Hugging Face's logo

onceuponai-dev
/
mlc-chat-Qra-1b-q4f32_0

Model card Files Community
onceuponai commited on
Commit
13e14aa
·
verified ·
1 Parent(s): 35dba73

add model

Browse files
Files changed (29) hide show
  1. mlc-chat-config.json +22 -0
  2. ndarray-cache.json +2651 -0
  3. params_shard_0.bin +3 -0
  4. params_shard_1.bin +3 -0
  5. params_shard_10.bin +3 -0
  6. params_shard_11.bin +3 -0
  7. params_shard_12.bin +3 -0
  8. params_shard_13.bin +3 -0
  9. params_shard_14.bin +3 -0
  10. params_shard_15.bin +3 -0
  11. params_shard_16.bin +3 -0
  12. params_shard_17.bin +3 -0
  13. params_shard_18.bin +3 -0
  14. params_shard_19.bin +3 -0
  15. params_shard_2.bin +3 -0
  16. params_shard_20.bin +3 -0
  17. params_shard_21.bin +3 -0
  18. params_shard_22.bin +3 -0
  19. params_shard_23.bin +3 -0
  20. params_shard_3.bin +3 -0
  21. params_shard_4.bin +3 -0
  22. params_shard_5.bin +3 -0
  23. params_shard_6.bin +3 -0
  24. params_shard_7.bin +3 -0
  25. params_shard_8.bin +3 -0
  26. params_shard_9.bin +3 -0
  27. tokenizer.json +0 -0
  28. tokenizer.model +3 -0
  29. tokenizer_config.json +42 -0
mlc-chat-config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_lib": "Qra-1b-q4f32_0",
3
+ "local_id": "Qra-1b-q4f32_0",
4
+ "conv_template": "llama-2",
5
+ "temperature": 0.7,
6
+ "repetition_penalty": 1.0,
7
+ "top_p": 0.95,
8
+ "mean_gen_len": 128,
9
+ "max_gen_len": 2048,
10
+ "num_shards": 1,
11
+ "use_presharded_weights": false,
12
+ "shift_fill_factor": 0.3,
13
+ "tokenizer_files": [
14
+ "tokenizer.model",
15
+ "tokenizer.json"
16
+ ],
17
+ "model_category": "llama",
18
+ "model_name": "Qra-1b",
19
+ "vocab_size": 32000,
20
+ "prefill_chunk_size": -1,
21
+ "max_window_size": 2048
22
+ }
ndarray-cache.json ADDED
@@ -0,0 +1,2651 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 227
4
+ },
5
+ "records": [
6
+ {
7
+ "dataPath": "params_shard_0.bin",
8
+ "format": "raw-shard",
9
+ "nbytes": 32768000,
10
+ "records": [
11
+ {
12
+ "name": "param_0",
13
+ "shape": [
14
+ 32000,
15
+ 256
16
+ ],
17
+ "dtype": "uint32",
18
+ "format": "raw",
19
+ "nbytes": 32768000,
20
+ "byteOffset": 0
21
+ }
22
+ ],
23
+ "md5sum": "569ec712551eb7b7cababd04e1a31176"
24
+ },
25
+ {
26
+ "dataPath": "params_shard_1.bin",
27
+ "format": "raw-shard",
28
+ "nbytes": 28508160,
29
+ "records": [
30
+ {
31
+ "name": "param_1",
32
+ "shape": [
33
+ 32000,
34
+ 64
35
+ ],
36
+ "dtype": "uint32",
37
+ "format": "raw",
38
+ "nbytes": 8192000,
39
+ "byteOffset": 0
40
+ },
41
+ {
42
+ "name": "param_2",
43
+ "shape": [
44
+ 256,
45
+ 2560
46
+ ],
47
+ "dtype": "uint32",
48
+ "format": "raw",
49
+ "nbytes": 2621440,
50
+ "byteOffset": 8192000
51
+ },
52
+ {
53
+ "name": "param_3",
54
+ "shape": [
55
+ 64,
56
+ 2560
57
+ ],
58
+ "dtype": "uint32",
59
+ "format": "raw",
60
+ "nbytes": 655360,
61
+ "byteOffset": 10813440
62
+ },
63
+ {
64
+ "name": "param_4",
65
+ "shape": [
66
+ 256,
67
+ 2048
68
+ ],
69
+ "dtype": "uint32",
70
+ "format": "raw",
71
+ "nbytes": 2097152,
72
+ "byteOffset": 11468800
73
+ },
74
+ {
75
+ "name": "param_5",
76
+ "shape": [
77
+ 64,
78
+ 2048
79
+ ],
80
+ "dtype": "uint32",
81
+ "format": "raw",
82
+ "nbytes": 524288,
83
+ "byteOffset": 13565952
84
+ },
85
+ {
86
+ "name": "param_6",
87
+ "shape": [
88
+ 256,
89
+ 11264
90
+ ],
91
+ "dtype": "uint32",
92
+ "format": "raw",
93
+ "nbytes": 11534336,
94
+ "byteOffset": 14090240
95
+ },
96
+ {
97
+ "name": "param_7",
98
+ "shape": [
99
+ 64,
100
+ 11264
101
+ ],
102
+ "dtype": "uint32",
103
+ "format": "raw",
104
+ "nbytes": 2883584,
105
+ "byteOffset": 25624576
106
+ }
107
+ ],
108
+ "md5sum": "25cb89b3b054b0f7e42ca1582e918a33"
109
+ },
110
+ {
111
+ "dataPath": "params_shard_2.bin",
112
+ "format": "raw-shard",
113
+ "nbytes": 33308672,
114
+ "records": [
115
+ {
116
+ "name": "param_8",
117
+ "shape": [
118
+ 704,
119
+ 2048
120
+ ],
121
+ "dtype": "uint32",
122
+ "format": "raw",
123
+ "nbytes": 5767168,
124
+ "byteOffset": 0
125
+ },
126
+ {
127
+ "name": "param_9",
128
+ "shape": [
129
+ 176,
130
+ 2048
131
+ ],
132
+ "dtype": "uint32",
133
+ "format": "raw",
134
+ "nbytes": 1441792,
135
+ "byteOffset": 5767168
136
+ },
137
+ {
138
+ "name": "param_10",
139
+ "shape": [
140
+ 2048
141
+ ],
142
+ "dtype": "float32",
143
+ "format": "raw",
144
+ "nbytes": 8192,
145
+ "byteOffset": 7208960
146
+ },
147
+ {
148
+ "name": "param_11",
149
+ "shape": [
150
+ 2048
151
+ ],
152
+ "dtype": "float32",
153
+ "format": "raw",
154
+ "nbytes": 8192,
155
+ "byteOffset": 7217152
156
+ },
157
+ {
158
+ "name": "param_12",
159
+ "shape": [
160
+ 256,
161
+ 2560
162
+ ],
163
+ "dtype": "uint32",
164
+ "format": "raw",
165
+ "nbytes": 2621440,
166
+ "byteOffset": 7225344
167
+ },
168
+ {
169
+ "name": "param_13",
170
+ "shape": [
171
+ 64,
172
+ 2560
173
+ ],
174
+ "dtype": "uint32",
175
+ "format": "raw",
176
+ "nbytes": 655360,
177
+ "byteOffset": 9846784
178
+ },
179
+ {
180
+ "name": "param_14",
181
+ "shape": [
182
+ 256,
183
+ 2048
184
+ ],
185
+ "dtype": "uint32",
186
+ "format": "raw",
187
+ "nbytes": 2097152,
188
+ "byteOffset": 10502144
189
+ },
190
+ {
191
+ "name": "param_15",
192
+ "shape": [
193
+ 64,
194
+ 2048
195
+ ],
196
+ "dtype": "uint32",
197
+ "format": "raw",
198
+ "nbytes": 524288,
199
+ "byteOffset": 12599296
200
+ },
201
+ {
202
+ "name": "param_16",
203
+ "shape": [
204
+ 256,
205
+ 11264
206
+ ],
207
+ "dtype": "uint32",
208
+ "format": "raw",
209
+ "nbytes": 11534336,
210
+ "byteOffset": 13123584
211
+ },
212
+ {
213
+ "name": "param_17",
214
+ "shape": [
215
+ 64,
216
+ 11264
217
+ ],
218
+ "dtype": "uint32",
219
+ "format": "raw",
220
+ "nbytes": 2883584,
221
+ "byteOffset": 24657920
222
+ },
223
+ {
224
+ "name": "param_18",
225
+ "shape": [
226
+ 704,
227
+ 2048
228
+ ],
229
+ "dtype": "uint32",
230
+ "format": "raw",
231
+ "nbytes": 5767168,
232
+ "byteOffset": 27541504
233
+ }
234
+ ],
235
+ "md5sum": "66a03e73a6f63f8c14c1aca809e6dea4"
236
+ },
237
+ {
238
+ "dataPath": "params_shard_3.bin",
239
+ "format": "raw-shard",
240
+ "nbytes": 32276480,
241
+ "records": [
242
+ {
243
+ "name": "param_19",
244
+ "shape": [
245
+ 176,
246
+ 2048
247
+ ],
248
+ "dtype": "uint32",
249
+ "format": "raw",
250
+ "nbytes": 1441792,
251
+ "byteOffset": 0
252
+ },
253
+ {
254
+ "name": "param_20",
255
+ "shape": [
256
+ 2048
257
+ ],
258
+ "dtype": "float32",
259
+ "format": "raw",
260
+ "nbytes": 8192,
261
+ "byteOffset": 1441792
262
+ },
263
+ {
264
+ "name": "param_21",
265
+ "shape": [
266
+ 2048
267
+ ],
268
+ "dtype": "float32",
269
+ "format": "raw",
270
+ "nbytes": 8192,
271
+ "byteOffset": 1449984
272
+ },
273
+ {
274
+ "name": "param_22",
275
+ "shape": [
276
+ 256,
277
+ 2560
278
+ ],
279
+ "dtype": "uint32",
280
+ "format": "raw",
281
+ "nbytes": 2621440,
282
+ "byteOffset": 1458176
283
+ },
284
+ {
285
+ "name": "param_23",
286
+ "shape": [
287
+ 64,
288
+ 2560
289
+ ],
290
+ "dtype": "uint32",
291
+ "format": "raw",
292
+ "nbytes": 655360,
293
+ "byteOffset": 4079616
294
+ },
295
+ {
296
+ "name": "param_24",
297
+ "shape": [
298
+ 256,
299
+ 2048
300
+ ],
301
+ "dtype": "uint32",
302
+ "format": "raw",
303
+ "nbytes": 2097152,
304
+ "byteOffset": 4734976
305
+ },
306
+ {
307
+ "name": "param_25",
308
+ "shape": [
309
+ 64,
310
+ 2048
311
+ ],
312
+ "dtype": "uint32",
313
+ "format": "raw",
314
+ "nbytes": 524288,
315
+ "byteOffset": 6832128
316
+ },
317
+ {
318
+ "name": "param_26",
319
+ "shape": [
320
+ 256,
321
+ 11264
322
+ ],
323
+ "dtype": "uint32",
324
+ "format": "raw",
325
+ "nbytes": 11534336,
326
+ "byteOffset": 7356416
327
+ },
328
+ {
329
+ "name": "param_27",
330
+ "shape": [
331
+ 64,
332
+ 11264
333
+ ],
334
+ "dtype": "uint32",
335
+ "format": "raw",
336
+ "nbytes": 2883584,
337
+ "byteOffset": 18890752
338
+ },
339
+ {
340
+ "name": "param_28",
341
+ "shape": [
342
+ 704,
343
+ 2048
344
+ ],
345
+ "dtype": "uint32",
346
+ "format": "raw",
347
+ "nbytes": 5767168,
348
+ "byteOffset": 21774336
349
+ },
350
+ {
351
+ "name": "param_29",
352
+ "shape": [
353
+ 176,
354
+ 2048
355
+ ],
356
+ "dtype": "uint32",
357
+ "format": "raw",
358
+ "nbytes": 1441792,
359
+ "byteOffset": 27541504
360
+ },
361
+ {
362
+ "name": "param_30",
363
+ "shape": [
364
+ 2048
365
+ ],
366
+ "dtype": "float32",
367
+ "format": "raw",
368
+ "nbytes": 8192,
369
+ "byteOffset": 28983296
370
+ },
371
+ {
372
+ "name": "param_31",
373
+ "shape": [
374
+ 2048
375
+ ],
376
+ "dtype": "float32",
377
+ "format": "raw",
378
+ "nbytes": 8192,
379
+ "byteOffset": 28991488
380
+ },
381
+ {
382
+ "name": "param_32",
383
+ "shape": [
384
+ 256,
385
+ 2560
386
+ ],
387
+ "dtype": "uint32",
388
+ "format": "raw",
389
+ "nbytes": 2621440,
390
+ "byteOffset": 28999680
391
+ },
392
+ {
393
+ "name": "param_33",
394
+ "shape": [
395
+ 64,
396
+ 2560
397
+ ],
398
+ "dtype": "uint32",
399
+ "format": "raw",
400
+ "nbytes": 655360,
401
+ "byteOffset": 31621120
402
+ }
403
+ ],
404
+ "md5sum": "f98023fbfdce276f96ecf1d751d63200"
405
+ },
406
+ {
407
+ "dataPath": "params_shard_4.bin",
408
+ "format": "raw-shard",
409
+ "nbytes": 30162944,
410
+ "records": [
411
+ {
412
+ "name": "param_34",
413
+ "shape": [
414
+ 256,
415
+ 2048
416
+ ],
417
+ "dtype": "uint32",
418
+ "format": "raw",
419
+ "nbytes": 2097152,
420
+ "byteOffset": 0
421
+ },
422
+ {
423
+ "name": "param_35",
424
+ "shape": [
425
+ 64,
426
+ 2048
427
+ ],
428
+ "dtype": "uint32",
429
+ "format": "raw",
430
+ "nbytes": 524288,
431
+ "byteOffset": 2097152
432
+ },
433
+ {
434
+ "name": "param_36",
435
+ "shape": [
436
+ 256,
437
+ 11264
438
+ ],
439
+ "dtype": "uint32",
440
+ "format": "raw",
441
+ "nbytes": 11534336,
442
+ "byteOffset": 2621440
443
+ },
444
+ {
445
+ "name": "param_37",
446
+ "shape": [
447
+ 64,
448
+ 11264
449
+ ],
450
+ "dtype": "uint32",
451
+ "format": "raw",
452
+ "nbytes": 2883584,
453
+ "byteOffset": 14155776
454
+ },
455
+ {
456
+ "name": "param_38",
457
+ "shape": [
458
+ 704,
459
+ 2048
460
+ ],
461
+ "dtype": "uint32",
462
+ "format": "raw",
463
+ "nbytes": 5767168,
464
+ "byteOffset": 17039360
465
+ },
466
+ {
467
+ "name": "param_39",
468
+ "shape": [
469
+ 176,
470
+ 2048
471
+ ],
472
+ "dtype": "uint32",
473
+ "format": "raw",
474
+ "nbytes": 1441792,
475
+ "byteOffset": 22806528
476
+ },
477
+ {
478
+ "name": "param_40",
479
+ "shape": [
480
+ 2048
481
+ ],
482
+ "dtype": "float32",
483
+ "format": "raw",
484
+ "nbytes": 8192,
485
+ "byteOffset": 24248320
486
+ },
487
+ {
488
+ "name": "param_41",
489
+ "shape": [
490
+ 2048
491
+ ],
492
+ "dtype": "float32",
493
+ "format": "raw",
494
+ "nbytes": 8192,
495
+ "byteOffset": 24256512
496
+ },
497
+ {
498
+ "name": "param_42",
499
+ "shape": [
500
+ 256,
501
+ 2560
502
+ ],
503
+ "dtype": "uint32",
504
+ "format": "raw",
505
+ "nbytes": 2621440,
506
+ "byteOffset": 24264704
507
+ },
508
+ {
509
+ "name": "param_43",
510
+ "shape": [
511
+ 64,
512
+ 2560
513
+ ],
514
+ "dtype": "uint32",
515
+ "format": "raw",
516
+ "nbytes": 655360,
517
+ "byteOffset": 26886144
518
+ },
519
+ {
520
+ "name": "param_44",
521
+ "shape": [
522
+ 256,
523
+ 2048
524
+ ],
525
+ "dtype": "uint32",
526
+ "format": "raw",
527
+ "nbytes": 2097152,
528
+ "byteOffset": 27541504
529
+ },
530
+ {
531
+ "name": "param_45",
532
+ "shape": [
533
+ 64,
534
+ 2048
535
+ ],
536
+ "dtype": "uint32",
537
+ "format": "raw",
538
+ "nbytes": 524288,
539
+ "byteOffset": 29638656
540
+ }
541
+ ],
542
+ "md5sum": "10ae6f2821b72e3d9f132424e3baa7ca"
543
+ },
544
+ {
545
+ "dataPath": "params_shard_5.bin",
546
+ "format": "raw-shard",
547
+ "nbytes": 27541504,
548
+ "records": [
549
+ {
550
+ "name": "param_46",
551
+ "shape": [
552
+ 256,
553
+ 11264
554
+ ],
555
+ "dtype": "uint32",
556
+ "format": "raw",
557
+ "nbytes": 11534336,
558
+ "byteOffset": 0
559
+ },
560
+ {
561
+ "name": "param_47",
562
+ "shape": [
563
+ 64,
564
+ 11264
565
+ ],
566
+ "dtype": "uint32",
567
+ "format": "raw",
568
+ "nbytes": 2883584,
569
+ "byteOffset": 11534336
570
+ },
571
+ {
572
+ "name": "param_48",
573
+ "shape": [
574
+ 704,
575
+ 2048
576
+ ],
577
+ "dtype": "uint32",
578
+ "format": "raw",
579
+ "nbytes": 5767168,
580
+ "byteOffset": 14417920
581
+ },
582
+ {
583
+ "name": "param_49",
584
+ "shape": [
585
+ 176,
586
+ 2048
587
+ ],
588
+ "dtype": "uint32",
589
+ "format": "raw",
590
+ "nbytes": 1441792,
591
+ "byteOffset": 20185088
592
+ },
593
+ {
594
+ "name": "param_50",
595
+ "shape": [
596
+ 2048
597
+ ],
598
+ "dtype": "float32",
599
+ "format": "raw",
600
+ "nbytes": 8192,
601
+ "byteOffset": 21626880
602
+ },
603
+ {
604
+ "name": "param_51",
605
+ "shape": [
606
+ 2048
607
+ ],
608
+ "dtype": "float32",
609
+ "format": "raw",
610
+ "nbytes": 8192,
611
+ "byteOffset": 21635072
612
+ },
613
+ {
614
+ "name": "param_52",
615
+ "shape": [
616
+ 256,
617
+ 2560
618
+ ],
619
+ "dtype": "uint32",
620
+ "format": "raw",
621
+ "nbytes": 2621440,
622
+ "byteOffset": 21643264
623
+ },
624
+ {
625
+ "name": "param_53",
626
+ "shape": [
627
+ 64,
628
+ 2560
629
+ ],
630
+ "dtype": "uint32",
631
+ "format": "raw",
632
+ "nbytes": 655360,
633
+ "byteOffset": 24264704
634
+ },
635
+ {
636
+ "name": "param_54",
637
+ "shape": [
638
+ 256,
639
+ 2048
640
+ ],
641
+ "dtype": "uint32",
642
+ "format": "raw",
643
+ "nbytes": 2097152,
644
+ "byteOffset": 24920064
645
+ },
646
+ {
647
+ "name": "param_55",
648
+ "shape": [
649
+ 64,
650
+ 2048
651
+ ],
652
+ "dtype": "uint32",
653
+ "format": "raw",
654
+ "nbytes": 524288,
655
+ "byteOffset": 27017216
656
+ }
657
+ ],
658
+ "md5sum": "6d039289d1776b630ae96210c4db6e2b"
659
+ },
660
+ {
661
+ "dataPath": "params_shard_6.bin",
662
+ "format": "raw-shard",
663
+ "nbytes": 27541504,
664
+ "records": [
665
+ {
666
+ "name": "param_56",
667
+ "shape": [
668
+ 256,
669
+ 11264
670
+ ],
671
+ "dtype": "uint32",
672
+ "format": "raw",
673
+ "nbytes": 11534336,
674
+ "byteOffset": 0
675
+ },
676
+ {
677
+ "name": "param_57",
678
+ "shape": [
679
+ 64,
680
+ 11264
681
+ ],
682
+ "dtype": "uint32",
683
+ "format": "raw",
684
+ "nbytes": 2883584,
685
+ "byteOffset": 11534336
686
+ },
687
+ {
688
+ "name": "param_58",
689
+ "shape": [
690
+ 704,
691
+ 2048
692
+ ],
693
+ "dtype": "uint32",
694
+ "format": "raw",
695
+ "nbytes": 5767168,
696
+ "byteOffset": 14417920
697
+ },
698
+ {
699
+ "name": "param_59",
700
+ "shape": [
701
+ 176,
702
+ 2048
703
+ ],
704
+ "dtype": "uint32",
705
+ "format": "raw",
706
+ "nbytes": 1441792,
707
+ "byteOffset": 20185088
708
+ },
709
+ {
710
+ "name": "param_60",
711
+ "shape": [
712
+ 2048
713
+ ],
714
+ "dtype": "float32",
715
+ "format": "raw",
716
+ "nbytes": 8192,
717
+ "byteOffset": 21626880
718
+ },
719
+ {
720
+ "name": "param_61",
721
+ "shape": [
722
+ 2048
723
+ ],
724
+ "dtype": "float32",
725
+ "format": "raw",
726
+ "nbytes": 8192,
727
+ "byteOffset": 21635072
728
+ },
729
+ {
730
+ "name": "param_62",
731
+ "shape": [
732
+ 256,
733
+ 2560
734
+ ],
735
+ "dtype": "uint32",
736
+ "format": "raw",
737
+ "nbytes": 2621440,
738
+ "byteOffset": 21643264
739
+ },
740
+ {
741
+ "name": "param_63",
742
+ "shape": [
743
+ 64,
744
+ 2560
745
+ ],
746
+ "dtype": "uint32",
747
+ "format": "raw",
748
+ "nbytes": 655360,
749
+ "byteOffset": 24264704
750
+ },
751
+ {
752
+ "name": "param_64",
753
+ "shape": [
754
+ 256,
755
+ 2048
756
+ ],
757
+ "dtype": "uint32",
758
+ "format": "raw",
759
+ "nbytes": 2097152,
760
+ "byteOffset": 24920064
761
+ },
762
+ {
763
+ "name": "param_65",
764
+ "shape": [
765
+ 64,
766
+ 2048
767
+ ],
768
+ "dtype": "uint32",
769
+ "format": "raw",
770
+ "nbytes": 524288,
771
+ "byteOffset": 27017216
772
+ }
773
+ ],
774
+ "md5sum": "e01540071aa308c6e0e36d6098323c6e"
775
+ },
776
+ {
777
+ "dataPath": "params_shard_7.bin",
778
+ "format": "raw-shard",
779
+ "nbytes": 27541504,
780
+ "records": [
781
+ {
782
+ "name": "param_66",
783
+ "shape": [
784
+ 256,
785
+ 11264
786
+ ],
787
+ "dtype": "uint32",
788
+ "format": "raw",
789
+ "nbytes": 11534336,
790
+ "byteOffset": 0
791
+ },
792
+ {
793
+ "name": "param_67",
794
+ "shape": [
795
+ 64,
796
+ 11264
797
+ ],
798
+ "dtype": "uint32",
799
+ "format": "raw",
800
+ "nbytes": 2883584,
801
+ "byteOffset": 11534336
802
+ },
803
+ {
804
+ "name": "param_68",
805
+ "shape": [
806
+ 704,
807
+ 2048
808
+ ],
809
+ "dtype": "uint32",
810
+ "format": "raw",
811
+ "nbytes": 5767168,
812
+ "byteOffset": 14417920
813
+ },
814
+ {
815
+ "name": "param_69",
816
+ "shape": [
817
+ 176,
818
+ 2048
819
+ ],
820
+ "dtype": "uint32",
821
+ "format": "raw",
822
+ "nbytes": 1441792,
823
+ "byteOffset": 20185088
824
+ },
825
+ {
826
+ "name": "param_70",
827
+ "shape": [
828
+ 2048
829
+ ],
830
+ "dtype": "float32",
831
+ "format": "raw",
832
+ "nbytes": 8192,
833
+ "byteOffset": 21626880
834
+ },
835
+ {
836
+ "name": "param_71",
837
+ "shape": [
838
+ 2048
839
+ ],
840
+ "dtype": "float32",
841
+ "format": "raw",
842
+ "nbytes": 8192,
843
+ "byteOffset": 21635072
844
+ },
845
+ {
846
+ "name": "param_72",
847
+ "shape": [
848
+ 256,
849
+ 2560
850
+ ],
851
+ "dtype": "uint32",
852
+ "format": "raw",
853
+ "nbytes": 2621440,
854
+ "byteOffset": 21643264
855
+ },
856
+ {
857
+ "name": "param_73",
858
+ "shape": [
859
+ 64,
860
+ 2560
861
+ ],
862
+ "dtype": "uint32",
863
+ "format": "raw",
864
+ "nbytes": 655360,
865
+ "byteOffset": 24264704
866
+ },
867
+ {
868
+ "name": "param_74",
869
+ "shape": [
870
+ 256,
871
+ 2048
872
+ ],
873
+ "dtype": "uint32",
874
+ "format": "raw",
875
+ "nbytes": 2097152,
876
+ "byteOffset": 24920064
877
+ },
878
+ {
879
+ "name": "param_75",
880
+ "shape": [
881
+ 64,
882
+ 2048
883
+ ],
884
+ "dtype": "uint32",
885
+ "format": "raw",
886
+ "nbytes": 524288,
887
+ "byteOffset": 27017216
888
+ }
889
+ ],
890
+ "md5sum": "e60680032757fb49881d4fc2756d69c3"
891
+ },
892
+ {
893
+ "dataPath": "params_shard_8.bin",
894
+ "format": "raw-shard",
895
+ "nbytes": 27541504,
896
+ "records": [
897
+ {
898
+ "name": "param_76",
899
+ "shape": [
900
+ 256,
901
+ 11264
902
+ ],
903
+ "dtype": "uint32",
904
+ "format": "raw",
905
+ "nbytes": 11534336,
906
+ "byteOffset": 0
907
+ },
908
+ {
909
+ "name": "param_77",
910
+ "shape": [
911
+ 64,
912
+ 11264
913
+ ],
914
+ "dtype": "uint32",
915
+ "format": "raw",
916
+ "nbytes": 2883584,
917
+ "byteOffset": 11534336
918
+ },
919
+ {
920
+ "name": "param_78",
921
+ "shape": [
922
+ 704,
923
+ 2048
924
+ ],
925
+ "dtype": "uint32",
926
+ "format": "raw",
927
+ "nbytes": 5767168,
928
+ "byteOffset": 14417920
929
+ },
930
+ {
931
+ "name": "param_79",
932
+ "shape": [
933
+ 176,
934
+ 2048
935
+ ],
936
+ "dtype": "uint32",
937
+ "format": "raw",
938
+ "nbytes": 1441792,
939
+ "byteOffset": 20185088
940
+ },
941
+ {
942
+ "name": "param_80",
943
+ "shape": [
944
+ 2048
945
+ ],
946
+ "dtype": "float32",
947
+ "format": "raw",
948
+ "nbytes": 8192,
949
+ "byteOffset": 21626880
950
+ },
951
+ {
952
+ "name": "param_81",
953
+ "shape": [
954
+ 2048
955
+ ],
956
+ "dtype": "float32",
957
+ "format": "raw",
958
+ "nbytes": 8192,
959
+ "byteOffset": 21635072
960
+ },
961
+ {
962
+ "name": "param_82",
963
+ "shape": [
964
+ 256,
965
+ 2560
966
+ ],
967
+ "dtype": "uint32",
968
+ "format": "raw",
969
+ "nbytes": 2621440,
970
+ "byteOffset": 21643264
971
+ },
972
+ {
973
+ "name": "param_83",
974
+ "shape": [
975
+ 64,
976
+ 2560
977
+ ],
978
+ "dtype": "uint32",
979
+ "format": "raw",
980
+ "nbytes": 655360,
981
+ "byteOffset": 24264704
982
+ },
983
+ {
984
+ "name": "param_84",
985
+ "shape": [
986
+ 256,
987
+ 2048
988
+ ],
989
+ "dtype": "uint32",
990
+ "format": "raw",
991
+ "nbytes": 2097152,
992
+ "byteOffset": 24920064
993
+ },
994
+ {
995
+ "name": "param_85",
996
+ "shape": [
997
+ 64,
998
+ 2048
999
+ ],
1000
+ "dtype": "uint32",
1001
+ "format": "raw",
1002
+ "nbytes": 524288,
1003
+ "byteOffset": 27017216
1004
+ }
1005
+ ],
1006
+ "md5sum": "80cbc528dffda03f25f9edd4283a807c"
1007
+ },
1008
+ {
1009
+ "dataPath": "params_shard_9.bin",
1010
+ "format": "raw-shard",
1011
+ "nbytes": 27541504,
1012
+ "records": [
1013
+ {
1014
+ "name": "param_86",
1015
+ "shape": [
1016
+ 256,
1017
+ 11264
1018
+ ],
1019
+ "dtype": "uint32",
1020
+ "format": "raw",
1021
+ "nbytes": 11534336,
1022
+ "byteOffset": 0
1023
+ },
1024
+ {
1025
+ "name": "param_87",
1026
+ "shape": [
1027
+ 64,
1028
+ 11264
1029
+ ],
1030
+ "dtype": "uint32",
1031
+ "format": "raw",
1032
+ "nbytes": 2883584,
1033
+ "byteOffset": 11534336
1034
+ },
1035
+ {
1036
+ "name": "param_88",
1037
+ "shape": [
1038
+ 704,
1039
+ 2048
1040
+ ],
1041
+ "dtype": "uint32",
1042
+ "format": "raw",
1043
+ "nbytes": 5767168,
1044
+ "byteOffset": 14417920
1045
+ },
1046
+ {
1047
+ "name": "param_89",
1048
+ "shape": [
1049
+ 176,
1050
+ 2048
1051
+ ],
1052
+ "dtype": "uint32",
1053
+ "format": "raw",
1054
+ "nbytes": 1441792,
1055
+ "byteOffset": 20185088
1056
+ },
1057
+ {
1058
+ "name": "param_90",
1059
+ "shape": [
1060
+ 2048
1061
+ ],
1062
+ "dtype": "float32",
1063
+ "format": "raw",
1064
+ "nbytes": 8192,
1065
+ "byteOffset": 21626880
1066
+ },
1067
+ {
1068
+ "name": "param_91",
1069
+ "shape": [
1070
+ 2048
1071
+ ],
1072
+ "dtype": "float32",
1073
+ "format": "raw",
1074
+ "nbytes": 8192,
1075
+ "byteOffset": 21635072
1076
+ },
1077
+ {
1078
+ "name": "param_92",
1079
+ "shape": [
1080
+ 256,
1081
+ 2560
1082
+ ],
1083
+ "dtype": "uint32",
1084
+ "format": "raw",
1085
+ "nbytes": 2621440,
1086
+ "byteOffset": 21643264
1087
+ },
1088
+ {
1089
+ "name": "param_93",
1090
+ "shape": [
1091
+ 64,
1092
+ 2560
1093
+ ],
1094
+ "dtype": "uint32",
1095
+ "format": "raw",
1096
+ "nbytes": 655360,
1097
+ "byteOffset": 24264704
1098
+ },
1099
+ {
1100
+ "name": "param_94",
1101
+ "shape": [
1102
+ 256,
1103
+ 2048
1104
+ ],
1105
+ "dtype": "uint32",
1106
+ "format": "raw",
1107
+ "nbytes": 2097152,
1108
+ "byteOffset": 24920064
1109
+ },
1110
+ {
1111
+ "name": "param_95",
1112
+ "shape": [
1113
+ 64,
1114
+ 2048
1115
+ ],
1116
+ "dtype": "uint32",
1117
+ "format": "raw",
1118
+ "nbytes": 524288,
1119
+ "byteOffset": 27017216
1120
+ }
1121
+ ],
1122
+ "md5sum": "654924ba601dd5e32187b0ce3e322ba2"
1123
+ },
1124
+ {
1125
+ "dataPath": "params_shard_10.bin",
1126
+ "format": "raw-shard",
1127
+ "nbytes": 27541504,
1128
+ "records": [
1129
+ {
1130
+ "name": "param_96",
1131
+ "shape": [
1132
+ 256,
1133
+ 11264
1134
+ ],
1135
+ "dtype": "uint32",
1136
+ "format": "raw",
1137
+ "nbytes": 11534336,
1138
+ "byteOffset": 0
1139
+ },
1140
+ {
1141
+ "name": "param_97",
1142
+ "shape": [
1143
+ 64,
1144
+ 11264
1145
+ ],
1146
+ "dtype": "uint32",
1147
+ "format": "raw",
1148
+ "nbytes": 2883584,
1149
+ "byteOffset": 11534336
1150
+ },
1151
+ {
1152
+ "name": "param_98",
1153
+ "shape": [
1154
+ 704,
1155
+ 2048
1156
+ ],
1157
+ "dtype": "uint32",
1158
+ "format": "raw",
1159
+ "nbytes": 5767168,
1160
+ "byteOffset": 14417920
1161
+ },
1162
+ {
1163
+ "name": "param_99",
1164
+ "shape": [
1165
+ 176,
1166
+ 2048
1167
+ ],
1168
+ "dtype": "uint32",
1169
+ "format": "raw",
1170
+ "nbytes": 1441792,
1171
+ "byteOffset": 20185088
1172
+ },
1173
+ {
1174
+ "name": "param_100",
1175
+ "shape": [
1176
+ 2048
1177
+ ],
1178
+ "dtype": "float32",
1179
+ "format": "raw",
1180
+ "nbytes": 8192,
1181
+ "byteOffset": 21626880
1182
+ },
1183
+ {
1184
+ "name": "param_101",
1185
+ "shape": [
1186
+ 2048
1187
+ ],
1188
+ "dtype": "float32",
1189
+ "format": "raw",
1190
+ "nbytes": 8192,
1191
+ "byteOffset": 21635072
1192
+ },
1193
+ {
1194
+ "name": "param_102",
1195
+ "shape": [
1196
+ 256,
1197
+ 2560
1198
+ ],
1199
+ "dtype": "uint32",
1200
+ "format": "raw",
1201
+ "nbytes": 2621440,
1202
+ "byteOffset": 21643264
1203
+ },
1204
+ {
1205
+ "name": "param_103",
1206
+ "shape": [
1207
+ 64,
1208
+ 2560
1209
+ ],
1210
+ "dtype": "uint32",
1211
+ "format": "raw",
1212
+ "nbytes": 655360,
1213
+ "byteOffset": 24264704
1214
+ },
1215
+ {
1216
+ "name": "param_104",
1217
+ "shape": [
1218
+ 256,
1219
+ 2048
1220
+ ],
1221
+ "dtype": "uint32",
1222
+ "format": "raw",
1223
+ "nbytes": 2097152,
1224
+ "byteOffset": 24920064
1225
+ },
1226
+ {
1227
+ "name": "param_105",
1228
+ "shape": [
1229
+ 64,
1230
+ 2048
1231
+ ],
1232
+ "dtype": "uint32",
1233
+ "format": "raw",
1234
+ "nbytes": 524288,
1235
+ "byteOffset": 27017216
1236
+ }
1237
+ ],
1238
+ "md5sum": "257be54b52ccd32e12bbcd53d063050d"
1239
+ },
1240
+ {
1241
+ "dataPath": "params_shard_11.bin",
1242
+ "format": "raw-shard",
1243
+ "nbytes": 27541504,
1244
+ "records": [
1245
+ {
1246
+ "name": "param_106",
1247
+ "shape": [
1248
+ 256,
1249
+ 11264
1250
+ ],
1251
+ "dtype": "uint32",
1252
+ "format": "raw",
1253
+ "nbytes": 11534336,
1254
+ "byteOffset": 0
1255
+ },
1256
+ {
1257
+ "name": "param_107",
1258
+ "shape": [
1259
+ 64,
1260
+ 11264
1261
+ ],
1262
+ "dtype": "uint32",
1263
+ "format": "raw",
1264
+ "nbytes": 2883584,
1265
+ "byteOffset": 11534336
1266
+ },
1267
+ {
1268
+ "name": "param_108",
1269
+ "shape": [
1270
+ 704,
1271
+ 2048
1272
+ ],
1273
+ "dtype": "uint32",
1274
+ "format": "raw",
1275
+ "nbytes": 5767168,
1276
+ "byteOffset": 14417920
1277
+ },
1278
+ {
1279
+ "name": "param_109",
1280
+ "shape": [
1281
+ 176,
1282
+ 2048
1283
+ ],
1284
+ "dtype": "uint32",
1285
+ "format": "raw",
1286
+ "nbytes": 1441792,
1287
+ "byteOffset": 20185088
1288
+ },
1289
+ {
1290
+ "name": "param_110",
1291
+ "shape": [
1292
+ 2048
1293
+ ],
1294
+ "dtype": "float32",
1295
+ "format": "raw",
1296
+ "nbytes": 8192,
1297
+ "byteOffset": 21626880
1298
+ },
1299
+ {
1300
+ "name": "param_111",
1301
+ "shape": [
1302
+ 2048
1303
+ ],
1304
+ "dtype": "float32",
1305
+ "format": "raw",
1306
+ "nbytes": 8192,
1307
+ "byteOffset": 21635072
1308
+ },
1309
+ {
1310
+ "name": "param_112",
1311
+ "shape": [
1312
+ 256,
1313
+ 2560
1314
+ ],
1315
+ "dtype": "uint32",
1316
+ "format": "raw",
1317
+ "nbytes": 2621440,
1318
+ "byteOffset": 21643264
1319
+ },
1320
+ {
1321
+ "name": "param_113",
1322
+ "shape": [
1323
+ 64,
1324
+ 2560
1325
+ ],
1326
+ "dtype": "uint32",
1327
+ "format": "raw",
1328
+ "nbytes": 655360,
1329
+ "byteOffset": 24264704
1330
+ },
1331
+ {
1332
+ "name": "param_114",
1333
+ "shape": [
1334
+ 256,
1335
+ 2048
1336
+ ],
1337
+ "dtype": "uint32",
1338
+ "format": "raw",
1339
+ "nbytes": 2097152,
1340
+ "byteOffset": 24920064
1341
+ },
1342
+ {
1343
+ "name": "param_115",
1344
+ "shape": [
1345
+ 64,
1346
+ 2048
1347
+ ],
1348
+ "dtype": "uint32",
1349
+ "format": "raw",
1350
+ "nbytes": 524288,
1351
+ "byteOffset": 27017216
1352
+ }
1353
+ ],
1354
+ "md5sum": "c897c8f8ab19d02ba67fad212d254476"
1355
+ },
1356
+ {
1357
+ "dataPath": "params_shard_12.bin",
1358
+ "format": "raw-shard",
1359
+ "nbytes": 27541504,
1360
+ "records": [
1361
+ {
1362
+ "name": "param_116",
1363
+ "shape": [
1364
+ 256,
1365
+ 11264
1366
+ ],
1367
+ "dtype": "uint32",
1368
+ "format": "raw",
1369
+ "nbytes": 11534336,
1370
+ "byteOffset": 0
1371
+ },
1372
+ {
1373
+ "name": "param_117",
1374
+ "shape": [
1375
+ 64,
1376
+ 11264
1377
+ ],
1378
+ "dtype": "uint32",
1379
+ "format": "raw",
1380
+ "nbytes": 2883584,
1381
+ "byteOffset": 11534336
1382
+ },
1383
+ {
1384
+ "name": "param_118",
1385
+ "shape": [
1386
+ 704,
1387
+ 2048
1388
+ ],
1389
+ "dtype": "uint32",
1390
+ "format": "raw",
1391
+ "nbytes": 5767168,
1392
+ "byteOffset": 14417920
1393
+ },
1394
+ {
1395
+ "name": "param_119",
1396
+ "shape": [
1397
+ 176,
1398
+ 2048
1399
+ ],
1400
+ "dtype": "uint32",
1401
+ "format": "raw",
1402
+ "nbytes": 1441792,
1403
+ "byteOffset": 20185088
1404
+ },
1405
+ {
1406
+ "name": "param_120",
1407
+ "shape": [
1408
+ 2048
1409
+ ],
1410
+ "dtype": "float32",
1411
+ "format": "raw",
1412
+ "nbytes": 8192,
1413
+ "byteOffset": 21626880
1414
+ },
1415
+ {
1416
+ "name": "param_121",
1417
+ "shape": [
1418
+ 2048
1419
+ ],
1420
+ "dtype": "float32",
1421
+ "format": "raw",
1422
+ "nbytes": 8192,
1423
+ "byteOffset": 21635072
1424
+ },
1425
+ {
1426
+ "name": "param_122",
1427
+ "shape": [
1428
+ 256,
1429
+ 2560
1430
+ ],
1431
+ "dtype": "uint32",
1432
+ "format": "raw",
1433
+ "nbytes": 2621440,
1434
+ "byteOffset": 21643264
1435
+ },
1436
+ {
1437
+ "name": "param_123",
1438
+ "shape": [
1439
+ 64,
1440
+ 2560
1441
+ ],
1442
+ "dtype": "uint32",
1443
+ "format": "raw",
1444
+ "nbytes": 655360,
1445
+ "byteOffset": 24264704
1446
+ },
1447
+ {
1448
+ "name": "param_124",
1449
+ "shape": [
1450
+ 256,
1451
+ 2048
1452
+ ],
1453
+ "dtype": "uint32",
1454
+ "format": "raw",
1455
+ "nbytes": 2097152,
1456
+ "byteOffset": 24920064
1457
+ },
1458
+ {
1459
+ "name": "param_125",
1460
+ "shape": [
1461
+ 64,
1462
+ 2048
1463
+ ],
1464
+ "dtype": "uint32",
1465
+ "format": "raw",
1466
+ "nbytes": 524288,
1467
+ "byteOffset": 27017216
1468
+ }
1469
+ ],
1470
+ "md5sum": "2716d8194c451c50e88be8ee2700a6e0"
1471
+ },
1472
+ {
1473
+ "dataPath": "params_shard_13.bin",
1474
+ "format": "raw-shard",
1475
+ "nbytes": 27541504,
1476
+ "records": [
1477
+ {
1478
+ "name": "param_126",
1479
+ "shape": [
1480
+ 256,
1481
+ 11264
1482
+ ],
1483
+ "dtype": "uint32",
1484
+ "format": "raw",
1485
+ "nbytes": 11534336,
1486
+ "byteOffset": 0
1487
+ },
1488
+ {
1489
+ "name": "param_127",
1490
+ "shape": [
1491
+ 64,
1492
+ 11264
1493
+ ],
1494
+ "dtype": "uint32",
1495
+ "format": "raw",
1496
+ "nbytes": 2883584,
1497
+ "byteOffset": 11534336
1498
+ },
1499
+ {
1500
+ "name": "param_128",
1501
+ "shape": [
1502
+ 704,
1503
+ 2048
1504
+ ],
1505
+ "dtype": "uint32",
1506
+ "format": "raw",
1507
+ "nbytes": 5767168,
1508
+ "byteOffset": 14417920
1509
+ },
1510
+ {
1511
+ "name": "param_129",
1512
+ "shape": [
1513
+ 176,
1514
+ 2048
1515
+ ],
1516
+ "dtype": "uint32",
1517
+ "format": "raw",
1518
+ "nbytes": 1441792,
1519
+ "byteOffset": 20185088
1520
+ },
1521
+ {
1522
+ "name": "param_130",
1523
+ "shape": [
1524
+ 2048
1525
+ ],
1526
+ "dtype": "float32",
1527
+ "format": "raw",
1528
+ "nbytes": 8192,
1529
+ "byteOffset": 21626880
1530
+ },
1531
+ {
1532
+ "name": "param_131",
1533
+ "shape": [
1534
+ 2048
1535
+ ],
1536
+ "dtype": "float32",
1537
+ "format": "raw",
1538
+ "nbytes": 8192,
1539
+ "byteOffset": 21635072
1540
+ },
1541
+ {
1542
+ "name": "param_132",
1543
+ "shape": [
1544
+ 256,
1545
+ 2560
1546
+ ],
1547
+ "dtype": "uint32",
1548
+ "format": "raw",
1549
+ "nbytes": 2621440,
1550
+ "byteOffset": 21643264
1551
+ },
1552
+ {
1553
+ "name": "param_133",
1554
+ "shape": [
1555
+ 64,
1556
+ 2560
1557
+ ],
1558
+ "dtype": "uint32",
1559
+ "format": "raw",
1560
+ "nbytes": 655360,
1561
+ "byteOffset": 24264704
1562
+ },
1563
+ {
1564
+ "name": "param_134",
1565
+ "shape": [
1566
+ 256,
1567
+ 2048
1568
+ ],
1569
+ "dtype": "uint32",
1570
+ "format": "raw",
1571
+ "nbytes": 2097152,
1572
+ "byteOffset": 24920064
1573
+ },
1574
+ {
1575
+ "name": "param_135",
1576
+ "shape": [
1577
+ 64,
1578
+ 2048
1579
+ ],
1580
+ "dtype": "uint32",
1581
+ "format": "raw",
1582
+ "nbytes": 524288,
1583
+ "byteOffset": 27017216
1584
+ }
1585
+ ],
1586
+ "md5sum": "7568cb649cc7f195a1bf0e73a2af1c03"
1587
+ },
1588
+ {
1589
+ "dataPath": "params_shard_14.bin",
1590
+ "format": "raw-shard",
1591
+ "nbytes": 27541504,
1592
+ "records": [
1593
+ {
1594
+ "name": "param_136",
1595
+ "shape": [
1596
+ 256,
1597
+ 11264
1598
+ ],
1599
+ "dtype": "uint32",
1600
+ "format": "raw",
1601
+ "nbytes": 11534336,
1602
+ "byteOffset": 0
1603
+ },
1604
+ {
1605
+ "name": "param_137",
1606
+ "shape": [
1607
+ 64,
1608
+ 11264
1609
+ ],
1610
+ "dtype": "uint32",
1611
+ "format": "raw",
1612
+ "nbytes": 2883584,
1613
+ "byteOffset": 11534336
1614
+ },
1615
+ {
1616
+ "name": "param_138",
1617
+ "shape": [
1618
+ 704,
1619
+ 2048
1620
+ ],
1621
+ "dtype": "uint32",
1622
+ "format": "raw",
1623
+ "nbytes": 5767168,
1624
+ "byteOffset": 14417920
1625
+ },
1626
+ {
1627
+ "name": "param_139",
1628
+ "shape": [
1629
+ 176,
1630
+ 2048
1631
+ ],
1632
+ "dtype": "uint32",
1633
+ "format": "raw",
1634
+ "nbytes": 1441792,
1635
+ "byteOffset": 20185088
1636
+ },
1637
+ {
1638
+ "name": "param_140",
1639
+ "shape": [
1640
+ 2048
1641
+ ],
1642
+ "dtype": "float32",
1643
+ "format": "raw",
1644
+ "nbytes": 8192,
1645
+ "byteOffset": 21626880
1646
+ },
1647
+ {
1648
+ "name": "param_141",
1649
+ "shape": [
1650
+ 2048
1651
+ ],
1652
+ "dtype": "float32",
1653
+ "format": "raw",
1654
+ "nbytes": 8192,
1655
+ "byteOffset": 21635072
1656
+ },
1657
+ {
1658
+ "name": "param_142",
1659
+ "shape": [
1660
+ 256,
1661
+ 2560
1662
+ ],
1663
+ "dtype": "uint32",
1664
+ "format": "raw",
1665
+ "nbytes": 2621440,
1666
+ "byteOffset": 21643264
1667
+ },
1668
+ {
1669
+ "name": "param_143",
1670
+ "shape": [
1671
+ 64,
1672
+ 2560
1673
+ ],
1674
+ "dtype": "uint32",
1675
+ "format": "raw",
1676
+ "nbytes": 655360,
1677
+ "byteOffset": 24264704
1678
+ },
1679
+ {
1680
+ "name": "param_144",
1681
+ "shape": [
1682
+ 256,
1683
+ 2048
1684
+ ],
1685
+ "dtype": "uint32",
1686
+ "format": "raw",
1687
+ "nbytes": 2097152,
1688
+ "byteOffset": 24920064
1689
+ },
1690
+ {
1691
+ "name": "param_145",
1692
+ "shape": [
1693
+ 64,
1694
+ 2048
1695
+ ],
1696
+ "dtype": "uint32",
1697
+ "format": "raw",
1698
+ "nbytes": 524288,
1699
+ "byteOffset": 27017216
1700
+ }
1701
+ ],
1702
+ "md5sum": "df92b77ff5f3ec8839fe34df2cc7be71"
1703
+ },
1704
+ {
1705
+ "dataPath": "params_shard_15.bin",
1706
+ "format": "raw-shard",
1707
+ "nbytes": 27541504,
1708
+ "records": [
1709
+ {
1710
+ "name": "param_146",
1711
+ "shape": [
1712
+ 256,
1713
+ 11264
1714
+ ],
1715
+ "dtype": "uint32",
1716
+ "format": "raw",
1717
+ "nbytes": 11534336,
1718
+ "byteOffset": 0
1719
+ },
1720
+ {
1721
+ "name": "param_147",
1722
+ "shape": [
1723
+ 64,
1724
+ 11264
1725
+ ],
1726
+ "dtype": "uint32",
1727
+ "format": "raw",
1728
+ "nbytes": 2883584,
1729
+ "byteOffset": 11534336
1730
+ },
1731
+ {
1732
+ "name": "param_148",
1733
+ "shape": [
1734
+ 704,
1735
+ 2048
1736
+ ],
1737
+ "dtype": "uint32",
1738
+ "format": "raw",
1739
+ "nbytes": 5767168,
1740
+ "byteOffset": 14417920
1741
+ },
1742
+ {
1743
+ "name": "param_149",
1744
+ "shape": [
1745
+ 176,
1746
+ 2048
1747
+ ],
1748
+ "dtype": "uint32",
1749
+ "format": "raw",
1750
+ "nbytes": 1441792,
1751
+ "byteOffset": 20185088
1752
+ },
1753
+ {
1754
+ "name": "param_150",
1755
+ "shape": [
1756
+ 2048
1757
+ ],
1758
+ "dtype": "float32",
1759
+ "format": "raw",
1760
+ "nbytes": 8192,
1761
+ "byteOffset": 21626880
1762
+ },
1763
+ {
1764
+ "name": "param_151",
1765
+ "shape": [
1766
+ 2048
1767
+ ],
1768
+ "dtype": "float32",
1769
+ "format": "raw",
1770
+ "nbytes": 8192,
1771
+ "byteOffset": 21635072
1772
+ },
1773
+ {
1774
+ "name": "param_152",
1775
+ "shape": [
1776
+ 256,
1777
+ 2560
1778
+ ],
1779
+ "dtype": "uint32",
1780
+ "format": "raw",
1781
+ "nbytes": 2621440,
1782
+ "byteOffset": 21643264
1783
+ },
1784
+ {
1785
+ "name": "param_153",
1786
+ "shape": [
1787
+ 64,
1788
+ 2560
1789
+ ],
1790
+ "dtype": "uint32",
1791
+ "format": "raw",
1792
+ "nbytes": 655360,
1793
+ "byteOffset": 24264704
1794
+ },
1795
+ {
1796
+ "name": "param_154",
1797
+ "shape": [
1798
+ 256,
1799
+ 2048
1800
+ ],
1801
+ "dtype": "uint32",
1802
+ "format": "raw",
1803
+ "nbytes": 2097152,
1804
+ "byteOffset": 24920064
1805
+ },
1806
+ {
1807
+ "name": "param_155",
1808
+ "shape": [
1809
+ 64,
1810
+ 2048
1811
+ ],
1812
+ "dtype": "uint32",
1813
+ "format": "raw",
1814
+ "nbytes": 524288,
1815
+ "byteOffset": 27017216
1816
+ }
1817
+ ],
1818
+ "md5sum": "3ba07d8b0a7531d8706155b6a619b90b"
1819
+ },
1820
+ {
1821
+ "dataPath": "params_shard_16.bin",
1822
+ "format": "raw-shard",
1823
+ "nbytes": 27541504,
1824
+ "records": [
1825
+ {
1826
+ "name": "param_156",
1827
+ "shape": [
1828
+ 256,
1829
+ 11264
1830
+ ],
1831
+ "dtype": "uint32",
1832
+ "format": "raw",
1833
+ "nbytes": 11534336,
1834
+ "byteOffset": 0
1835
+ },
1836
+ {
1837
+ "name": "param_157",
1838
+ "shape": [
1839
+ 64,
1840
+ 11264
1841
+ ],
1842
+ "dtype": "uint32",
1843
+ "format": "raw",
1844
+ "nbytes": 2883584,
1845
+ "byteOffset": 11534336
1846
+ },
1847
+ {
1848
+ "name": "param_158",
1849
+ "shape": [
1850
+ 704,
1851
+ 2048
1852
+ ],
1853
+ "dtype": "uint32",
1854
+ "format": "raw",
1855
+ "nbytes": 5767168,
1856
+ "byteOffset": 14417920
1857
+ },
1858
+ {
1859
+ "name": "param_159",
1860
+ "shape": [
1861
+ 176,
1862
+ 2048
1863
+ ],
1864
+ "dtype": "uint32",
1865
+ "format": "raw",
1866
+ "nbytes": 1441792,
1867
+ "byteOffset": 20185088
1868
+ },
1869
+ {
1870
+ "name": "param_160",
1871
+ "shape": [
1872
+ 2048
1873
+ ],
1874
+ "dtype": "float32",
1875
+ "format": "raw",
1876
+ "nbytes": 8192,
1877
+ "byteOffset": 21626880
1878
+ },
1879
+ {
1880
+ "name": "param_161",
1881
+ "shape": [
1882
+ 2048
1883
+ ],
1884
+ "dtype": "float32",
1885
+ "format": "raw",
1886
+ "nbytes": 8192,
1887
+ "byteOffset": 21635072
1888
+ },
1889
+ {
1890
+ "name": "param_162",
1891
+ "shape": [
1892
+ 256,
1893
+ 2560
1894
+ ],
1895
+ "dtype": "uint32",
1896
+ "format": "raw",
1897
+ "nbytes": 2621440,
1898
+ "byteOffset": 21643264
1899
+ },
1900
+ {
1901
+ "name": "param_163",
1902
+ "shape": [
1903
+ 64,
1904
+ 2560
1905
+ ],
1906
+ "dtype": "uint32",
1907
+ "format": "raw",
1908
+ "nbytes": 655360,
1909
+ "byteOffset": 24264704
1910
+ },
1911
+ {
1912
+ "name": "param_164",
1913
+ "shape": [
1914
+ 256,
1915
+ 2048
1916
+ ],
1917
+ "dtype": "uint32",
1918
+ "format": "raw",
1919
+ "nbytes": 2097152,
1920
+ "byteOffset": 24920064
1921
+ },
1922
+ {
1923
+ "name": "param_165",
1924
+ "shape": [
1925
+ 64,
1926
+ 2048
1927
+ ],
1928
+ "dtype": "uint32",
1929
+ "format": "raw",
1930
+ "nbytes": 524288,
1931
+ "byteOffset": 27017216
1932
+ }
1933
+ ],
1934
+ "md5sum": "e4634aa4de3450d7d50c58c41ac9e7cc"
1935
+ },
1936
+ {
1937
+ "dataPath": "params_shard_17.bin",
1938
+ "format": "raw-shard",
1939
+ "nbytes": 27541504,
1940
+ "records": [
1941
+ {
1942
+ "name": "param_166",
1943
+ "shape": [
1944
+ 256,
1945
+ 11264
1946
+ ],
1947
+ "dtype": "uint32",
1948
+ "format": "raw",
1949
+ "nbytes": 11534336,
1950
+ "byteOffset": 0
1951
+ },
1952
+ {
1953
+ "name": "param_167",
1954
+ "shape": [
1955
+ 64,
1956
+ 11264
1957
+ ],
1958
+ "dtype": "uint32",
1959
+ "format": "raw",
1960
+ "nbytes": 2883584,
1961
+ "byteOffset": 11534336
1962
+ },
1963
+ {
1964
+ "name": "param_168",
1965
+ "shape": [
1966
+ 704,
1967
+ 2048
1968
+ ],
1969
+ "dtype": "uint32",
1970
+ "format": "raw",
1971
+ "nbytes": 5767168,
1972
+ "byteOffset": 14417920
1973
+ },
1974
+ {
1975
+ "name": "param_169",
1976
+ "shape": [
1977
+ 176,
1978
+ 2048
1979
+ ],
1980
+ "dtype": "uint32",
1981
+ "format": "raw",
1982
+ "nbytes": 1441792,
1983
+ "byteOffset": 20185088
1984
+ },
1985
+ {
1986
+ "name": "param_170",
1987
+ "shape": [
1988
+ 2048
1989
+ ],
1990
+ "dtype": "float32",
1991
+ "format": "raw",
1992
+ "nbytes": 8192,
1993
+ "byteOffset": 21626880
1994
+ },
1995
+ {
1996
+ "name": "param_171",
1997
+ "shape": [
1998
+ 2048
1999
+ ],
2000
+ "dtype": "float32",
2001
+ "format": "raw",
2002
+ "nbytes": 8192,
2003
+ "byteOffset": 21635072
2004
+ },
2005
+ {
2006
+ "name": "param_172",
2007
+ "shape": [
2008
+ 256,
2009
+ 2560
2010
+ ],
2011
+ "dtype": "uint32",
2012
+ "format": "raw",
2013
+ "nbytes": 2621440,
2014
+ "byteOffset": 21643264
2015
+ },
2016
+ {
2017
+ "name": "param_173",
2018
+ "shape": [
2019
+ 64,
2020
+ 2560
2021
+ ],
2022
+ "dtype": "uint32",
2023
+ "format": "raw",
2024
+ "nbytes": 655360,
2025
+ "byteOffset": 24264704
2026
+ },
2027
+ {
2028
+ "name": "param_174",
2029
+ "shape": [
2030
+ 256,
2031
+ 2048
2032
+ ],
2033
+ "dtype": "uint32",
2034
+ "format": "raw",
2035
+ "nbytes": 2097152,
2036
+ "byteOffset": 24920064
2037
+ },
2038
+ {
2039
+ "name": "param_175",
2040
+ "shape": [
2041
+ 64,
2042
+ 2048
2043
+ ],
2044
+ "dtype": "uint32",
2045
+ "format": "raw",
2046
+ "nbytes": 524288,
2047
+ "byteOffset": 27017216
2048
+ }
2049
+ ],
2050
+ "md5sum": "1b3bbe10e98577fdb672158123c360aa"
2051
+ },
2052
+ {
2053
+ "dataPath": "params_shard_18.bin",
2054
+ "format": "raw-shard",
2055
+ "nbytes": 27541504,
2056
+ "records": [
2057
+ {
2058
+ "name": "param_176",
2059
+ "shape": [
2060
+ 256,
2061
+ 11264
2062
+ ],
2063
+ "dtype": "uint32",
2064
+ "format": "raw",
2065
+ "nbytes": 11534336,
2066
+ "byteOffset": 0
2067
+ },
2068
+ {
2069
+ "name": "param_177",
2070
+ "shape": [
2071
+ 64,
2072
+ 11264
2073
+ ],
2074
+ "dtype": "uint32",
2075
+ "format": "raw",
2076
+ "nbytes": 2883584,
2077
+ "byteOffset": 11534336
2078
+ },
2079
+ {
2080
+ "name": "param_178",
2081
+ "shape": [
2082
+ 704,
2083
+ 2048
2084
+ ],
2085
+ "dtype": "uint32",
2086
+ "format": "raw",
2087
+ "nbytes": 5767168,
2088
+ "byteOffset": 14417920
2089
+ },
2090
+ {
2091
+ "name": "param_179",
2092
+ "shape": [
2093
+ 176,
2094
+ 2048
2095
+ ],
2096
+ "dtype": "uint32",
2097
+ "format": "raw",
2098
+ "nbytes": 1441792,
2099
+ "byteOffset": 20185088
2100
+ },
2101
+ {
2102
+ "name": "param_180",
2103
+ "shape": [
2104
+ 2048
2105
+ ],
2106
+ "dtype": "float32",
2107
+ "format": "raw",
2108
+ "nbytes": 8192,
2109
+ "byteOffset": 21626880
2110
+ },
2111
+ {
2112
+ "name": "param_181",
2113
+ "shape": [
2114
+ 2048
2115
+ ],
2116
+ "dtype": "float32",
2117
+ "format": "raw",
2118
+ "nbytes": 8192,
2119
+ "byteOffset": 21635072
2120
+ },
2121
+ {
2122
+ "name": "param_182",
2123
+ "shape": [
2124
+ 256,
2125
+ 2560
2126
+ ],
2127
+ "dtype": "uint32",
2128
+ "format": "raw",
2129
+ "nbytes": 2621440,
2130
+ "byteOffset": 21643264
2131
+ },
2132
+ {
2133
+ "name": "param_183",
2134
+ "shape": [
2135
+ 64,
2136
+ 2560
2137
+ ],
2138
+ "dtype": "uint32",
2139
+ "format": "raw",
2140
+ "nbytes": 655360,
2141
+ "byteOffset": 24264704
2142
+ },
2143
+ {
2144
+ "name": "param_184",
2145
+ "shape": [
2146
+ 256,
2147
+ 2048
2148
+ ],
2149
+ "dtype": "uint32",
2150
+ "format": "raw",
2151
+ "nbytes": 2097152,
2152
+ "byteOffset": 24920064
2153
+ },
2154
+ {
2155
+ "name": "param_185",
2156
+ "shape": [
2157
+ 64,
2158
+ 2048
2159
+ ],
2160
+ "dtype": "uint32",
2161
+ "format": "raw",
2162
+ "nbytes": 524288,
2163
+ "byteOffset": 27017216
2164
+ }
2165
+ ],
2166
+ "md5sum": "9e56e0e0c8d6a636068e85aaede85f61"
2167
+ },
2168
+ {
2169
+ "dataPath": "params_shard_19.bin",
2170
+ "format": "raw-shard",
2171
+ "nbytes": 27541504,
2172
+ "records": [
2173
+ {
2174
+ "name": "param_186",
2175
+ "shape": [
2176
+ 256,
2177
+ 11264
2178
+ ],
2179
+ "dtype": "uint32",
2180
+ "format": "raw",
2181
+ "nbytes": 11534336,
2182
+ "byteOffset": 0
2183
+ },
2184
+ {
2185
+ "name": "param_187",
2186
+ "shape": [
2187
+ 64,
2188
+ 11264
2189
+ ],
2190
+ "dtype": "uint32",
2191
+ "format": "raw",
2192
+ "nbytes": 2883584,
2193
+ "byteOffset": 11534336
2194
+ },
2195
+ {
2196
+ "name": "param_188",
2197
+ "shape": [
2198
+ 704,
2199
+ 2048
2200
+ ],
2201
+ "dtype": "uint32",
2202
+ "format": "raw",
2203
+ "nbytes": 5767168,
2204
+ "byteOffset": 14417920
2205
+ },
2206
+ {
2207
+ "name": "param_189",
2208
+ "shape": [
2209
+ 176,
2210
+ 2048
2211
+ ],
2212
+ "dtype": "uint32",
2213
+ "format": "raw",
2214
+ "nbytes": 1441792,
2215
+ "byteOffset": 20185088
2216
+ },
2217
+ {
2218
+ "name": "param_190",
2219
+ "shape": [
2220
+ 2048
2221
+ ],
2222
+ "dtype": "float32",
2223
+ "format": "raw",
2224
+ "nbytes": 8192,
2225
+ "byteOffset": 21626880
2226
+ },
2227
+ {
2228
+ "name": "param_191",
2229
+ "shape": [
2230
+ 2048
2231
+ ],
2232
+ "dtype": "float32",
2233
+ "format": "raw",
2234
+ "nbytes": 8192,
2235
+ "byteOffset": 21635072
2236
+ },
2237
+ {
2238
+ "name": "param_192",
2239
+ "shape": [
2240
+ 256,
2241
+ 2560
2242
+ ],
2243
+ "dtype": "uint32",
2244
+ "format": "raw",
2245
+ "nbytes": 2621440,
2246
+ "byteOffset": 21643264
2247
+ },
2248
+ {
2249
+ "name": "param_193",
2250
+ "shape": [
2251
+ 64,
2252
+ 2560
2253
+ ],
2254
+ "dtype": "uint32",
2255
+ "format": "raw",
2256
+ "nbytes": 655360,
2257
+ "byteOffset": 24264704
2258
+ },
2259
+ {
2260
+ "name": "param_194",
2261
+ "shape": [
2262
+ 256,
2263
+ 2048
2264
+ ],
2265
+ "dtype": "uint32",
2266
+ "format": "raw",
2267
+ "nbytes": 2097152,
2268
+ "byteOffset": 24920064
2269
+ },
2270
+ {
2271
+ "name": "param_195",
2272
+ "shape": [
2273
+ 64,
2274
+ 2048
2275
+ ],
2276
+ "dtype": "uint32",
2277
+ "format": "raw",
2278
+ "nbytes": 524288,
2279
+ "byteOffset": 27017216
2280
+ }
2281
+ ],
2282
+ "md5sum": "88375119f9863565ede6b8748db4365b"
2283
+ },
2284
+ {
2285
+ "dataPath": "params_shard_20.bin",
2286
+ "format": "raw-shard",
2287
+ "nbytes": 27541504,
2288
+ "records": [
2289
+ {
2290
+ "name": "param_196",
2291
+ "shape": [
2292
+ 256,
2293
+ 11264
2294
+ ],
2295
+ "dtype": "uint32",
2296
+ "format": "raw",
2297
+ "nbytes": 11534336,
2298
+ "byteOffset": 0
2299
+ },
2300
+ {
2301
+ "name": "param_197",
2302
+ "shape": [
2303
+ 64,
2304
+ 11264
2305
+ ],
2306
+ "dtype": "uint32",
2307
+ "format": "raw",
2308
+ "nbytes": 2883584,
2309
+ "byteOffset": 11534336
2310
+ },
2311
+ {
2312
+ "name": "param_198",
2313
+ "shape": [
2314
+ 704,
2315
+ 2048
2316
+ ],
2317
+ "dtype": "uint32",
2318
+ "format": "raw",
2319
+ "nbytes": 5767168,
2320
+ "byteOffset": 14417920
2321
+ },
2322
+ {
2323
+ "name": "param_199",
2324
+ "shape": [
2325
+ 176,
2326
+ 2048
2327
+ ],
2328
+ "dtype": "uint32",
2329
+ "format": "raw",
2330
+ "nbytes": 1441792,
2331
+ "byteOffset": 20185088
2332
+ },
2333
+ {
2334
+ "name": "param_200",
2335
+ "shape": [
2336
+ 2048
2337
+ ],
2338
+ "dtype": "float32",
2339
+ "format": "raw",
2340
+ "nbytes": 8192,
2341
+ "byteOffset": 21626880
2342
+ },
2343
+ {
2344
+ "name": "param_201",
2345
+ "shape": [
2346
+ 2048
2347
+ ],
2348
+ "dtype": "float32",
2349
+ "format": "raw",
2350
+ "nbytes": 8192,
2351
+ "byteOffset": 21635072
2352
+ },
2353
+ {
2354
+ "name": "param_202",
2355
+ "shape": [
2356
+ 256,
2357
+ 2560
2358
+ ],
2359
+ "dtype": "uint32",
2360
+ "format": "raw",
2361
+ "nbytes": 2621440,
2362
+ "byteOffset": 21643264
2363
+ },
2364
+ {
2365
+ "name": "param_203",
2366
+ "shape": [
2367
+ 64,
2368
+ 2560
2369
+ ],
2370
+ "dtype": "uint32",
2371
+ "format": "raw",
2372
+ "nbytes": 655360,
2373
+ "byteOffset": 24264704
2374
+ },
2375
+ {
2376
+ "name": "param_204",
2377
+ "shape": [
2378
+ 256,
2379
+ 2048
2380
+ ],
2381
+ "dtype": "uint32",
2382
+ "format": "raw",
2383
+ "nbytes": 2097152,
2384
+ "byteOffset": 24920064
2385
+ },
2386
+ {
2387
+ "name": "param_205",
2388
+ "shape": [
2389
+ 64,
2390
+ 2048
2391
+ ],
2392
+ "dtype": "uint32",
2393
+ "format": "raw",
2394
+ "nbytes": 524288,
2395
+ "byteOffset": 27017216
2396
+ }
2397
+ ],
2398
+ "md5sum": "d4ea1ab7075b736b0f4e35e7a020161e"
2399
+ },
2400
+ {
2401
+ "dataPath": "params_shard_21.bin",
2402
+ "format": "raw-shard",
2403
+ "nbytes": 27541504,
2404
+ "records": [
2405
+ {
2406
+ "name": "param_206",
2407
+ "shape": [
2408
+ 256,
2409
+ 11264
2410
+ ],
2411
+ "dtype": "uint32",
2412
+ "format": "raw",
2413
+ "nbytes": 11534336,
2414
+ "byteOffset": 0
2415
+ },
2416
+ {
2417
+ "name": "param_207",
2418
+ "shape": [
2419
+ 64,
2420
+ 11264
2421
+ ],
2422
+ "dtype": "uint32",
2423
+ "format": "raw",
2424
+ "nbytes": 2883584,
2425
+ "byteOffset": 11534336
2426
+ },
2427
+ {
2428
+ "name": "param_208",
2429
+ "shape": [
2430
+ 704,
2431
+ 2048
2432
+ ],
2433
+ "dtype": "uint32",
2434
+ "format": "raw",
2435
+ "nbytes": 5767168,
2436
+ "byteOffset": 14417920
2437
+ },
2438
+ {
2439
+ "name": "param_209",
2440
+ "shape": [
2441
+ 176,
2442
+ 2048
2443
+ ],
2444
+ "dtype": "uint32",
2445
+ "format": "raw",
2446
+ "nbytes": 1441792,
2447
+ "byteOffset": 20185088
2448
+ },
2449
+ {
2450
+ "name": "param_210",
2451
+ "shape": [
2452
+ 2048
2453
+ ],
2454
+ "dtype": "float32",
2455
+ "format": "raw",
2456
+ "nbytes": 8192,
2457
+ "byteOffset": 21626880
2458
+ },
2459
+ {
2460
+ "name": "param_211",
2461
+ "shape": [
2462
+ 2048
2463
+ ],
2464
+ "dtype": "float32",
2465
+ "format": "raw",
2466
+ "nbytes": 8192,
2467
+ "byteOffset": 21635072
2468
+ },
2469
+ {
2470
+ "name": "param_212",
2471
+ "shape": [
2472
+ 256,
2473
+ 2560
2474
+ ],
2475
+ "dtype": "uint32",
2476
+ "format": "raw",
2477
+ "nbytes": 2621440,
2478
+ "byteOffset": 21643264
2479
+ },
2480
+ {
2481
+ "name": "param_213",
2482
+ "shape": [
2483
+ 64,
2484
+ 2560
2485
+ ],
2486
+ "dtype": "uint32",
2487
+ "format": "raw",
2488
+ "nbytes": 655360,
2489
+ "byteOffset": 24264704
2490
+ },
2491
+ {
2492
+ "name": "param_214",
2493
+ "shape": [
2494
+ 256,
2495
+ 2048
2496
+ ],
2497
+ "dtype": "uint32",
2498
+ "format": "raw",
2499
+ "nbytes": 2097152,
2500
+ "byteOffset": 24920064
2501
+ },
2502
+ {
2503
+ "name": "param_215",
2504
+ "shape": [
2505
+ 64,
2506
+ 2048
2507
+ ],
2508
+ "dtype": "uint32",
2509
+ "format": "raw",
2510
+ "nbytes": 524288,
2511
+ "byteOffset": 27017216
2512
+ }
2513
+ ],
2514
+ "md5sum": "a8b7995d16937e8f260af256277d293c"
2515
+ },
2516
+ {
2517
+ "dataPath": "params_shard_22.bin",
2518
+ "format": "raw-shard",
2519
+ "nbytes": 32768000,
2520
+ "records": [
2521
+ {
2522
+ "name": "param_223",
2523
+ "shape": [
2524
+ 256,
2525
+ 32000
2526
+ ],
2527
+ "dtype": "uint32",
2528
+ "format": "raw",
2529
+ "nbytes": 32768000,
2530
+ "byteOffset": 0
2531
+ }
2532
+ ],
2533
+ "md5sum": "edbb534655606d1fdd38384804de021d"
2534
+ },
2535
+ {
2536
+ "dataPath": "params_shard_23.bin",
2537
+ "format": "raw-shard",
2538
+ "nbytes": 30892032,
2539
+ "records": [
2540
+ {
2541
+ "name": "param_216",
2542
+ "shape": [
2543
+ 256,
2544
+ 11264
2545
+ ],
2546
+ "dtype": "uint32",
2547
+ "format": "raw",
2548
+ "nbytes": 11534336,
2549
+ "byteOffset": 0
2550
+ },
2551
+ {
2552
+ "name": "param_217",
2553
+ "shape": [
2554
+ 64,
2555
+ 11264
2556
+ ],
2557
+ "dtype": "uint32",
2558
+ "format": "raw",
2559
+ "nbytes": 2883584,
2560
+ "byteOffset": 11534336
2561
+ },
2562
+ {
2563
+ "name": "param_218",
2564
+ "shape": [
2565
+ 704,
2566
+ 2048
2567
+ ],
2568
+ "dtype": "uint32",
2569
+ "format": "raw",
2570
+ "nbytes": 5767168,
2571
+ "byteOffset": 14417920
2572
+ },
2573
+ {
2574
+ "name": "param_219",
2575
+ "shape": [
2576
+ 176,
2577
+ 2048
2578
+ ],
2579
+ "dtype": "uint32",
2580
+ "format": "raw",
2581
+ "nbytes": 1441792,
2582
+ "byteOffset": 20185088
2583
+ },
2584
+ {
2585
+ "name": "param_220",
2586
+ "shape": [
2587
+ 2048
2588
+ ],
2589
+ "dtype": "float32",
2590
+ "format": "raw",
2591
+ "nbytes": 8192,
2592
+ "byteOffset": 21626880
2593
+ },
2594
+ {
2595
+ "name": "param_221",
2596
+ "shape": [
2597
+ 2048
2598
+ ],
2599
+ "dtype": "float32",
2600
+ "format": "raw",
2601
+ "nbytes": 8192,
2602
+ "byteOffset": 21635072
2603
+ },
2604
+ {
2605
+ "name": "param_222",
2606
+ "shape": [
2607
+ 2048
2608
+ ],
2609
+ "dtype": "float32",
2610
+ "format": "raw",
2611
+ "nbytes": 8192,
2612
+ "byteOffset": 21643264
2613
+ },
2614
+ {
2615
+ "name": "param_224",
2616
+ "shape": [
2617
+ 64,
2618
+ 32000
2619
+ ],
2620
+ "dtype": "uint32",
2621
+ "format": "raw",
2622
+ "nbytes": 8192000,
2623
+ "byteOffset": 21651456
2624
+ },
2625
+ {
2626
+ "name": "param_225",
2627
+ "shape": [
2628
+ 2048,
2629
+ 64
2630
+ ],
2631
+ "dtype": "float32",
2632
+ "format": "raw",
2633
+ "nbytes": 524288,
2634
+ "byteOffset": 29843456
2635
+ },
2636
+ {
2637
+ "name": "param_226",
2638
+ "shape": [
2639
+ 2048,
2640
+ 64
2641
+ ],
2642
+ "dtype": "float32",
2643
+ "format": "raw",
2644
+ "nbytes": 524288,
2645
+ "byteOffset": 30367744
2646
+ }
2647
+ ],
2648
+ "md5sum": "3cbd9b5d8f5d6710d43afc037163c3c2"
2649
+ }
2650
+ ]
2651
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb4b3367b7f56d3c97fdb307e02fe40d45023d08cdbd481494ea1aa93a885146
3
+ size 32768000
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e6ef035c972761666f2594c2a73a5ae6e1c04889f9fd3b1716ffb40d9079263
3
+ size 28508160
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa9cfd9b02bc7cd3f4182af1ca3e612d656da8135521867cacd4cc61501ed87c
3
+ size 27541504
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42ec06b20d34fcf7de0840e513294f3ec2b22e60571fa34d8e398c161c5434b2
3
+ size 27541504
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6619445cac1659ae6ac188a8262daca4bd6b38aa4bf3bdce744d9e622d86089
3
+ size 27541504
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bff5d6f093aceb0324eb9cbff706c3de1cf06ac6782812b5ac08c5037bd5f2b4
3
+ size 27541504
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08f2c552fc8f4c01be12059bffe3edd2ba695196c6431d6e7a1c00f283376760
3
+ size 27541504
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78c0f99dd5b88d88bcb523430b75a0677a762be34cc18abc8ef2b8f015365b6f
3
+ size 27541504
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a8c0f6e382fae489a2759cdb6fe8317affb57f031e8d4961ad22a81fbb274a5
3
+ size 27541504
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23dd026acf30e267d3e69d03e27d091c715b7871363dd7595a5a507d4a0e34ae
3
+ size 27541504
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c7dd1b6a7482a3b408d3ba5431e5a3a4e629bf16b8830266ea18d523191c036
3
+ size 27541504
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b2d39304134a11033d9294e4ff7f8ff7d349e87ab2c95acf21bf4add8f5fe90
3
+ size 27541504
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dee2ba9408e285f997d5f2509d03e8f1d7c79f51db1a5085eb59c4d019853fb9
3
+ size 33308672
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d0756adfe38e22f50c468289b855e0395451de43a8c924a0b39740a5bdcb135
3
+ size 27541504
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:715e72b4b337089406836625b736506f918009476e1ca7ed9780ec527c6a4391
3
+ size 27541504
params_shard_22.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eaa56bbfcc65e5cc2d74d3cf377db9b271228a067794e0624f3400d978f088c6
3
+ size 32768000
params_shard_23.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14f254395b2e3192fcd64d23c8c841f7164668e41fd57a15d0bb4f10367b0c2a
3
+ size 30892032
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f937077a2ba2d06de219b5e87b3734c67507bd792a32d364b72b2f04a49fcb0
3
+ size 32276480
params_shard_4.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5fb7829135cb1172dd8db2adbf5213074ef6632d2c4a5fcd43eb95fdf3e8b72
3
+ size 30162944
params_shard_5.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d11843f4d5d7703b05dc0bae6cd9b2393a44f26169c6380a4c7938fc14744e4
3
+ size 27541504
params_shard_6.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7e4ad00b565394083e9cefff9eb2f198e3306c033fe7e45c0c0ef0af3ede492
3
+ size 27541504
params_shard_7.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb7f4ea208c7825c31fb8931df89836953f44e3c0a2277901a7da568e3e6cf8f
3
+ size 27541504
params_shard_8.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ce84e0c926065185c00de0e35fc701bc9b2bdda3e9b05f9118cb6a20e7693f2
3
+ size 27541504
params_shard_9.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:246b4c50615b3e07e9080e889597d4bbaf0d20401b6541aeae367010b6fd7dde
3
+ size 27541504
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "clean_up_tokenization_spaces": false,
32
+ "eos_token": "</s>",
33
+ "legacy": false,
34
+ "model_max_length": 4096,
35
+ "pad_token": "</s>",
36
+ "padding_side": "right",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "unk_token": "<unk>",
41
+ "use_default_system_prompt": false
42
+ }