ln_1.weight": "pytorch_model-00002-of-00005.bin",
145
+ "transformer.h.19.ln_2.bias": "pytorch_model-00002-of-00005.bin",
146
+ "transformer.h.19.ln_2.weight": "pytorch_model-00002-of-00005.bin",
147
+ "transformer.h.19.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
148
+ "transformer.h.19.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
149
+ "transformer.h.19.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
150
+ "transformer.h.19.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
151
+ "transformer.h.2.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
152
+ "transformer.h.2.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
153
+ "transformer.h.2.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
154
+ "transformer.h.2.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
155
+ "transformer.h.2.ln_1.bias": "pytorch_model-00001-of-00005.bin",
156
+ "transformer.h.2.ln_1.weight": "pytorch_model-00001-of-00005.bin",
157
+ "transformer.h.2.ln_2.bias": "pytorch_model-00001-of-00005.bin",
158
+ "transformer.h.2.ln_2.weight": "pytorch_model-00001-of-00005.bin",
159
+ "transformer.h.2.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
160
+ "transformer.h.2.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
161
+ "transformer.h.2.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
162
+ "transformer.h.2.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
163
+ "transformer.h.20.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
164
+ "transformer.h.20.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
165
+ "transformer.h.20.attn.c_proj.bias": "pytorch_model-00002-of-00005.bin",
166
+ "transformer.h.20.attn.c_proj.weight": "pytorch_model-00002-of-00005.bin",
167
+ "transformer.h.20.ln_1.bias": "pytorch_model-00002-of-00005.bin",
168
+ "transformer.h.20.ln_1.weight": "pytorch_model-00002-of-00005.bin",
169
+ "transformer.h.20.ln_2.bias": "pytorch_model-00002-of-00005.bin",
170
+ "transformer.h.20.ln_2.weight": "pytorch_model-00002-of-00005.bin",
171
+ "transformer.h.20.mlp.c_fc.bias": "pytorch_model-00002-of-00005.bin",
172
+ "transformer.h.20.mlp.c_fc.weight": "pytorch_model-00002-of-00005.bin",
173
+ "transformer.h.20.mlp.c_proj.bias": "pytorch_model-00002-of-00005.bin",
174
+ "transformer.h.20.mlp.c_proj.weight": "pytorch_model-00002-of-00005.bin",
175
+ "transformer.h.21.attn.c_attn.bias": "pytorch_model-00002-of-00005.bin",
176
+ "transformer.h.21.attn.c_attn.weight": "pytorch_model-00002-of-00005.bin",
177
+ "transformer.h.21.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
178
+ "transformer.h.21.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
179
+ "transformer.h.21.ln_1.bias": "pytorch_model-00002-of-00005.bin",
180
+ "transformer.h.21.ln_1.weight": "pytorch_model-00002-of-00005.bin",
181
+ "transformer.h.21.ln_2.bias": "pytorch_model-00003-of-00005.bin",
182
+ "transformer.h.21.ln_2.weight": "pytorch_model-00003-of-00005.bin",
183
+ "transformer.h.21.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
184
+ "transformer.h.21.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
185
+ "transformer.h.21.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
186
+ "transformer.h.21.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
187
+ "transformer.h.22.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
188
+ "transformer.h.22.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
189
+ "transformer.h.22.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
190
+ "transformer.h.22.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
191
+ "transformer.h.22.ln_1.bias": "pytorch_model-00003-of-00005.bin",
192
+ "transformer.h.22.ln_1.weight": "pytorch_model-00003-of-00005.bin",
193
+ "transformer.h.22.ln_2.bias": "pytorch_model-00003-of-00005.bin",
194
+ "transformer.h.22.ln_2.weight": "pytorch_model-00003-of-00005.bin",
195
+ "transformer.h.22.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
196
+ "transformer.h.22.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
197
+ "transformer.h.22.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
198
+ "transformer.h.22.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
199
+ "transformer.h.23.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
200
+ "transformer.h.23.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
201
+ "transformer.h.23.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
202
+ "transformer.h.23.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
203
+ "transformer.h.23.ln_1.bias": "pytorch_model-00003-of-00005.bin",
204
+ "transformer.h.23.ln_1.weight": "pytorch_model-00003-of-00005.bin",
205
+ "transformer.h.23.ln_2.bias": "pytorch_model-00003-of-00005.bin",
206
+ "transformer.h.23.ln_2.weight": "pytorch_model-00003-of-00005.bin",
207
+ "transformer.h.23.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
208
+ "transformer.h.23.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
209
+ "transformer.h.23.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
210
+ "transformer.h.23.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
211
+ "transformer.h.24.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
212
+ "transformer.h.24.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
213
+ "transformer.h.24.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
214
+ "transformer.h.24.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
215
+ "transformer.h.24.ln_1.bias": "pytorch_model-00003-of-00005.bin",
216
+ "transformer.h.24.ln_1.weight": "pytorch_model-00003-of-00005.bin",
217
+ "transformer.h.24.ln_2.bias": "pytorch_model-00003-of-00005.bin",
218
+ "transformer.h.24.ln_2.weight": "pytorch_model-00003-of-00005.bin",
219
+ "transformer.h.24.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
220
+ "transformer.h.24.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
221
+ "transformer.h.24.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
222
+ "transformer.h.24.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
223
+ "transformer.h.25.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
224
+ "transformer.h.25.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
225
+ "transformer.h.25.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
226
+ "transformer.h.25.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
227
+ "transformer.h.25.ln_1.bias": "pytorch_model-00003-of-00005.bin",
228
+ "transformer.h.25.ln_1.weight": "pytorch_model-00003-of-00005.bin",
229
+ "transformer.h.25.ln_2.bias": "pytorch_model-00003-of-00005.bin",
230
+ "transformer.h.25.ln_2.weight": "pytorch_model-00003-of-00005.bin",
231
+ "transformer.h.25.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
232
+ "transformer.h.25.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
233
+ "transformer.h.25.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
234
+ "transformer.h.25.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
235
+ "transformer.h.26.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
236
+ "transformer.h.26.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
237
+ "transformer.h.26.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
238
+ "transformer.h.26.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
239
+ "transformer.h.26.ln_1.bias": "pytorch_model-00003-of-00005.bin",
240
+ "transformer.h.26.ln_1.weight": "pytorch_model-00003-of-00005.bin",
241
+ "transformer.h.26.ln_2.bias": "pytorch_model-00003-of-00005.bin",
242
+ "transformer.h.26.ln_2.weight": "pytorch_model-00003-of-00005.bin",
243
+ "transformer.h.26.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
244
+ "transformer.h.26.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
245
+ "transformer.h.26.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
246
+ "transformer.h.26.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
247
+ "transformer.h.27.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
248
+ "transformer.h.27.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
249
+ "transformer.h.27.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
250
+ "transformer.h.27.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
251
+ "transformer.h.27.ln_1.bias": "pytorch_model-00003-of-00005.bin",
252
+ "transformer.h.27.ln_1.weight": "pytorch_model-00003-of-00005.bin",
253
+ "transformer.h.27.ln_2.bias": "pytorch_model-00003-of-00005.bin",
254
+ "transformer.h.27.ln_2.weight": "pytorch_model-00003-of-00005.bin",
255
+ "transformer.h.27.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
256
+ "transformer.h.27.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
257
+ "transformer.h.27.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
258
+ "transformer.h.27.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
259
+ "transformer.h.28.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
260
+ "transformer.h.28.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
261
+ "transformer.h.28.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
262
+ "transformer.h.28.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
263
+ "transformer.h.28.ln_1.bias": "pytorch_model-00003-of-00005.bin",
264
+ "transformer.h.28.ln_1.weight": "pytorch_model-00003-of-00005.bin",
265
+ "transformer.h.28.ln_2.bias": "pytorch_model-00003-of-00005.bin",
266
+ "transformer.h.28.ln_2.weight": "pytorch_model-00003-of-00005.bin",
267
+ "transformer.h.28.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
268
+ "transformer.h.28.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
269
+ "transformer.h.28.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
270
+ "transformer.h.28.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
271
+ "transformer.h.29.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
272
+ "transformer.h.29.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
273
+ "transformer.h.29.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
274
+ "transformer.h.29.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
275
+ "transformer.h.29.ln_1.bias": "pytorch_model-00003-of-00005.bin",
276
+ "transformer.h.29.ln_1.weight": "pytorch_model-00003-of-00005.bin",
277
+ "transformer.h.29.ln_2.bias": "pytorch_model-00003-of-00005.bin",
278
+ "transformer.h.29.ln_2.weight": "pytorch_model-00003-of-00005.bin",
279
+ "transformer.h.29.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
280
+ "transformer.h.29.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
281
+ "transformer.h.29.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
282
+ "transformer.h.29.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
283
+ "transformer.h.3.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
284
+ "transformer.h.3.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
285
+ "transformer.h.3.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
286
+ "transformer.h.3.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
287
+ "transformer.h.3.ln_1.bias": "pytorch_model-00001-of-00005.bin",
288
+ "transformer.h.3.ln_1.weight": "pytorch_model-00001-of-00005.bin",
289
+ "transformer.h.3.ln_2.bias": "pytorch_model-00001-of-00005.bin",
290
+ "transformer.h.3.ln_2.weight": "pytorch_model-00001-of-00005.bin",
291
+ "transformer.h.3.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
292
+ "transformer.h.3.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
293
+ "transformer.h.3.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
294
+ "transformer.h.3.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
295
+ "transformer.h.30.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
296
+ "transformer.h.30.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
297
+ "transformer.h.30.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
298
+ "transformer.h.30.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
299
+ "transformer.h.30.ln_1.bias": "pytorch_model-00003-of-00005.bin",
300
+ "transformer.h.30.ln_1.weight": "pytorch_model-00003-of-00005.bin",
301
+ "transformer.h.30.ln_2.bias": "pytorch_model-00003-of-00005.bin",
302
+ "transformer.h.30.ln_2.weight": "pytorch_model-00003-of-00005.bin",
303
+ "transformer.h.30.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
304
+ "transformer.h.30.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
305
+ "transformer.h.30.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
306
+ "transformer.h.30.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
307
+ "transformer.h.31.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
308
+ "transformer.h.31.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
309
+ "transformer.h.31.attn.c_proj.bias": "pytorch_model-00003-of-00005.bin",
310
+ "transformer.h.31.attn.c_proj.weight": "pytorch_model-00003-of-00005.bin",
311
+ "transformer.h.31.ln_1.bias": "pytorch_model-00003-of-00005.bin",
312
+ "transformer.h.31.ln_1.weight": "pytorch_model-00003-of-00005.bin",
313
+ "transformer.h.31.ln_2.bias": "pytorch_model-00003-of-00005.bin",
314
+ "transformer.h.31.ln_2.weight": "pytorch_model-00003-of-00005.bin",
315
+ "transformer.h.31.mlp.c_fc.bias": "pytorch_model-00003-of-00005.bin",
316
+ "transformer.h.31.mlp.c_fc.weight": "pytorch_model-00003-of-00005.bin",
317
+ "transformer.h.31.mlp.c_proj.bias": "pytorch_model-00003-of-00005.bin",
318
+ "transformer.h.31.mlp.c_proj.weight": "pytorch_model-00003-of-00005.bin",
319
+ "transformer.h.32.attn.c_attn.bias": "pytorch_model-00003-of-00005.bin",
320
+ "transformer.h.32.attn.c_attn.weight": "pytorch_model-00003-of-00005.bin",
321
+ "transformer.h.32.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
322
+ "transformer.h.32.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
323
+ "transformer.h.32.ln_1.bias": "pytorch_model-00003-of-00005.bin",
324
+ "transformer.h.32.ln_1.weight": "pytorch_model-00003-of-00005.bin",
325
+ "transformer.h.32.ln_2.bias": "pytorch_model-00004-of-00005.bin",
326
+ "transformer.h.32.ln_2.weight": "pytorch_model-00004-of-00005.bin",
327
+ "transformer.h.32.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
328
+ "transformer.h.32.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
329
+ "transformer.h.32.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
330
+ "transformer.h.32.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
331
+ "transformer.h.33.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
332
+ "transformer.h.33.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
333
+ "transformer.h.33.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
334
+ "transformer.h.33.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
335
+ "transformer.h.33.ln_1.bias": "pytorch_model-00004-of-00005.bin",
336
+ "transformer.h.33.ln_1.weight": "pytorch_model-00004-of-00005.bin",
337
+ "transformer.h.33.ln_2.bias": "pytorch_model-00004-of-00005.bin",
338
+ "transformer.h.33.ln_2.weight": "pytorch_model-00004-of-00005.bin",
339
+ "transformer.h.33.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
340
+ "transformer.h.33.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
341
+ "transformer.h.33.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
342
+ "transformer.h.33.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
343
+ "transformer.h.34.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
344
+ "transformer.h.34.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
345
+ "transformer.h.34.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
346
+ "transformer.h.34.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
347
+ "transformer.h.34.ln_1.bias": "pytorch_model-00004-of-00005.bin",
348
+ "transformer.h.34.ln_1.weight": "pytorch_model-00004-of-00005.bin",
349
+ "transformer.h.34.ln_2.bias": "pytorch_model-00004-of-00005.bin",
350
+ "transformer.h.34.ln_2.weight": "pytorch_model-00004-of-00005.bin",
351
+ "transformer.h.34.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
352
+ "transformer.h.34.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
353
+ "transformer.h.34.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
354
+ "transformer.h.34.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
355
+ "transformer.h.35.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
356
+ "transformer.h.35.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
357
+ "transformer.h.35.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
358
+ "transformer.h.35.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
359
+ "transformer.h.35.ln_1.bias": "pytorch_model-00004-of-00005.bin",
360
+ "transformer.h.35.ln_1.weight": "pytorch_model-00004-of-00005.bin",
361
+ "transformer.h.35.ln_2.bias": "pytorch_model-00004-of-00005.bin",
362
+ "transformer.h.35.ln_2.weight": "pytorch_model-00004-of-00005.bin",
363
+ "transformer.h.35.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
364
+ "transformer.h.35.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
365
+ "transformer.h.35.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
366
+ "transformer.h.35.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
367
+ "transformer.h.36.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
368
+ "transformer.h.36.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
369
+ "transformer.h.36.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
370
+ "transformer.h.36.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
371
+ "transformer.h.36.ln_1.bias": "pytorch_model-00004-of-00005.bin",
372
+ "transformer.h.36.ln_1.weight": "pytorch_model-00004-of-00005.bin",
373
+ "transformer.h.36.ln_2.bias": "pytorch_model-00004-of-00005.bin",
374
+ "transformer.h.36.ln_2.weight": "pytorch_model-00004-of-00005.bin",
375
+ "transformer.h.36.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
376
+ "transformer.h.36.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
377
+ "transformer.h.36.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
378
+ "transformer.h.36.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
379
+ "transformer.h.37.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
380
+ "transformer.h.37.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
381
+ "transformer.h.37.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
382
+ "transformer.h.37.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
383
+ "transformer.h.37.ln_1.bias": "pytorch_model-00004-of-00005.bin",
384
+ "transformer.h.37.ln_1.weight": "pytorch_model-00004-of-00005.bin",
385
+ "transformer.h.37.ln_2.bias": "pytorch_model-00004-of-00005.bin",
386
+ "transformer.h.37.ln_2.weight": "pytorch_model-00004-of-00005.bin",
387
+ "transformer.h.37.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
388
+ "transformer.h.37.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
389
+ "transformer.h.37.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
390
+ "transformer.h.37.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
391
+ "transformer.h.38.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
392
+ "transformer.h.38.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
393
+ "transformer.h.38.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
394
+ "transformer.h.38.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
395
+ "transformer.h.38.ln_1.bias": "pytorch_model-00004-of-00005.bin",
396
+ "transformer.h.38.ln_1.weight": "pytorch_model-00004-of-00005.bin",
397
+ "transformer.h.38.ln_2.bias": "pytorch_model-00004-of-00005.bin",
398
+ "transformer.h.38.ln_2.weight": "pytorch_model-00004-of-00005.bin",
399
+ "transformer.h.38.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
400
+ "transformer.h.38.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
401
+ "transformer.h.38.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
402
+ "transformer.h.38.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
403
+ "transformer.h.39.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
404
+ "transformer.h.39.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
405
+ "transformer.h.39.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
406
+ "transformer.h.39.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
407
+ "transformer.h.39.ln_1.bias": "pytorch_model-00004-of-00005.bin",
408
+ "transformer.h.39.ln_1.weight": "pytorch_model-00004-of-00005.bin",
409
+ "transformer.h.39.ln_2.bias": "pytorch_model-00004-of-00005.bin",
410
+ "transformer.h.39.ln_2.weight": "pytorch_model-00004-of-00005.bin",
411
+ "transformer.h.39.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
412
+ "transformer.h.39.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
413
+ "transformer.h.39.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
414
+ "transformer.h.39.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
415
+ "transformer.h.4.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
416
+ "transformer.h.4.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
417
+ "transformer.h.4.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
418
+ "transformer.h.4.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
419
+ "transformer.h.4.ln_1.bias": "pytorch_model-00001-of-00005.bin",
420
+ "transformer.h.4.ln_1.weight": "pytorch_model-00001-of-00005.bin",
421
+ "transformer.h.4.ln_2.bias": "pytorch_model-00001-of-00005.bin",
422
+ "transformer.h.4.ln_2.weight": "pytorch_model-00001-of-00005.bin",
423
+ "transformer.h.4.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
424
+ "transformer.h.4.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
425
+ "transformer.h.4.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
426
+ "transformer.h.4.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
427
+ "transformer.h.40.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
428
+ "transformer.h.40.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
429
+ "transformer.h.40.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
430
+ "transformer.h.40.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
431
+ "transformer.h.40.ln_1.bias": "pytorch_model-00004-of-00005.bin",
432
+ "transformer.h.40.ln_1.weight": "pytorch_model-00004-of-00005.bin",
433
+ "transformer.h.40.ln_2.bias": "pytorch_model-00004-of-00005.bin",
434
+ "transformer.h.40.ln_2.weight": "pytorch_model-00004-of-00005.bin",
435
+ "transformer.h.40.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
436
+ "transformer.h.40.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
437
+ "transformer.h.40.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
438
+ "transformer.h.40.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
439
+ "transformer.h.41.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
440
+ "transformer.h.41.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
441
+ "transformer.h.41.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
442
+ "transformer.h.41.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
443
+ "transformer.h.41.ln_1.bias": "pytorch_model-00004-of-00005.bin",
444
+ "transformer.h.41.ln_1.weight": "pytorch_model-00004-of-00005.bin",
445
+ "transformer.h.41.ln_2.bias": "pytorch_model-00004-of-00005.bin",
446
+ "transformer.h.41.ln_2.weight": "pytorch_model-00004-of-00005.bin",
447
+ "transformer.h.41.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
448
+ "transformer.h.41.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
449
+ "transformer.h.41.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
450
+ "transformer.h.41.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
451
+ "transformer.h.42.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
452
+ "transformer.h.42.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
453
+ "transformer.h.42.attn.c_proj.bias": "pytorch_model-00004-of-00005.bin",
454
+ "transformer.h.42.attn.c_proj.weight": "pytorch_model-00004-of-00005.bin",
455
+ "transformer.h.42.ln_1.bias": "pytorch_model-00004-of-00005.bin",
456
+ "transformer.h.42.ln_1.weight": "pytorch_model-00004-of-00005.bin",
457
+ "transformer.h.42.ln_2.bias": "pytorch_model-00004-of-00005.bin",
458
+ "transformer.h.42.ln_2.weight": "pytorch_model-00004-of-00005.bin",
459
+ "transformer.h.42.mlp.c_fc.bias": "pytorch_model-00004-of-00005.bin",
460
+ "transformer.h.42.mlp.c_fc.weight": "pytorch_model-00004-of-00005.bin",
461
+ "transformer.h.42.mlp.c_proj.bias": "pytorch_model-00004-of-00005.bin",
462
+ "transformer.h.42.mlp.c_proj.weight": "pytorch_model-00004-of-00005.bin",
463
+ "transformer.h.43.attn.c_attn.bias": "pytorch_model-00004-of-00005.bin",
464
+ "transformer.h.43.attn.c_attn.weight": "pytorch_model-00004-of-00005.bin",
465
+ "transformer.h.43.attn.c_proj.bias": "pytorch_model-00005-of-00005.bin",
466
+ "transformer.h.43.attn.c_proj.weight": "pytorch_model-00005-of-00005.bin",
467
+ "transformer.h.43.ln_1.bias": "pytorch_model-00004-of-00005.bin",
468
+ "transformer.h.43.ln_1.weight": "pytorch_model-00004-of-00005.bin",
469
+ "transformer.h.43.ln_2.bias": "pytorch_model-00005-of-00005.bin",
470
+ "transformer.h.43.ln_2.weight": "pytorch_model-00005-of-00005.bin",
471
+ "transformer.h.43.mlp.c_fc.bias": "pytorch_model-00005-of-00005.bin",
472
+ "transformer.h.43.mlp.c_fc.weight": "pytorch_model-00005-of-00005.bin",
473
+ "transformer.h.43.mlp.c_proj.bias": "pytorch_model-00005-of-00005.bin",
474
+ "transformer.h.43.mlp.c_proj.weight": "pytorch_model-00005-of-00005.bin",
475
+ "transformer.h.5.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
476
+ "transformer.h.5.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
477
+ "transformer.h.5.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
478
+ "transformer.h.5.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
479
+ "transformer.h.5.ln_1.bias": "pytorch_model-00001-of-00005.bin",
480
+ "transformer.h.5.ln_1.weight": "pytorch_model-00001-of-00005.bin",
481
+ "transformer.h.5.ln_2.bias": "pytorch_model-00001-of-00005.bin",
482
+ "transformer.h.5.ln_2.weight": "pytorch_model-00001-of-00005.bin",
483
+ "transformer.h.5.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
484
+ "transformer.h.5.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
485
+ "transformer.h.5.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
486
+ "transformer.h.5.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
487
+ "transformer.h.6.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
488
+ "transformer.h.6.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
489
+ "transformer.h.6.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
490
+ "transformer.h.6.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
491
+ "transformer.h.6.ln_1.bias": "pytorch_model-00001-of-00005.bin",
492
+ "transformer.h.6.ln_1.weight": "pytorch_model-00001-of-00005.bin",
493
+ "transformer.h.6.ln_2.bias": "pytorch_model-00001-of-00005.bin",
494
+ "transformer.h.6.ln_2.weight": "pytorch_model-00001-of-00005.bin",
495
+ "transformer.h.6.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
496
+ "transformer.h.6.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
497
+ "transformer.h.6.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
498
+ "transformer.h.6.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
499
+ "transformer.h.7.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
500
+ "transformer.h.7.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
501
+ "transformer.h.7.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
502
+ "transformer.h.7.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
503
+ "transformer.h.7.ln_1.bias": "pytorch_model-00001-of-00005.bin",
504
+ "transformer.h.7.ln_1.weight": "pytorch_model-00001-of-00005.bin",
505
+ "transformer.h.7.ln_2.bias": "pytorch_model-00001-of-00005.bin",
506
+ "transformer.h.7.ln_2.weight": "pytorch_model-00001-of-00005.bin",
507
+ "transformer.h.7.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
508
+ "transformer.h.7.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
509
+ "transformer.h.7.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
510
+ "transformer.h.7.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
511
+ "transformer.h.8.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
512
+ "transformer.h.8.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
513
+ "transformer.h.8.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
514
+ "transformer.h.8.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
515
+ "transformer.h.8.ln_1.bias": "pytorch_model-00001-of-00005.bin",
516
+ "transformer.h.8.ln_1.weight": "pytorch_model-00001-of-00005.bin",
517
+ "transformer.h.8.ln_2.bias": "pytorch_model-00001-of-00005.bin",
518
+ "transformer.h.8.ln_2.weight": "pytorch_model-00001-of-00005.bin",
519
+ "transformer.h.8.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
520
+ "transformer.h.8.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
521
+ "transformer.h.8.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
522
+ "transformer.h.8.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
523
+ "transformer.h.9.attn.c_attn.bias": "pytorch_model-00001-of-00005.bin",
524
+ "transformer.h.9.attn.c_attn.weight": "pytorch_model-00001-of-00005.bin",
525
+ "transformer.h.9.attn.c_proj.bias": "pytorch_model-00001-of-00005.bin",
526
+ "transformer.h.9.attn.c_proj.weight": "pytorch_model-00001-of-00005.bin",
527
+ "transformer.h.9.ln_1.bias": "pytorch_model-00001-of-00005.bin",
528
+ "transformer.h.9.ln_1.weight": "pytorch_model-00001-of-00005.bin",
529
+ "transformer.h.9.ln_2.bias": "pytorch_model-00001-of-00005.bin",
530
+ "transformer.h.9.ln_2.weight": "pytorch_model-00001-of-00005.bin",
531
+ "transformer.h.9.mlp.c_fc.bias": "pytorch_model-00001-of-00005.bin",
532
+ "transformer.h.9.mlp.c_fc.weight": "pytorch_model-00001-of-00005.bin",
533
+ "transformer.h.9.mlp.c_proj.bias": "pytorch_model-00001-of-00005.bin",
534
+ "transformer.h.9.mlp.c_proj.weight": "pytorch_model-00001-of-00005.bin",
535
+ "transformer.ln_f.bias": "pytorch_model-00005-of-00005.bin",
536
+ "transformer.ln_f.weight": "pytorch_model-00005-of-00005.bin",
537
+ "transformer.wpe.weight": "pytorch_model-00001-of-00005.bin",
538
+ "transformer.wte.weight": "pytorch_model-00001-of-00005.bin"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
539
  }
540
  }