Hugging Face's logo

sambanovasystems
/
SN-13B-8k-Instruct

Feature Extraction
Transformers
PyTorch
gpt2
text-generation-inference
Model card Files Community
2

Upload folder using huggingface_hub

#2
by jayr014 - opened
base: refs/heads/main
←
from: refs/pr/2
Discussion Files changed
+631
-0
Files changed (8) hide show
  1. config.json +42 -0
  2. pytorch_model-00001-of-00006.bin +3 -0
  3. pytorch_model-00002-of-00006.bin +3 -0
  4. pytorch_model-00003-of-00006.bin +3 -0
  5. pytorch_model-00004-of-00006.bin +3 -0
  6. pytorch_model-00005-of-00006.bin +3 -0
  7. pytorch_model-00006-of-00006.bin +3 -0
  8. pytorch_model.bin.index.json +571 -0
config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/scratch1/jayr/gpt13b_itv3.5",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPT2Model"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 50256,
9
+ "embd_pdrop": 0.0,
10
+ "eos_token_id": 50256,
11
+ "gradient_checkpointing": false,
12
+ "initializer_range": 0.02,
13
+ "large_bs_inference": false,
14
+ "layer_norm_epsilon": 1e-05,
15
+ "model_type": "gpt2",
16
+ "n_ctx": 8192,
17
+ "n_embd": 5120,
18
+ "n_head": 40,
19
+ "n_inner": null,
20
+ "n_layer": 40,
21
+ "n_positions": 8192,
22
+ "reorder_and_upcast_attn": false,
23
+ "resid_pdrop": 0.1,
24
+ "return_dict": false,
25
+ "scale_attn_by_inverse_layer_idx": false,
26
+ "scale_attn_weights": true,
27
+ "summary_activation": null,
28
+ "summary_first_dropout": 0.1,
29
+ "summary_proj_to_labels": true,
30
+ "summary_type": "cls_index",
31
+ "summary_use_proj": true,
32
+ "task_specific_params": {
33
+ "text-generation": {
34
+ "do_sample": true,
35
+ "max_length": 50
36
+ }
37
+ },
38
+ "torch_dtype": "float32",
39
+ "transformers_version": "4.18.0",
40
+ "use_cache": false,
41
+ "vocab_size": 50260
42
+ }
pytorch_model-00001-of-00006.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1bfa526df470ca695e792a56cdbd89c41c023768661a763fea6a593ca59277f1
3
+ size 9637826281
pytorch_model-00002-of-00006.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f39b4b18ae8677a75bcf76fef72ad3302a65d83bddbb2a7875dcb052d2d01c30
3
+ size 9699205981
pytorch_model-00003-of-00006.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc656acbae27131d84153e8daada6a446d0716728d1a1c2ad083e0bc0156d738
3
+ size 9766295653
pytorch_model-00004-of-00006.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f38d0df96c2f08a1f31c5bbe31adde0029e9e71519097dd39340204c8140e6a
3
+ size 9699248065
pytorch_model-00005-of-00006.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3cb7ce956c0c1480fbe92c1e43199ca5d2586d5813c030e837c031b22e3e2537
3
+ size 9699206045
pytorch_model-00006-of-00006.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed7ca6af6180c9923d18c69eb68353948455d7546bb5419eaf30d3b1aba1dbe5
3
+ size 5722174405
pytorch_model.bin.index.json ADDED
@@ -0,0 +1,571 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 54223790240
4
+ },
5
+ "weight_map": {
6
+ "h.0.attn.bias": "pytorch_model-00001-of-00006.bin",
7
+ "h.0.attn.c_attn.bias": "pytorch_model-00001-of-00006.bin",
8
+ "h.0.attn.c_attn.weight": "pytorch_model-00001-of-00006.bin",
9
+ "h.0.attn.c_proj.bias": "pytorch_model-00001-of-00006.bin",
10
+ "h.0.attn.c_proj.weight": "pytorch_model-00001-of-00006.bin",
11
+ "h.0.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
12
+ "h.0.ln_1.bias": "pytorch_model-00001-of-00006.bin",
13
+ "h.0.ln_1.weight": "pytorch_model-00001-of-00006.bin",
14
+ "h.0.ln_2.bias": "pytorch_model-00001-of-00006.bin",
15
+ "h.0.ln_2.weight": "pytorch_model-00001-of-00006.bin",
16
+ "h.0.mlp.c_fc.bias": "pytorch_model-00001-of-00006.bin",
17
+ "h.0.mlp.c_fc.weight": "pytorch_model-00001-of-00006.bin",
18
+ "h.0.mlp.c_proj.bias": "pytorch_model-00001-of-00006.bin",
19
+ "h.0.mlp.c_proj.weight": "pytorch_model-00001-of-00006.bin",
20
+ "h.1.attn.bias": "pytorch_model-00001-of-00006.bin",
21
+ "h.1.attn.c_attn.bias": "pytorch_model-00001-of-00006.bin",
22
+ "h.1.attn.c_attn.weight": "pytorch_model-00001-of-00006.bin",
23
+ "h.1.attn.c_proj.bias": "pytorch_model-00001-of-00006.bin",
24
+ "h.1.attn.c_proj.weight": "pytorch_model-00001-of-00006.bin",
25
+ "h.1.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
26
+ "h.1.ln_1.bias": "pytorch_model-00001-of-00006.bin",
27
+ "h.1.ln_1.weight": "pytorch_model-00001-of-00006.bin",
28
+ "h.1.ln_2.bias": "pytorch_model-00001-of-00006.bin",
29
+ "h.1.ln_2.weight": "pytorch_model-00001-of-00006.bin",
30
+ "h.1.mlp.c_fc.bias": "pytorch_model-00001-of-00006.bin",
31
+ "h.1.mlp.c_fc.weight": "pytorch_model-00001-of-00006.bin",
32
+ "h.1.mlp.c_proj.bias": "pytorch_model-00001-of-00006.bin",
33
+ "h.1.mlp.c_proj.weight": "pytorch_model-00001-of-00006.bin",
34
+ "h.10.attn.bias": "pytorch_model-00002-of-00006.bin",
35
+ "h.10.attn.c_attn.bias": "pytorch_model-00002-of-00006.bin",
36
+ "h.10.attn.c_attn.weight": "pytorch_model-00002-of-00006.bin",
37
+ "h.10.attn.c_proj.bias": "pytorch_model-00002-of-00006.bin",
38
+ "h.10.attn.c_proj.weight": "pytorch_model-00002-of-00006.bin",
39
+ "h.10.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
40
+ "h.10.ln_1.bias": "pytorch_model-00002-of-00006.bin",
41
+ "h.10.ln_1.weight": "pytorch_model-00002-of-00006.bin",
42
+ "h.10.ln_2.bias": "pytorch_model-00002-of-00006.bin",
43
+ "h.10.ln_2.weight": "pytorch_model-00002-of-00006.bin",
44
+ "h.10.mlp.c_fc.bias": "pytorch_model-00002-of-00006.bin",
45
+ "h.10.mlp.c_fc.weight": "pytorch_model-00002-of-00006.bin",
46
+ "h.10.mlp.c_proj.bias": "pytorch_model-00002-of-00006.bin",
47
+ "h.10.mlp.c_proj.weight": "pytorch_model-00002-of-00006.bin",
48
+ "h.11.attn.bias": "pytorch_model-00002-of-00006.bin",
49
+ "h.11.attn.c_attn.bias": "pytorch_model-00002-of-00006.bin",
50
+ "h.11.attn.c_attn.weight": "pytorch_model-00002-of-00006.bin",
51
+ "h.11.attn.c_proj.bias": "pytorch_model-00002-of-00006.bin",
52
+ "h.11.attn.c_proj.weight": "pytorch_model-00002-of-00006.bin",
53
+ "h.11.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
54
+ "h.11.ln_1.bias": "pytorch_model-00002-of-00006.bin",
55
+ "h.11.ln_1.weight": "pytorch_model-00002-of-00006.bin",
56
+ "h.11.ln_2.bias": "pytorch_model-00002-of-00006.bin",
57
+ "h.11.ln_2.weight": "pytorch_model-00002-of-00006.bin",
58
+ "h.11.mlp.c_fc.bias": "pytorch_model-00002-of-00006.bin",
59
+ "h.11.mlp.c_fc.weight": "pytorch_model-00002-of-00006.bin",
60
+ "h.11.mlp.c_proj.bias": "pytorch_model-00002-of-00006.bin",
61
+ "h.11.mlp.c_proj.weight": "pytorch_model-00002-of-00006.bin",
62
+ "h.12.attn.bias": "pytorch_model-00002-of-00006.bin",
63
+ "h.12.attn.c_attn.bias": "pytorch_model-00002-of-00006.bin",
64
+ "h.12.attn.c_attn.weight": "pytorch_model-00002-of-00006.bin",
65
+ "h.12.attn.c_proj.bias": "pytorch_model-00002-of-00006.bin",
66
+ "h.12.attn.c_proj.weight": "pytorch_model-00002-of-00006.bin",
67
+ "h.12.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
68
+ "h.12.ln_1.bias": "pytorch_model-00002-of-00006.bin",
69
+ "h.12.ln_1.weight": "pytorch_model-00002-of-00006.bin",
70
+ "h.12.ln_2.bias": "pytorch_model-00002-of-00006.bin",
71
+ "h.12.ln_2.weight": "pytorch_model-00002-of-00006.bin",
72
+ "h.12.mlp.c_fc.bias": "pytorch_model-00002-of-00006.bin",
73
+ "h.12.mlp.c_fc.weight": "pytorch_model-00002-of-00006.bin",
74
+ "h.12.mlp.c_proj.bias": "pytorch_model-00002-of-00006.bin",
75
+ "h.12.mlp.c_proj.weight": "pytorch_model-00002-of-00006.bin",
76
+ "h.13.attn.bias": "pytorch_model-00002-of-00006.bin",
77
+ "h.13.attn.c_attn.bias": "pytorch_model-00002-of-00006.bin",
78
+ "h.13.attn.c_attn.weight": "pytorch_model-00002-of-00006.bin",
79
+ "h.13.attn.c_proj.bias": "pytorch_model-00002-of-00006.bin",
80
+ "h.13.attn.c_proj.weight": "pytorch_model-00002-of-00006.bin",
81
+ "h.13.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
82
+ "h.13.ln_1.bias": "pytorch_model-00002-of-00006.bin",
83
+ "h.13.ln_1.weight": "pytorch_model-00002-of-00006.bin",
84
+ "h.13.ln_2.bias": "pytorch_model-00002-of-00006.bin",
85
+ "h.13.ln_2.weight": "pytorch_model-00002-of-00006.bin",
86
+ "h.13.mlp.c_fc.bias": "pytorch_model-00002-of-00006.bin",
87
+ "h.13.mlp.c_fc.weight": "pytorch_model-00002-of-00006.bin",
88
+ "h.13.mlp.c_proj.bias": "pytorch_model-00003-of-00006.bin",
89
+ "h.13.mlp.c_proj.weight": "pytorch_model-00003-of-00006.bin",
90
+ "h.14.attn.bias": "pytorch_model-00003-of-00006.bin",
91
+ "h.14.attn.c_attn.bias": "pytorch_model-00003-of-00006.bin",
92
+ "h.14.attn.c_attn.weight": "pytorch_model-00003-of-00006.bin",
93
+ "h.14.attn.c_proj.bias": "pytorch_model-00003-of-00006.bin",
94
+ "h.14.attn.c_proj.weight": "pytorch_model-00003-of-00006.bin",
95
+ "h.14.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
96
+ "h.14.ln_1.bias": "pytorch_model-00003-of-00006.bin",
97
+ "h.14.ln_1.weight": "pytorch_model-00003-of-00006.bin",
98
+ "h.14.ln_2.bias": "pytorch_model-00003-of-00006.bin",
99
+ "h.14.ln_2.weight": "pytorch_model-00003-of-00006.bin",
100
+ "h.14.mlp.c_fc.bias": "pytorch_model-00003-of-00006.bin",
101
+ "h.14.mlp.c_fc.weight": "pytorch_model-00003-of-00006.bin",
102
+ "h.14.mlp.c_proj.bias": "pytorch_model-00003-of-00006.bin",
103
+ "h.14.mlp.c_proj.weight": "pytorch_model-00003-of-00006.bin",
104
+ "h.15.attn.bias": "pytorch_model-00003-of-00006.bin",
105
+ "h.15.attn.c_attn.bias": "pytorch_model-00003-of-00006.bin",
106
+ "h.15.attn.c_attn.weight": "pytorch_model-00003-of-00006.bin",
107
+ "h.15.attn.c_proj.bias": "pytorch_model-00003-of-00006.bin",
108
+ "h.15.attn.c_proj.weight": "pytorch_model-00003-of-00006.bin",
109
+ "h.15.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
110
+ "h.15.ln_1.bias": "pytorch_model-00003-of-00006.bin",
111
+ "h.15.ln_1.weight": "pytorch_model-00003-of-00006.bin",
112
+ "h.15.ln_2.bias": "pytorch_model-00003-of-00006.bin",
113
+ "h.15.ln_2.weight": "pytorch_model-00003-of-00006.bin",
114
+ "h.15.mlp.c_fc.bias": "pytorch_model-00003-of-00006.bin",
115
+ "h.15.mlp.c_fc.weight": "pytorch_model-00003-of-00006.bin",
116
+ "h.15.mlp.c_proj.bias": "pytorch_model-00003-of-00006.bin",
117
+ "h.15.mlp.c_proj.weight": "pytorch_model-00003-of-00006.bin",
118
+ "h.16.attn.bias": "pytorch_model-00003-of-00006.bin",
119
+ "h.16.attn.c_attn.bias": "pytorch_model-00003-of-00006.bin",
120
+ "h.16.attn.c_attn.weight": "pytorch_model-00003-of-00006.bin",
121
+ "h.16.attn.c_proj.bias": "pytorch_model-00003-of-00006.bin",
122
+ "h.16.attn.c_proj.weight": "pytorch_model-00003-of-00006.bin",
123
+ "h.16.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
124
+ "h.16.ln_1.bias": "pytorch_model-00003-of-00006.bin",
125
+ "h.16.ln_1.weight": "pytorch_model-00003-of-00006.bin",
126
+ "h.16.ln_2.bias": "pytorch_model-00003-of-00006.bin",
127
+ "h.16.ln_2.weight": "pytorch_model-00003-of-00006.bin",
128
+ "h.16.mlp.c_fc.bias": "pytorch_model-00003-of-00006.bin",
129
+ "h.16.mlp.c_fc.weight": "pytorch_model-00003-of-00006.bin",
130
+ "h.16.mlp.c_proj.bias": "pytorch_model-00003-of-00006.bin",
131
+ "h.16.mlp.c_proj.weight": "pytorch_model-00003-of-00006.bin",
132
+ "h.17.attn.bias": "pytorch_model-00003-of-00006.bin",
133
+ "h.17.attn.c_attn.bias": "pytorch_model-00003-of-00006.bin",
134
+ "h.17.attn.c_attn.weight": "pytorch_model-00003-of-00006.bin",
135
+ "h.17.attn.c_proj.bias": "pytorch_model-00003-of-00006.bin",
136
+ "h.17.attn.c_proj.weight": "pytorch_model-00003-of-00006.bin",
137
+ "h.17.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
138
+ "h.17.ln_1.bias": "pytorch_model-00003-of-00006.bin",
139
+ "h.17.ln_1.weight": "pytorch_model-00003-of-00006.bin",
140
+ "h.17.ln_2.bias": "pytorch_model-00003-of-00006.bin",
141
+ "h.17.ln_2.weight": "pytorch_model-00003-of-00006.bin",
142
+ "h.17.mlp.c_fc.bias": "pytorch_model-00003-of-00006.bin",
143
+ "h.17.mlp.c_fc.weight": "pytorch_model-00003-of-00006.bin",
144
+ "h.17.mlp.c_proj.bias": "pytorch_model-00003-of-00006.bin",
145
+ "h.17.mlp.c_proj.weight": "pytorch_model-00003-of-00006.bin",
146
+ "h.18.attn.bias": "pytorch_model-00003-of-00006.bin",
147
+ "h.18.attn.c_attn.bias": "pytorch_model-00003-of-00006.bin",
148
+ "h.18.attn.c_attn.weight": "pytorch_model-00003-of-00006.bin",
149
+ "h.18.attn.c_proj.bias": "pytorch_model-00003-of-00006.bin",
150
+ "h.18.attn.c_proj.weight": "pytorch_model-00003-of-00006.bin",
151
+ "h.18.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
152
+ "h.18.ln_1.bias": "pytorch_model-00003-of-00006.bin",
153
+ "h.18.ln_1.weight": "pytorch_model-00003-of-00006.bin",
154
+ "h.18.ln_2.bias": "pytorch_model-00003-of-00006.bin",
155
+ "h.18.ln_2.weight": "pytorch_model-00003-of-00006.bin",
156
+ "h.18.mlp.c_fc.bias": "pytorch_model-00003-of-00006.bin",
157
+ "h.18.mlp.c_fc.weight": "pytorch_model-00003-of-00006.bin",
158
+ "h.18.mlp.c_proj.bias": "pytorch_model-00003-of-00006.bin",
159
+ "h.18.mlp.c_proj.weight": "pytorch_model-00003-of-00006.bin",
160
+ "h.19.attn.bias": "pytorch_model-00003-of-00006.bin",
161
+ "h.19.attn.c_attn.bias": "pytorch_model-00003-of-00006.bin",
162
+ "h.19.attn.c_attn.weight": "pytorch_model-00003-of-00006.bin",
163
+ "h.19.attn.c_proj.bias": "pytorch_model-00003-of-00006.bin",
164
+ "h.19.attn.c_proj.weight": "pytorch_model-00003-of-00006.bin",
165
+ "h.19.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
166
+ "h.19.ln_1.bias": "pytorch_model-00003-of-00006.bin",
167
+ "h.19.ln_1.weight": "pytorch_model-00003-of-00006.bin",
168
+ "h.19.ln_2.bias": "pytorch_model-00003-of-00006.bin",
169
+ "h.19.ln_2.weight": "pytorch_model-00003-of-00006.bin",
170
+ "h.19.mlp.c_fc.bias": "pytorch_model-00003-of-00006.bin",
171
+ "h.19.mlp.c_fc.weight": "pytorch_model-00003-of-00006.bin",
172
+ "h.19.mlp.c_proj.bias": "pytorch_model-00003-of-00006.bin",
173
+ "h.19.mlp.c_proj.weight": "pytorch_model-00003-of-00006.bin",
174
+ "h.2.attn.bias": "pytorch_model-00001-of-00006.bin",
175
+ "h.2.attn.c_attn.bias": "pytorch_model-00001-of-00006.bin",
176
+ "h.2.attn.c_attn.weight": "pytorch_model-00001-of-00006.bin",
177
+ "h.2.attn.c_proj.bias": "pytorch_model-00001-of-00006.bin",
178
+ "h.2.attn.c_proj.weight": "pytorch_model-00001-of-00006.bin",
179
+ "h.2.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
180
+ "h.2.ln_1.bias": "pytorch_model-00001-of-00006.bin",
181
+ "h.2.ln_1.weight": "pytorch_model-00001-of-00006.bin",
182
+ "h.2.ln_2.bias": "pytorch_model-00001-of-00006.bin",
183
+ "h.2.ln_2.weight": "pytorch_model-00001-of-00006.bin",
184
+ "h.2.mlp.c_fc.bias": "pytorch_model-00001-of-00006.bin",
185
+ "h.2.mlp.c_fc.weight": "pytorch_model-00001-of-00006.bin",
186
+ "h.2.mlp.c_proj.bias": "pytorch_model-00001-of-00006.bin",
187
+ "h.2.mlp.c_proj.weight": "pytorch_model-00001-of-00006.bin",
188
+ "h.20.attn.bias": "pytorch_model-00003-of-00006.bin",
189
+ "h.20.attn.c_attn.bias": "pytorch_model-00003-of-00006.bin",
190
+ "h.20.attn.c_attn.weight": "pytorch_model-00003-of-00006.bin",
191
+ "h.20.attn.c_proj.bias": "pytorch_model-00003-of-00006.bin",
192
+ "h.20.attn.c_proj.weight": "pytorch_model-00003-of-00006.bin",
193
+ "h.20.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
194
+ "h.20.ln_1.bias": "pytorch_model-00003-of-00006.bin",
195
+ "h.20.ln_1.weight": "pytorch_model-00003-of-00006.bin",
196
+ "h.20.ln_2.bias": "pytorch_model-00003-of-00006.bin",
197
+ "h.20.ln_2.weight": "pytorch_model-00003-of-00006.bin",
198
+ "h.20.mlp.c_fc.bias": "pytorch_model-00003-of-00006.bin",
199
+ "h.20.mlp.c_fc.weight": "pytorch_model-00003-of-00006.bin",
200
+ "h.20.mlp.c_proj.bias": "pytorch_model-00003-of-00006.bin",
201
+ "h.20.mlp.c_proj.weight": "pytorch_model-00003-of-00006.bin",
202
+ "h.21.attn.bias": "pytorch_model-00003-of-00006.bin",
203
+ "h.21.attn.c_attn.bias": "pytorch_model-00004-of-00006.bin",
204
+ "h.21.attn.c_attn.weight": "pytorch_model-00004-of-00006.bin",
205
+ "h.21.attn.c_proj.bias": "pytorch_model-00004-of-00006.bin",
206
+ "h.21.attn.c_proj.weight": "pytorch_model-00004-of-00006.bin",
207
+ "h.21.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
208
+ "h.21.ln_1.bias": "pytorch_model-00003-of-00006.bin",
209
+ "h.21.ln_1.weight": "pytorch_model-00003-of-00006.bin",
210
+ "h.21.ln_2.bias": "pytorch_model-00004-of-00006.bin",
211
+ "h.21.ln_2.weight": "pytorch_model-00004-of-00006.bin",
212
+ "h.21.mlp.c_fc.bias": "pytorch_model-00004-of-00006.bin",
213
+ "h.21.mlp.c_fc.weight": "pytorch_model-00004-of-00006.bin",
214
+ "h.21.mlp.c_proj.bias": "pytorch_model-00004-of-00006.bin",
215
+ "h.21.mlp.c_proj.weight": "pytorch_model-00004-of-00006.bin",
216
+ "h.22.attn.bias": "pytorch_model-00004-of-00006.bin",
217
+ "h.22.attn.c_attn.bias": "pytorch_model-00004-of-00006.bin",
218
+ "h.22.attn.c_attn.weight": "pytorch_model-00004-of-00006.bin",
219
+ "h.22.attn.c_proj.bias": "pytorch_model-00004-of-00006.bin",
220
+ "h.22.attn.c_proj.weight": "pytorch_model-00004-of-00006.bin",
221
+ "h.22.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
222
+ "h.22.ln_1.bias": "pytorch_model-00004-of-00006.bin",
223
+ "h.22.ln_1.weight": "pytorch_model-00004-of-00006.bin",
224
+ "h.22.ln_2.bias": "pytorch_model-00004-of-00006.bin",
225
+ "h.22.ln_2.weight": "pytorch_model-00004-of-00006.bin",
226
+ "h.22.mlp.c_fc.bias": "pytorch_model-00004-of-00006.bin",
227
+ "h.22.mlp.c_fc.weight": "pytorch_model-00004-of-00006.bin",
228
+ "h.22.mlp.c_proj.bias": "pytorch_model-00004-of-00006.bin",
229
+ "h.22.mlp.c_proj.weight": "pytorch_model-00004-of-00006.bin",
230
+ "h.23.attn.bias": "pytorch_model-00004-of-00006.bin",
231
+ "h.23.attn.c_attn.bias": "pytorch_model-00004-of-00006.bin",
232
+ "h.23.attn.c_attn.weight": "pytorch_model-00004-of-00006.bin",
233
+ "h.23.attn.c_proj.bias": "pytorch_model-00004-of-00006.bin",
234
+ "h.23.attn.c_proj.weight": "pytorch_model-00004-of-00006.bin",
235
+ "h.23.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
236
+ "h.23.ln_1.bias": "pytorch_model-00004-of-00006.bin",
237
+ "h.23.ln_1.weight": "pytorch_model-00004-of-00006.bin",
238
+ "h.23.ln_2.bias": "pytorch_model-00004-of-00006.bin",
239
+ "h.23.ln_2.weight": "pytorch_model-00004-of-00006.bin",
240
+ "h.23.mlp.c_fc.bias": "pytorch_model-00004-of-00006.bin",
241
+ "h.23.mlp.c_fc.weight": "pytorch_model-00004-of-00006.bin",
242
+ "h.23.mlp.c_proj.bias": "pytorch_model-00004-of-00006.bin",
243
+ "h.23.mlp.c_proj.weight": "pytorch_model-00004-of-00006.bin",
244
+ "h.24.attn.bias": "pytorch_model-00004-of-00006.bin",
245
+ "h.24.attn.c_attn.bias": "pytorch_model-00004-of-00006.bin",
246
+ "h.24.attn.c_attn.weight": "pytorch_model-00004-of-00006.bin",
247
+ "h.24.attn.c_proj.bias": "pytorch_model-00004-of-00006.bin",
248
+ "h.24.attn.c_proj.weight": "pytorch_model-00004-of-00006.bin",
249
+ "h.24.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
250
+ "h.24.ln_1.bias": "pytorch_model-00004-of-00006.bin",
251
+ "h.24.ln_1.weight": "pytorch_model-00004-of-00006.bin",
252
+ "h.24.ln_2.bias": "pytorch_model-00004-of-00006.bin",
253
+ "h.24.ln_2.weight": "pytorch_model-00004-of-00006.bin",
254
+ "h.24.mlp.c_fc.bias": "pytorch_model-00004-of-00006.bin",
255
+ "h.24.mlp.c_fc.weight": "pytorch_model-00004-of-00006.bin",
256
+ "h.24.mlp.c_proj.bias": "pytorch_model-00004-of-00006.bin",
257
+ "h.24.mlp.c_proj.weight": "pytorch_model-00004-of-00006.bin",
258
+ "h.25.attn.bias": "pytorch_model-00004-of-00006.bin",
259
+ "h.25.attn.c_attn.bias": "pytorch_model-00004-of-00006.bin",
260
+ "h.25.attn.c_attn.weight": "pytorch_model-00004-of-00006.bin",
261
+ "h.25.attn.c_proj.bias": "pytorch_model-00004-of-00006.bin",
262
+ "h.25.attn.c_proj.weight": "pytorch_model-00004-of-00006.bin",
263
+ "h.25.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
264
+ "h.25.ln_1.bias": "pytorch_model-00004-of-00006.bin",
265
+ "h.25.ln_1.weight": "pytorch_model-00004-of-00006.bin",
266
+ "h.25.ln_2.bias": "pytorch_model-00004-of-00006.bin",
267
+ "h.25.ln_2.weight": "pytorch_model-00004-of-00006.bin",
268
+ "h.25.mlp.c_fc.bias": "pytorch_model-00004-of-00006.bin",
269
+ "h.25.mlp.c_fc.weight": "pytorch_model-00004-of-00006.bin",
270
+ "h.25.mlp.c_proj.bias": "pytorch_model-00004-of-00006.bin",
271
+ "h.25.mlp.c_proj.weight": "pytorch_model-00004-of-00006.bin",
272
+ "h.26.attn.bias": "pytorch_model-00004-of-00006.bin",
273
+ "h.26.attn.c_attn.bias": "pytorch_model-00004-of-00006.bin",
274
+ "h.26.attn.c_attn.weight": "pytorch_model-00004-of-00006.bin",
275
+ "h.26.attn.c_proj.bias": "pytorch_model-00004-of-00006.bin",
276
+ "h.26.attn.c_proj.weight": "pytorch_model-00004-of-00006.bin",
277
+ "h.26.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
278
+ "h.26.ln_1.bias": "pytorch_model-00004-of-00006.bin",
279
+ "h.26.ln_1.weight": "pytorch_model-00004-of-00006.bin",
280
+ "h.26.ln_2.bias": "pytorch_model-00004-of-00006.bin",
281
+ "h.26.ln_2.weight": "pytorch_model-00004-of-00006.bin",
282
+ "h.26.mlp.c_fc.bias": "pytorch_model-00004-of-00006.bin",
283
+ "h.26.mlp.c_fc.weight": "pytorch_model-00004-of-00006.bin",
284
+ "h.26.mlp.c_proj.bias": "pytorch_model-00004-of-00006.bin",
285
+ "h.26.mlp.c_proj.weight": "pytorch_model-00004-of-00006.bin",
286
+ "h.27.attn.bias": "pytorch_model-00004-of-00006.bin",
287
+ "h.27.attn.c_attn.bias": "pytorch_model-00004-of-00006.bin",
288
+ "h.27.attn.c_attn.weight": "pytorch_model-00004-of-00006.bin",
289
+ "h.27.attn.c_proj.bias": "pytorch_model-00004-of-00006.bin",
290
+ "h.27.attn.c_proj.weight": "pytorch_model-00004-of-00006.bin",
291
+ "h.27.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
292
+ "h.27.ln_1.bias": "pytorch_model-00004-of-00006.bin",
293
+ "h.27.ln_1.weight": "pytorch_model-00004-of-00006.bin",
294
+ "h.27.ln_2.bias": "pytorch_model-00004-of-00006.bin",
295
+ "h.27.ln_2.weight": "pytorch_model-00004-of-00006.bin",
296
+ "h.27.mlp.c_fc.bias": "pytorch_model-00004-of-00006.bin",
297
+ "h.27.mlp.c_fc.weight": "pytorch_model-00004-of-00006.bin",
298
+ "h.27.mlp.c_proj.bias": "pytorch_model-00004-of-00006.bin",
299
+ "h.27.mlp.c_proj.weight": "pytorch_model-00004-of-00006.bin",
300
+ "h.28.attn.bias": "pytorch_model-00004-of-00006.bin",
301
+ "h.28.attn.c_attn.bias": "pytorch_model-00004-of-00006.bin",
302
+ "h.28.attn.c_attn.weight": "pytorch_model-00004-of-00006.bin",
303
+ "h.28.attn.c_proj.bias": "pytorch_model-00004-of-00006.bin",
304
+ "h.28.attn.c_proj.weight": "pytorch_model-00004-of-00006.bin",
305
+ "h.28.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
306
+ "h.28.ln_1.bias": "pytorch_model-00004-of-00006.bin",
307
+ "h.28.ln_1.weight": "pytorch_model-00004-of-00006.bin",
308
+ "h.28.ln_2.bias": "pytorch_model-00004-of-00006.bin",
309
+ "h.28.ln_2.weight": "pytorch_model-00004-of-00006.bin",
310
+ "h.28.mlp.c_fc.bias": "pytorch_model-00005-of-00006.bin",
311
+ "h.28.mlp.c_fc.weight": "pytorch_model-00005-of-00006.bin",
312
+ "h.28.mlp.c_proj.bias": "pytorch_model-00005-of-00006.bin",
313
+ "h.28.mlp.c_proj.weight": "pytorch_model-00005-of-00006.bin",
314
+ "h.29.attn.bias": "pytorch_model-00005-of-00006.bin",
315
+ "h.29.attn.c_attn.bias": "pytorch_model-00005-of-00006.bin",
316
+ "h.29.attn.c_attn.weight": "pytorch_model-00005-of-00006.bin",
317
+ "h.29.attn.c_proj.bias": "pytorch_model-00005-of-00006.bin",
318
+ "h.29.attn.c_proj.weight": "pytorch_model-00005-of-00006.bin",
319
+ "h.29.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
320
+ "h.29.ln_1.bias": "pytorch_model-00005-of-00006.bin",
321
+ "h.29.ln_1.weight": "pytorch_model-00005-of-00006.bin",
322
+ "h.29.ln_2.bias": "pytorch_model-00005-of-00006.bin",
323
+ "h.29.ln_2.weight": "pytorch_model-00005-of-00006.bin",
324
+ "h.29.mlp.c_fc.bias": "pytorch_model-00005-of-00006.bin",
325
+ "h.29.mlp.c_fc.weight": "pytorch_model-00005-of-00006.bin",
326
+ "h.29.mlp.c_proj.bias": "pytorch_model-00005-of-00006.bin",
327
+ "h.29.mlp.c_proj.weight": "pytorch_model-00005-of-00006.bin",
328
+ "h.3.attn.bias": "pytorch_model-00001-of-00006.bin",
329
+ "h.3.attn.c_attn.bias": "pytorch_model-00001-of-00006.bin",
330
+ "h.3.attn.c_attn.weight": "pytorch_model-00001-of-00006.bin",
331
+ "h.3.attn.c_proj.bias": "pytorch_model-00001-of-00006.bin",
332
+ "h.3.attn.c_proj.weight": "pytorch_model-00001-of-00006.bin",
333
+ "h.3.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
334
+ "h.3.ln_1.bias": "pytorch_model-00001-of-00006.bin",
335
+ "h.3.ln_1.weight": "pytorch_model-00001-of-00006.bin",
336
+ "h.3.ln_2.bias": "pytorch_model-00001-of-00006.bin",
337
+ "h.3.ln_2.weight": "pytorch_model-00001-of-00006.bin",
338
+ "h.3.mlp.c_fc.bias": "pytorch_model-00001-of-00006.bin",
339
+ "h.3.mlp.c_fc.weight": "pytorch_model-00001-of-00006.bin",
340
+ "h.3.mlp.c_proj.bias": "pytorch_model-00001-of-00006.bin",
341
+ "h.3.mlp.c_proj.weight": "pytorch_model-00001-of-00006.bin",
342
+ "h.30.attn.bias": "pytorch_model-00005-of-00006.bin",
343
+ "h.30.attn.c_attn.bias": "pytorch_model-00005-of-00006.bin",
344
+ "h.30.attn.c_attn.weight": "pytorch_model-00005-of-00006.bin",
345
+ "h.30.attn.c_proj.bias": "pytorch_model-00005-of-00006.bin",
346
+ "h.30.attn.c_proj.weight": "pytorch_model-00005-of-00006.bin",
347
+ "h.30.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
348
+ "h.30.ln_1.bias": "pytorch_model-00005-of-00006.bin",
349
+ "h.30.ln_1.weight": "pytorch_model-00005-of-00006.bin",
350
+ "h.30.ln_2.bias": "pytorch_model-00005-of-00006.bin",
351
+ "h.30.ln_2.weight": "pytorch_model-00005-of-00006.bin",
352
+ "h.30.mlp.c_fc.bias": "pytorch_model-00005-of-00006.bin",
353
+ "h.30.mlp.c_fc.weight": "pytorch_model-00005-of-00006.bin",
354
+ "h.30.mlp.c_proj.bias": "pytorch_model-00005-of-00006.bin",
355
+ "h.30.mlp.c_proj.weight": "pytorch_model-00005-of-00006.bin",
356
+ "h.31.attn.bias": "pytorch_model-00005-of-00006.bin",
357
+ "h.31.attn.c_attn.bias": "pytorch_model-00005-of-00006.bin",
358
+ "h.31.attn.c_attn.weight": "pytorch_model-00005-of-00006.bin",
359
+ "h.31.attn.c_proj.bias": "pytorch_model-00005-of-00006.bin",
360
+ "h.31.attn.c_proj.weight": "pytorch_model-00005-of-00006.bin",
361
+ "h.31.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
362
+ "h.31.ln_1.bias": "pytorch_model-00005-of-00006.bin",
363
+ "h.31.ln_1.weight": "pytorch_model-00005-of-00006.bin",
364
+ "h.31.ln_2.bias": "pytorch_model-00005-of-00006.bin",
365
+ "h.31.ln_2.weight": "pytorch_model-00005-of-00006.bin",
366
+ "h.31.mlp.c_fc.bias": "pytorch_model-00005-of-00006.bin",
367
+ "h.31.mlp.c_fc.weight": "pytorch_model-00005-of-00006.bin",
368
+ "h.31.mlp.c_proj.bias": "pytorch_model-00005-of-00006.bin",
369
+ "h.31.mlp.c_proj.weight": "pytorch_model-00005-of-00006.bin",
370
+ "h.32.attn.bias": "pytorch_model-00005-of-00006.bin",
371
+ "h.32.attn.c_attn.bias": "pytorch_model-00005-of-00006.bin",
372
+ "h.32.attn.c_attn.weight": "pytorch_model-00005-of-00006.bin",
373
+ "h.32.attn.c_proj.bias": "pytorch_model-00005-of-00006.bin",
374
+ "h.32.attn.c_proj.weight": "pytorch_model-00005-of-00006.bin",
375
+ "h.32.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
376
+ "h.32.ln_1.bias": "pytorch_model-00005-of-00006.bin",
377
+ "h.32.ln_1.weight": "pytorch_model-00005-of-00006.bin",
378
+ "h.32.ln_2.bias": "pytorch_model-00005-of-00006.bin",
379
+ "h.32.ln_2.weight": "pytorch_model-00005-of-00006.bin",
380
+ "h.32.mlp.c_fc.bias": "pytorch_model-00005-of-00006.bin",
381
+ "h.32.mlp.c_fc.weight": "pytorch_model-00005-of-00006.bin",
382
+ "h.32.mlp.c_proj.bias": "pytorch_model-00005-of-00006.bin",
383
+ "h.32.mlp.c_proj.weight": "pytorch_model-00005-of-00006.bin",
384
+ "h.33.attn.bias": "pytorch_model-00005-of-00006.bin",
385
+ "h.33.attn.c_attn.bias": "pytorch_model-00005-of-00006.bin",
386
+ "h.33.attn.c_attn.weight": "pytorch_model-00005-of-00006.bin",
387
+ "h.33.attn.c_proj.bias": "pytorch_model-00005-of-00006.bin",
388
+ "h.33.attn.c_proj.weight": "pytorch_model-00005-of-00006.bin",
389
+ "h.33.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
390
+ "h.33.ln_1.bias": "pytorch_model-00005-of-00006.bin",
391
+ "h.33.ln_1.weight": "pytorch_model-00005-of-00006.bin",
392
+ "h.33.ln_2.bias": "pytorch_model-00005-of-00006.bin",
393
+ "h.33.ln_2.weight": "pytorch_model-00005-of-00006.bin",
394
+ "h.33.mlp.c_fc.bias": "pytorch_model-00005-of-00006.bin",
395
+ "h.33.mlp.c_fc.weight": "pytorch_model-00005-of-00006.bin",
396
+ "h.33.mlp.c_proj.bias": "pytorch_model-00005-of-00006.bin",
397
+ "h.33.mlp.c_proj.weight": "pytorch_model-00005-of-00006.bin",
398
+ "h.34.attn.bias": "pytorch_model-00005-of-00006.bin",
399
+ "h.34.attn.c_attn.bias": "pytorch_model-00005-of-00006.bin",
400
+ "h.34.attn.c_attn.weight": "pytorch_model-00005-of-00006.bin",
401
+ "h.34.attn.c_proj.bias": "pytorch_model-00005-of-00006.bin",
402
+ "h.34.attn.c_proj.weight": "pytorch_model-00005-of-00006.bin",
403
+ "h.34.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
404
+ "h.34.ln_1.bias": "pytorch_model-00005-of-00006.bin",
405
+ "h.34.ln_1.weight": "pytorch_model-00005-of-00006.bin",
406
+ "h.34.ln_2.bias": "pytorch_model-00005-of-00006.bin",
407
+ "h.34.ln_2.weight": "pytorch_model-00005-of-00006.bin",
408
+ "h.34.mlp.c_fc.bias": "pytorch_model-00005-of-00006.bin",
409
+ "h.34.mlp.c_fc.weight": "pytorch_model-00005-of-00006.bin",
410
+ "h.34.mlp.c_proj.bias": "pytorch_model-00005-of-00006.bin",
411
+ "h.34.mlp.c_proj.weight": "pytorch_model-00005-of-00006.bin",
412
+ "h.35.attn.bias": "pytorch_model-00005-of-00006.bin",
413
+ "h.35.attn.c_attn.bias": "pytorch_model-00005-of-00006.bin",
414
+ "h.35.attn.c_attn.weight": "pytorch_model-00005-of-00006.bin",
415
+ "h.35.attn.c_proj.bias": "pytorch_model-00005-of-00006.bin",
416
+ "h.35.attn.c_proj.weight": "pytorch_model-00005-of-00006.bin",
417
+ "h.35.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
418
+ "h.35.ln_1.bias": "pytorch_model-00005-of-00006.bin",
419
+ "h.35.ln_1.weight": "pytorch_model-00005-of-00006.bin",
420
+ "h.35.ln_2.bias": "pytorch_model-00005-of-00006.bin",
421
+ "h.35.ln_2.weight": "pytorch_model-00005-of-00006.bin",
422
+ "h.35.mlp.c_fc.bias": "pytorch_model-00005-of-00006.bin",
423
+ "h.35.mlp.c_fc.weight": "pytorch_model-00005-of-00006.bin",
424
+ "h.35.mlp.c_proj.bias": "pytorch_model-00006-of-00006.bin",
425
+ "h.35.mlp.c_proj.weight": "pytorch_model-00006-of-00006.bin",
426
+ "h.36.attn.bias": "pytorch_model-00006-of-00006.bin",
427
+ "h.36.attn.c_attn.bias": "pytorch_model-00006-of-00006.bin",
428
+ "h.36.attn.c_attn.weight": "pytorch_model-00006-of-00006.bin",
429
+ "h.36.attn.c_proj.bias": "pytorch_model-00006-of-00006.bin",
430
+ "h.36.attn.c_proj.weight": "pytorch_model-00006-of-00006.bin",
431
+ "h.36.attn.masked_bias": "pytorch_model-00006-of-00006.bin",
432
+ "h.36.ln_1.bias": "pytorch_model-00006-of-00006.bin",
433
+ "h.36.ln_1.weight": "pytorch_model-00006-of-00006.bin",
434
+ "h.36.ln_2.bias": "pytorch_model-00006-of-00006.bin",
435
+ "h.36.ln_2.weight": "pytorch_model-00006-of-00006.bin",
436
+ "h.36.mlp.c_fc.bias": "pytorch_model-00006-of-00006.bin",
437
+ "h.36.mlp.c_fc.weight": "pytorch_model-00006-of-00006.bin",
438
+ "h.36.mlp.c_proj.bias": "pytorch_model-00006-of-00006.bin",
439
+ "h.36.mlp.c_proj.weight": "pytorch_model-00006-of-00006.bin",
440
+ "h.37.attn.bias": "pytorch_model-00006-of-00006.bin",
441
+ "h.37.attn.c_attn.bias": "pytorch_model-00006-of-00006.bin",
442
+ "h.37.attn.c_attn.weight": "pytorch_model-00006-of-00006.bin",
443
+ "h.37.attn.c_proj.bias": "pytorch_model-00006-of-00006.bin",
444
+ "h.37.attn.c_proj.weight": "pytorch_model-00006-of-00006.bin",
445
+ "h.37.attn.masked_bias": "pytorch_model-00006-of-00006.bin",
446
+ "h.37.ln_1.bias": "pytorch_model-00006-of-00006.bin",
447
+ "h.37.ln_1.weight": "pytorch_model-00006-of-00006.bin",
448
+ "h.37.ln_2.bias": "pytorch_model-00006-of-00006.bin",
449
+ "h.37.ln_2.weight": "pytorch_model-00006-of-00006.bin",
450
+ "h.37.mlp.c_fc.bias": "pytorch_model-00006-of-00006.bin",
451
+ "h.37.mlp.c_fc.weight": "pytorch_model-00006-of-00006.bin",
452
+ "h.37.mlp.c_proj.bias": "pytorch_model-00006-of-00006.bin",
453
+ "h.37.mlp.c_proj.weight": "pytorch_model-00006-of-00006.bin",
454
+ "h.38.attn.bias": "pytorch_model-00006-of-00006.bin",
455
+ "h.38.attn.c_attn.bias": "pytorch_model-00006-of-00006.bin",
456
+ "h.38.attn.c_attn.weight": "pytorch_model-00006-of-00006.bin",
457
+ "h.38.attn.c_proj.bias": "pytorch_model-00006-of-00006.bin",
458
+ "h.38.attn.c_proj.weight": "pytorch_model-00006-of-00006.bin",
459
+ "h.38.attn.masked_bias": "pytorch_model-00006-of-00006.bin",
460
+ "h.38.ln_1.bias": "pytorch_model-00006-of-00006.bin",
461
+ "h.38.ln_1.weight": "pytorch_model-00006-of-00006.bin",
462
+ "h.38.ln_2.bias": "pytorch_model-00006-of-00006.bin",
463
+ "h.38.ln_2.weight": "pytorch_model-00006-of-00006.bin",
464
+ "h.38.mlp.c_fc.bias": "pytorch_model-00006-of-00006.bin",
465
+ "h.38.mlp.c_fc.weight": "pytorch_model-00006-of-00006.bin",
466
+ "h.38.mlp.c_proj.bias": "pytorch_model-00006-of-00006.bin",
467
+ "h.38.mlp.c_proj.weight": "pytorch_model-00006-of-00006.bin",
468
+ "h.39.attn.bias": "pytorch_model-00006-of-00006.bin",
469
+ "h.39.attn.c_attn.bias": "pytorch_model-00006-of-00006.bin",
470
+ "h.39.attn.c_attn.weight": "pytorch_model-00006-of-00006.bin",
471
+ "h.39.attn.c_proj.bias": "pytorch_model-00006-of-00006.bin",
472
+ "h.39.attn.c_proj.weight": "pytorch_model-00006-of-00006.bin",
473
+ "h.39.attn.masked_bias": "pytorch_model-00006-of-00006.bin",
474
+ "h.39.ln_1.bias": "pytorch_model-00006-of-00006.bin",
475
+ "h.39.ln_1.weight": "pytorch_model-00006-of-00006.bin",
476
+ "h.39.ln_2.bias": "pytorch_model-00006-of-00006.bin",
477
+ "h.39.ln_2.weight": "pytorch_model-00006-of-00006.bin",
478
+ "h.39.mlp.c_fc.bias": "pytorch_model-00006-of-00006.bin",
479
+ "h.39.mlp.c_fc.weight": "pytorch_model-00006-of-00006.bin",
480
+ "h.39.mlp.c_proj.bias": "pytorch_model-00006-of-00006.bin",
481
+ "h.39.mlp.c_proj.weight": "pytorch_model-00006-of-00006.bin",
482
+ "h.4.attn.bias": "pytorch_model-00001-of-00006.bin",
483
+ "h.4.attn.c_attn.bias": "pytorch_model-00001-of-00006.bin",
484
+ "h.4.attn.c_attn.weight": "pytorch_model-00001-of-00006.bin",
485
+ "h.4.attn.c_proj.bias": "pytorch_model-00001-of-00006.bin",
486
+ "h.4.attn.c_proj.weight": "pytorch_model-00001-of-00006.bin",
487
+ "h.4.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
488
+ "h.4.ln_1.bias": "pytorch_model-00001-of-00006.bin",
489
+ "h.4.ln_1.weight": "pytorch_model-00001-of-00006.bin",
490
+ "h.4.ln_2.bias": "pytorch_model-00001-of-00006.bin",
491
+ "h.4.ln_2.weight": "pytorch_model-00001-of-00006.bin",
492
+ "h.4.mlp.c_fc.bias": "pytorch_model-00001-of-00006.bin",
493
+ "h.4.mlp.c_fc.weight": "pytorch_model-00001-of-00006.bin",
494
+ "h.4.mlp.c_proj.bias": "pytorch_model-00001-of-00006.bin",
495
+ "h.4.mlp.c_proj.weight": "pytorch_model-00001-of-00006.bin",
496
+ "h.5.attn.bias": "pytorch_model-00001-of-00006.bin",
497
+ "h.5.attn.c_attn.bias": "pytorch_model-00001-of-00006.bin",
498
+ "h.5.attn.c_attn.weight": "pytorch_model-00001-of-00006.bin",
499
+ "h.5.attn.c_proj.bias": "pytorch_model-00001-of-00006.bin",
500
+ "h.5.attn.c_proj.weight": "pytorch_model-00001-of-00006.bin",
501
+ "h.5.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
502
+ "h.5.ln_1.bias": "pytorch_model-00001-of-00006.bin",
503
+ "h.5.ln_1.weight": "pytorch_model-00001-of-00006.bin",
504
+ "h.5.ln_2.bias": "pytorch_model-00001-of-00006.bin",
505
+ "h.5.ln_2.weight": "pytorch_model-00001-of-00006.bin",
506
+ "h.5.mlp.c_fc.bias": "pytorch_model-00001-of-00006.bin",
507
+ "h.5.mlp.c_fc.weight": "pytorch_model-00001-of-00006.bin",
508
+ "h.5.mlp.c_proj.bias": "pytorch_model-00001-of-00006.bin",
509
+ "h.5.mlp.c_proj.weight": "pytorch_model-00001-of-00006.bin",
510
+ "h.6.attn.bias": "pytorch_model-00001-of-00006.bin",
511
+ "h.6.attn.c_attn.bias": "pytorch_model-00001-of-00006.bin",
512
+ "h.6.attn.c_attn.weight": "pytorch_model-00001-of-00006.bin",
513
+ "h.6.attn.c_proj.bias": "pytorch_model-00001-of-00006.bin",
514
+ "h.6.attn.c_proj.weight": "pytorch_model-00001-of-00006.bin",
515
+ "h.6.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
516
+ "h.6.ln_1.bias": "pytorch_model-00001-of-00006.bin",
517
+ "h.6.ln_1.weight": "pytorch_model-00001-of-00006.bin",
518
+ "h.6.ln_2.bias": "pytorch_model-00001-of-00006.bin",
519
+ "h.6.ln_2.weight": "pytorch_model-00001-of-00006.bin",
520
+ "h.6.mlp.c_fc.bias": "pytorch_model-00002-of-00006.bin",
521
+ "h.6.mlp.c_fc.weight": "pytorch_model-00002-of-00006.bin",
522
+ "h.6.mlp.c_proj.bias": "pytorch_model-00002-of-00006.bin",
523
+ "h.6.mlp.c_proj.weight": "pytorch_model-00002-of-00006.bin",
524
+ "h.7.attn.bias": "pytorch_model-00002-of-00006.bin",
525
+ "h.7.attn.c_attn.bias": "pytorch_model-00002-of-00006.bin",
526
+ "h.7.attn.c_attn.weight": "pytorch_model-00002-of-00006.bin",
527
+ "h.7.attn.c_proj.bias": "pytorch_model-00002-of-00006.bin",
528
+ "h.7.attn.c_proj.weight": "pytorch_model-00002-of-00006.bin",
529
+ "h.7.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
530
+ "h.7.ln_1.bias": "pytorch_model-00002-of-00006.bin",
531
+ "h.7.ln_1.weight": "pytorch_model-00002-of-00006.bin",
532
+ "h.7.ln_2.bias": "pytorch_model-00002-of-00006.bin",
533
+ "h.7.ln_2.weight": "pytorch_model-00002-of-00006.bin",
534
+ "h.7.mlp.c_fc.bias": "pytorch_model-00002-of-00006.bin",
535
+ "h.7.mlp.c_fc.weight": "pytorch_model-00002-of-00006.bin",
536
+ "h.7.mlp.c_proj.bias": "pytorch_model-00002-of-00006.bin",
537
+ "h.7.mlp.c_proj.weight": "pytorch_model-00002-of-00006.bin",
538
+ "h.8.attn.bias": "pytorch_model-00002-of-00006.bin",
539
+ "h.8.attn.c_attn.bias": "pytorch_model-00002-of-00006.bin",
540
+ "h.8.attn.c_attn.weight": "pytorch_model-00002-of-00006.bin",
541
+ "h.8.attn.c_proj.bias": "pytorch_model-00002-of-00006.bin",
542
+ "h.8.attn.c_proj.weight": "pytorch_model-00002-of-00006.bin",
543
+ "h.8.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
544
+ "h.8.ln_1.bias": "pytorch_model-00002-of-00006.bin",
545
+ "h.8.ln_1.weight": "pytorch_model-00002-of-00006.bin",
546
+ "h.8.ln_2.bias": "pytorch_model-00002-of-00006.bin",
547
+ "h.8.ln_2.weight": "pytorch_model-00002-of-00006.bin",
548
+ "h.8.mlp.c_fc.bias": "pytorch_model-00002-of-00006.bin",
549
+ "h.8.mlp.c_fc.weight": "pytorch_model-00002-of-00006.bin",
550
+ "h.8.mlp.c_proj.bias": "pytorch_model-00002-of-00006.bin",
551
+ "h.8.mlp.c_proj.weight": "pytorch_model-00002-of-00006.bin",
552
+ "h.9.attn.bias": "pytorch_model-00002-of-00006.bin",
553
+ "h.9.attn.c_attn.bias": "pytorch_model-00002-of-00006.bin",
554
+ "h.9.attn.c_attn.weight": "pytorch_model-00002-of-00006.bin",
555
+ "h.9.attn.c_proj.bias": "pytorch_model-00002-of-00006.bin",
556
+ "h.9.attn.c_proj.weight": "pytorch_model-00002-of-00006.bin",
557
+ "h.9.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
558
+ "h.9.ln_1.bias": "pytorch_model-00002-of-00006.bin",
559
+ "h.9.ln_1.weight": "pytorch_model-00002-of-00006.bin",
560
+ "h.9.ln_2.bias": "pytorch_model-00002-of-00006.bin",
561
+ "h.9.ln_2.weight": "pytorch_model-00002-of-00006.bin",
562
+ "h.9.mlp.c_fc.bias": "pytorch_model-00002-of-00006.bin",
563
+ "h.9.mlp.c_fc.weight": "pytorch_model-00002-of-00006.bin",
564
+ "h.9.mlp.c_proj.bias": "pytorch_model-00002-of-00006.bin",
565
+ "h.9.mlp.c_proj.weight": "pytorch_model-00002-of-00006.bin",
566
+ "ln_f.bias": "pytorch_model-00006-of-00006.bin",
567
+ "ln_f.weight": "pytorch_model-00006-of-00006.bin",
568
+ "wpe.weight": "pytorch_model-00001-of-00006.bin",
569
+ "wte.weight": "pytorch_model-00001-of-00006.bin"
570
+ }
571
+ }