Hugging Face's logo

mlc-ai
/
gemma-2b-it-q0f32-MLC

MLC-LLM
web-llm
Model card Files Community
CharlieFRuan commited on
Commit
9493b0e
•
1 Parent(s): 38a7d3d

Initial commit

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +1 -0
  2. logs.txt +0 -0
  3. mlc-chat-config.json +45 -0
  4. ndarray-cache-b16.json +1574 -0
  5. ndarray-cache.json +1574 -0
  6. params_shard_0.bin +3 -0
  7. params_shard_1.bin +3 -0
  8. params_shard_10.bin +3 -0
  9. params_shard_11.bin +3 -0
  10. params_shard_12.bin +3 -0
  11. params_shard_13.bin +3 -0
  12. params_shard_14.bin +3 -0
  13. params_shard_15.bin +3 -0
  14. params_shard_16.bin +3 -0
  15. params_shard_17.bin +3 -0
  16. params_shard_18.bin +3 -0
  17. params_shard_19.bin +3 -0
  18. params_shard_2.bin +3 -0
  19. params_shard_20.bin +3 -0
  20. params_shard_21.bin +3 -0
  21. params_shard_22.bin +3 -0
  22. params_shard_23.bin +3 -0
  23. params_shard_24.bin +3 -0
  24. params_shard_25.bin +3 -0
  25. params_shard_26.bin +3 -0
  26. params_shard_27.bin +3 -0
  27. params_shard_28.bin +3 -0
  28. params_shard_29.bin +3 -0
  29. params_shard_3.bin +3 -0
  30. params_shard_30.bin +3 -0
  31. params_shard_31.bin +3 -0
  32. params_shard_32.bin +3 -0
  33. params_shard_33.bin +3 -0
  34. params_shard_34.bin +3 -0
  35. params_shard_35.bin +3 -0
  36. params_shard_36.bin +3 -0
  37. params_shard_37.bin +3 -0
  38. params_shard_38.bin +3 -0
  39. params_shard_39.bin +3 -0
  40. params_shard_4.bin +3 -0
  41. params_shard_40.bin +3 -0
  42. params_shard_41.bin +3 -0
  43. params_shard_42.bin +3 -0
  44. params_shard_43.bin +3 -0
  45. params_shard_44.bin +3 -0
  46. params_shard_45.bin +3 -0
  47. params_shard_46.bin +3 -0
  48. params_shard_47.bin +3 -0
  49. params_shard_48.bin +3 -0
  50. params_shard_5.bin +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
logs.txt ADDED
The diff for this file is too large to render. See raw diff
 
mlc-chat-config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_type": "gemma",
3
+ "quantization": "q0f32",
4
+ "model_config": {
5
+ "hidden_size": 2048,
6
+ "hidden_act": "gelu",
7
+ "intermediate_size": 16384,
8
+ "attention_bias": false,
9
+ "num_attention_heads": 8,
10
+ "num_key_value_heads": 1,
11
+ "head_dim": 256,
12
+ "num_hidden_layers": 18,
13
+ "rms_norm_eps": 1e-06,
14
+ "vocab_size": 256000,
15
+ "position_embedding_base": 10000.0,
16
+ "context_window_size": 8192,
17
+ "prefill_chunk_size": 1024,
18
+ "tensor_parallel_shards": 1,
19
+ "max_batch_size": 80
20
+ },
21
+ "vocab_size": 256000,
22
+ "context_window_size": 8192,
23
+ "sliding_window_size": -1,
24
+ "prefill_chunk_size": 1024,
25
+ "attention_sink_size": -1,
26
+ "tensor_parallel_shards": 1,
27
+ "mean_gen_len": 128,
28
+ "max_gen_len": 512,
29
+ "shift_fill_factor": 0.3,
30
+ "temperature": 0.7,
31
+ "presence_penalty": 0.0,
32
+ "frequency_penalty": 0.0,
33
+ "repetition_penalty": 1.0,
34
+ "top_p": 0.95,
35
+ "conv_template": "gemma_instruction",
36
+ "pad_token_id": 0,
37
+ "bos_token_id": 2,
38
+ "eos_token_id": 1,
39
+ "tokenizer_files": [
40
+ "tokenizer.model",
41
+ "tokenizer.json",
42
+ "tokenizer_config.json"
43
+ ],
44
+ "version": "0.1.0"
45
+ }
ndarray-cache-b16.json ADDED
@@ -0,0 +1,1574 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 110,
4
+ "ParamBytes": 10024689664.0,
5
+ "BitsPerParam": 32.0
6
+ },
7
+ "records": [
8
+ {
9
+ "dataPath": "params_shard_0.bin",
10
+ "format": "raw-shard",
11
+ "nbytes": 1048576000,
12
+ "records": [
13
+ {
14
+ "name": "model.embed_tokens.weight",
15
+ "shape": [
16
+ 256000,
17
+ 2048
18
+ ],
19
+ "dtype": "bfloat16",
20
+ "format": "raw",
21
+ "nbytes": 1048576000,
22
+ "byteOffset": 0
23
+ }
24
+ ],
25
+ "md5sum": "21128c642a829dff80b86de466616d9d"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 67108864,
31
+ "records": [
32
+ {
33
+ "name": "model.layers.0.mlp.down_proj.weight",
34
+ "shape": [
35
+ 2048,
36
+ 16384
37
+ ],
38
+ "dtype": "bfloat16",
39
+ "format": "raw",
40
+ "nbytes": 67108864,
41
+ "byteOffset": 0
42
+ }
43
+ ],
44
+ "md5sum": "e3daa1938df1d775e3448e4fdd33527e"
45
+ },
46
+ {
47
+ "dataPath": "params_shard_2.bin",
48
+ "format": "raw-shard",
49
+ "nbytes": 134217728,
50
+ "records": [
51
+ {
52
+ "name": "model.layers.0.mlp.gate_up_proj.weight",
53
+ "shape": [
54
+ 32768,
55
+ 2048
56
+ ],
57
+ "dtype": "bfloat16",
58
+ "format": "raw",
59
+ "nbytes": 134217728,
60
+ "byteOffset": 0
61
+ }
62
+ ],
63
+ "md5sum": "5ebf170cde97e9025d2c3473b8357c13"
64
+ },
65
+ {
66
+ "dataPath": "params_shard_3.bin",
67
+ "format": "raw-shard",
68
+ "nbytes": 67108864,
69
+ "records": [
70
+ {
71
+ "name": "model.layers.1.mlp.down_proj.weight",
72
+ "shape": [
73
+ 2048,
74
+ 16384
75
+ ],
76
+ "dtype": "bfloat16",
77
+ "format": "raw",
78
+ "nbytes": 67108864,
79
+ "byteOffset": 0
80
+ }
81
+ ],
82
+ "md5sum": "b11672a46d1224e14cd2a5c8590a02ea"
83
+ },
84
+ {
85
+ "dataPath": "params_shard_4.bin",
86
+ "format": "raw-shard",
87
+ "nbytes": 134217728,
88
+ "records": [
89
+ {
90
+ "name": "model.layers.1.mlp.gate_up_proj.weight",
91
+ "shape": [
92
+ 32768,
93
+ 2048
94
+ ],
95
+ "dtype": "bfloat16",
96
+ "format": "raw",
97
+ "nbytes": 134217728,
98
+ "byteOffset": 0
99
+ }
100
+ ],
101
+ "md5sum": "9970e7f9c5a4be1cb1acbb075fe39a60"
102
+ },
103
+ {
104
+ "dataPath": "params_shard_5.bin",
105
+ "format": "raw-shard",
106
+ "nbytes": 29376512,
107
+ "records": [
108
+ {
109
+ "name": "model.layers.0.input_layernorm.weight",
110
+ "shape": [
111
+ 2048
112
+ ],
113
+ "dtype": "bfloat16",
114
+ "format": "raw",
115
+ "nbytes": 4096,
116
+ "byteOffset": 0
117
+ },
118
+ {
119
+ "name": "model.layers.0.post_attention_layernorm.weight",
120
+ "shape": [
121
+ 2048
122
+ ],
123
+ "dtype": "bfloat16",
124
+ "format": "raw",
125
+ "nbytes": 4096,
126
+ "byteOffset": 4096
127
+ },
128
+ {
129
+ "name": "model.layers.0.self_attn.qkv_proj.weight",
130
+ "shape": [
131
+ 2560,
132
+ 2048
133
+ ],
134
+ "dtype": "bfloat16",
135
+ "format": "raw",
136
+ "nbytes": 10485760,
137
+ "byteOffset": 8192
138
+ },
139
+ {
140
+ "name": "model.layers.0.self_attn.o_proj.weight",
141
+ "shape": [
142
+ 2048,
143
+ 2048
144
+ ],
145
+ "dtype": "bfloat16",
146
+ "format": "raw",
147
+ "nbytes": 8388608,
148
+ "byteOffset": 10493952
149
+ },
150
+ {
151
+ "name": "model.layers.1.input_layernorm.weight",
152
+ "shape": [
153
+ 2048
154
+ ],
155
+ "dtype": "bfloat16",
156
+ "format": "raw",
157
+ "nbytes": 4096,
158
+ "byteOffset": 18882560
159
+ },
160
+ {
161
+ "name": "model.layers.1.post_attention_layernorm.weight",
162
+ "shape": [
163
+ 2048
164
+ ],
165
+ "dtype": "bfloat16",
166
+ "format": "raw",
167
+ "nbytes": 4096,
168
+ "byteOffset": 18886656
169
+ },
170
+ {
171
+ "name": "model.layers.1.self_attn.qkv_proj.weight",
172
+ "shape": [
173
+ 2560,
174
+ 2048
175
+ ],
176
+ "dtype": "bfloat16",
177
+ "format": "raw",
178
+ "nbytes": 10485760,
179
+ "byteOffset": 18890752
180
+ }
181
+ ],
182
+ "md5sum": "690bd198a683ab8104715e9fa33e3eaf"
183
+ },
184
+ {
185
+ "dataPath": "params_shard_6.bin",
186
+ "format": "raw-shard",
187
+ "nbytes": 67108864,
188
+ "records": [
189
+ {
190
+ "name": "model.layers.10.mlp.down_proj.weight",
191
+ "shape": [
192
+ 2048,
193
+ 16384
194
+ ],
195
+ "dtype": "bfloat16",
196
+ "format": "raw",
197
+ "nbytes": 67108864,
198
+ "byteOffset": 0
199
+ }
200
+ ],
201
+ "md5sum": "71c4717b75b1547985ea79fb4451908d"
202
+ },
203
+ {
204
+ "dataPath": "params_shard_7.bin",
205
+ "format": "raw-shard",
206
+ "nbytes": 134217728,
207
+ "records": [
208
+ {
209
+ "name": "model.layers.10.mlp.gate_up_proj.weight",
210
+ "shape": [
211
+ 32768,
212
+ 2048
213
+ ],
214
+ "dtype": "bfloat16",
215
+ "format": "raw",
216
+ "nbytes": 134217728,
217
+ "byteOffset": 0
218
+ }
219
+ ],
220
+ "md5sum": "0385b2a5124f03324c0cbbf8f8afb4d5"
221
+ },
222
+ {
223
+ "dataPath": "params_shard_8.bin",
224
+ "format": "raw-shard",
225
+ "nbytes": 67108864,
226
+ "records": [
227
+ {
228
+ "name": "model.layers.11.mlp.down_proj.weight",
229
+ "shape": [
230
+ 2048,
231
+ 16384
232
+ ],
233
+ "dtype": "bfloat16",
234
+ "format": "raw",
235
+ "nbytes": 67108864,
236
+ "byteOffset": 0
237
+ }
238
+ ],
239
+ "md5sum": "9d7bbb224840a3c5f24b33f029fc8d3d"
240
+ },
241
+ {
242
+ "dataPath": "params_shard_9.bin",
243
+ "format": "raw-shard",
244
+ "nbytes": 134217728,
245
+ "records": [
246
+ {
247
+ "name": "model.layers.11.mlp.gate_up_proj.weight",
248
+ "shape": [
249
+ 32768,
250
+ 2048
251
+ ],
252
+ "dtype": "bfloat16",
253
+ "format": "raw",
254
+ "nbytes": 134217728,
255
+ "byteOffset": 0
256
+ }
257
+ ],
258
+ "md5sum": "babd2c92010bfd919b25fbd9daaaa8d9"
259
+ },
260
+ {
261
+ "dataPath": "params_shard_10.bin",
262
+ "format": "raw-shard",
263
+ "nbytes": 27279360,
264
+ "records": [
265
+ {
266
+ "name": "model.layers.1.self_attn.o_proj.weight",
267
+ "shape": [
268
+ 2048,
269
+ 2048
270
+ ],
271
+ "dtype": "bfloat16",
272
+ "format": "raw",
273
+ "nbytes": 8388608,
274
+ "byteOffset": 0
275
+ },
276
+ {
277
+ "name": "model.layers.10.input_layernorm.weight",
278
+ "shape": [
279
+ 2048
280
+ ],
281
+ "dtype": "bfloat16",
282
+ "format": "raw",
283
+ "nbytes": 4096,
284
+ "byteOffset": 8388608
285
+ },
286
+ {
287
+ "name": "model.layers.10.post_attention_layernorm.weight",
288
+ "shape": [
289
+ 2048
290
+ ],
291
+ "dtype": "bfloat16",
292
+ "format": "raw",
293
+ "nbytes": 4096,
294
+ "byteOffset": 8392704
295
+ },
296
+ {
297
+ "name": "model.layers.10.self_attn.qkv_proj.weight",
298
+ "shape": [
299
+ 2560,
300
+ 2048
301
+ ],
302
+ "dtype": "bfloat16",
303
+ "format": "raw",
304
+ "nbytes": 10485760,
305
+ "byteOffset": 8396800
306
+ },
307
+ {
308
+ "name": "model.layers.10.self_attn.o_proj.weight",
309
+ "shape": [
310
+ 2048,
311
+ 2048
312
+ ],
313
+ "dtype": "bfloat16",
314
+ "format": "raw",
315
+ "nbytes": 8388608,
316
+ "byteOffset": 18882560
317
+ },
318
+ {
319
+ "name": "model.layers.11.input_layernorm.weight",
320
+ "shape": [
321
+ 2048
322
+ ],
323
+ "dtype": "bfloat16",
324
+ "format": "raw",
325
+ "nbytes": 4096,
326
+ "byteOffset": 27271168
327
+ },
328
+ {
329
+ "name": "model.layers.11.post_attention_layernorm.weight",
330
+ "shape": [
331
+ 2048
332
+ ],
333
+ "dtype": "bfloat16",
334
+ "format": "raw",
335
+ "nbytes": 4096,
336
+ "byteOffset": 27275264
337
+ }
338
+ ],
339
+ "md5sum": "73e3bd854f6acf1969dae0923e8bb11b"
340
+ },
341
+ {
342
+ "dataPath": "params_shard_11.bin",
343
+ "format": "raw-shard",
344
+ "nbytes": 67108864,
345
+ "records": [
346
+ {
347
+ "name": "model.layers.12.mlp.down_proj.weight",
348
+ "shape": [
349
+ 2048,
350
+ 16384
351
+ ],
352
+ "dtype": "bfloat16",
353
+ "format": "raw",
354
+ "nbytes": 67108864,
355
+ "byteOffset": 0
356
+ }
357
+ ],
358
+ "md5sum": "113f6d9072ff3bd46cb356305c3d6435"
359
+ },
360
+ {
361
+ "dataPath": "params_shard_12.bin",
362
+ "format": "raw-shard",
363
+ "nbytes": 134217728,
364
+ "records": [
365
+ {
366
+ "name": "model.layers.12.mlp.gate_up_proj.weight",
367
+ "shape": [
368
+ 32768,
369
+ 2048
370
+ ],
371
+ "dtype": "bfloat16",
372
+ "format": "raw",
373
+ "nbytes": 134217728,
374
+ "byteOffset": 0
375
+ }
376
+ ],
377
+ "md5sum": "2c179b332604d046c25fdb8fa7e8a38d"
378
+ },
379
+ {
380
+ "dataPath": "params_shard_13.bin",
381
+ "format": "raw-shard",
382
+ "nbytes": 29368320,
383
+ "records": [
384
+ {
385
+ "name": "model.layers.11.self_attn.qkv_proj.weight",
386
+ "shape": [
387
+ 2560,
388
+ 2048
389
+ ],
390
+ "dtype": "bfloat16",
391
+ "format": "raw",
392
+ "nbytes": 10485760,
393
+ "byteOffset": 0
394
+ },
395
+ {
396
+ "name": "model.layers.11.self_attn.o_proj.weight",
397
+ "shape": [
398
+ 2048,
399
+ 2048
400
+ ],
401
+ "dtype": "bfloat16",
402
+ "format": "raw",
403
+ "nbytes": 8388608,
404
+ "byteOffset": 10485760
405
+ },
406
+ {
407
+ "name": "model.layers.12.input_layernorm.weight",
408
+ "shape": [
409
+ 2048
410
+ ],
411
+ "dtype": "bfloat16",
412
+ "format": "raw",
413
+ "nbytes": 4096,
414
+ "byteOffset": 18874368
415
+ },
416
+ {
417
+ "name": "model.layers.12.post_attention_layernorm.weight",
418
+ "shape": [
419
+ 2048
420
+ ],
421
+ "dtype": "bfloat16",
422
+ "format": "raw",
423
+ "nbytes": 4096,
424
+ "byteOffset": 18878464
425
+ },
426
+ {
427
+ "name": "model.layers.12.self_attn.qkv_proj.weight",
428
+ "shape": [
429
+ 2560,
430
+ 2048
431
+ ],
432
+ "dtype": "bfloat16",
433
+ "format": "raw",
434
+ "nbytes": 10485760,
435
+ "byteOffset": 18882560
436
+ }
437
+ ],
438
+ "md5sum": "b155622e091c8a222b1f4de1814cbb49"
439
+ },
440
+ {
441
+ "dataPath": "params_shard_14.bin",
442
+ "format": "raw-shard",
443
+ "nbytes": 67108864,
444
+ "records": [
445
+ {
446
+ "name": "model.layers.13.mlp.down_proj.weight",
447
+ "shape": [
448
+ 2048,
449
+ 16384
450
+ ],
451
+ "dtype": "bfloat16",
452
+ "format": "raw",
453
+ "nbytes": 67108864,
454
+ "byteOffset": 0
455
+ }
456
+ ],
457
+ "md5sum": "fb8e2063f991dd07a06fe38ea7365789"
458
+ },
459
+ {
460
+ "dataPath": "params_shard_15.bin",
461
+ "format": "raw-shard",
462
+ "nbytes": 134217728,
463
+ "records": [
464
+ {
465
+ "name": "model.layers.13.mlp.gate_up_proj.weight",
466
+ "shape": [
467
+ 32768,
468
+ 2048
469
+ ],
470
+ "dtype": "bfloat16",
471
+ "format": "raw",
472
+ "nbytes": 134217728,
473
+ "byteOffset": 0
474
+ }
475
+ ],
476
+ "md5sum": "9d5ceb3198e4924fccc31e3f2421ff14"
477
+ },
478
+ {
479
+ "dataPath": "params_shard_16.bin",
480
+ "format": "raw-shard",
481
+ "nbytes": 67108864,
482
+ "records": [
483
+ {
484
+ "name": "model.layers.14.mlp.down_proj.weight",
485
+ "shape": [
486
+ 2048,
487
+ 16384
488
+ ],
489
+ "dtype": "bfloat16",
490
+ "format": "raw",
491
+ "nbytes": 67108864,
492
+ "byteOffset": 0
493
+ }
494
+ ],
495
+ "md5sum": "9fb12a797891bebed74f0e65eefb538b"
496
+ },
497
+ {
498
+ "dataPath": "params_shard_17.bin",
499
+ "format": "raw-shard",
500
+ "nbytes": 134217728,
501
+ "records": [
502
+ {
503
+ "name": "model.layers.14.mlp.gate_up_proj.weight",
504
+ "shape": [
505
+ 32768,
506
+ 2048
507
+ ],
508
+ "dtype": "bfloat16",
509
+ "format": "raw",
510
+ "nbytes": 134217728,
511
+ "byteOffset": 0
512
+ }
513
+ ],
514
+ "md5sum": "59630bc8e808b139cf23c4a4d048a81f"
515
+ },
516
+ {
517
+ "dataPath": "params_shard_18.bin",
518
+ "format": "raw-shard",
519
+ "nbytes": 27279360,
520
+ "records": [
521
+ {
522
+ "name": "model.layers.12.self_attn.o_proj.weight",
523
+ "shape": [
524
+ 2048,
525
+ 2048
526
+ ],
527
+ "dtype": "bfloat16",
528
+ "format": "raw",
529
+ "nbytes": 8388608,
530
+ "byteOffset": 0
531
+ },
532
+ {
533
+ "name": "model.layers.13.input_layernorm.weight",
534
+ "shape": [
535
+ 2048
536
+ ],
537
+ "dtype": "bfloat16",
538
+ "format": "raw",
539
+ "nbytes": 4096,
540
+ "byteOffset": 8388608
541
+ },
542
+ {
543
+ "name": "model.layers.13.post_attention_layernorm.weight",
544
+ "shape": [
545
+ 2048
546
+ ],
547
+ "dtype": "bfloat16",
548
+ "format": "raw",
549
+ "nbytes": 4096,
550
+ "byteOffset": 8392704
551
+ },
552
+ {
553
+ "name": "model.layers.13.self_attn.qkv_proj.weight",
554
+ "shape": [
555
+ 2560,
556
+ 2048
557
+ ],
558
+ "dtype": "bfloat16",
559
+ "format": "raw",
560
+ "nbytes": 10485760,
561
+ "byteOffset": 8396800
562
+ },
563
+ {
564
+ "name": "model.layers.13.self_attn.o_proj.weight",
565
+ "shape": [
566
+ 2048,
567
+ 2048
568
+ ],
569
+ "dtype": "bfloat16",
570
+ "format": "raw",
571
+ "nbytes": 8388608,
572
+ "byteOffset": 18882560
573
+ },
574
+ {
575
+ "name": "model.layers.14.input_layernorm.weight",
576
+ "shape": [
577
+ 2048
578
+ ],
579
+ "dtype": "bfloat16",
580
+ "format": "raw",
581
+ "nbytes": 4096,
582
+ "byteOffset": 27271168
583
+ },
584
+ {
585
+ "name": "model.layers.14.post_attention_layernorm.weight",
586
+ "shape": [
587
+ 2048
588
+ ],
589
+ "dtype": "bfloat16",
590
+ "format": "raw",
591
+ "nbytes": 4096,
592
+ "byteOffset": 27275264
593
+ }
594
+ ],
595
+ "md5sum": "17759e68d24e2eb936f9129a722522f5"
596
+ },
597
+ {
598
+ "dataPath": "params_shard_19.bin",
599
+ "format": "raw-shard",
600
+ "nbytes": 67108864,
601
+ "records": [
602
+ {
603
+ "name": "model.layers.15.mlp.down_proj.weight",
604
+ "shape": [
605
+ 2048,
606
+ 16384
607
+ ],
608
+ "dtype": "bfloat16",
609
+ "format": "raw",
610
+ "nbytes": 67108864,
611
+ "byteOffset": 0
612
+ }
613
+ ],
614
+ "md5sum": "321ba4846c90abaa26d7179a514625dd"
615
+ },
616
+ {
617
+ "dataPath": "params_shard_20.bin",
618
+ "format": "raw-shard",
619
+ "nbytes": 134217728,
620
+ "records": [
621
+ {
622
+ "name": "model.layers.15.mlp.gate_up_proj.weight",
623
+ "shape": [
624
+ 32768,
625
+ 2048
626
+ ],
627
+ "dtype": "bfloat16",
628
+ "format": "raw",
629
+ "nbytes": 134217728,
630
+ "byteOffset": 0
631
+ }
632
+ ],
633
+ "md5sum": "f8665cbb955f5b9470e4fb4f2bebfd3a"
634
+ },
635
+ {
636
+ "dataPath": "params_shard_21.bin",
637
+ "format": "raw-shard",
638
+ "nbytes": 29368320,
639
+ "records": [
640
+ {
641
+ "name": "model.layers.14.self_attn.qkv_proj.weight",
642
+ "shape": [
643
+ 2560,
644
+ 2048
645
+ ],
646
+ "dtype": "bfloat16",
647
+ "format": "raw",
648
+ "nbytes": 10485760,
649
+ "byteOffset": 0
650
+ },
651
+ {
652
+ "name": "model.layers.14.self_attn.o_proj.weight",
653
+ "shape": [
654
+ 2048,
655
+ 2048
656
+ ],
657
+ "dtype": "bfloat16",
658
+ "format": "raw",
659
+ "nbytes": 8388608,
660
+ "byteOffset": 10485760
661
+ },
662
+ {
663
+ "name": "model.layers.15.input_layernorm.weight",
664
+ "shape": [
665
+ 2048
666
+ ],
667
+ "dtype": "bfloat16",
668
+ "format": "raw",
669
+ "nbytes": 4096,
670
+ "byteOffset": 18874368
671
+ },
672
+ {
673
+ "name": "model.layers.15.post_attention_layernorm.weight",
674
+ "shape": [
675
+ 2048
676
+ ],
677
+ "dtype": "bfloat16",
678
+ "format": "raw",
679
+ "nbytes": 4096,
680
+ "byteOffset": 18878464
681
+ },
682
+ {
683
+ "name": "model.layers.15.self_attn.qkv_proj.weight",
684
+ "shape": [
685
+ 2560,
686
+ 2048
687
+ ],
688
+ "dtype": "bfloat16",
689
+ "format": "raw",
690
+ "nbytes": 10485760,
691
+ "byteOffset": 18882560
692
+ }
693
+ ],
694
+ "md5sum": "1d686c17f8ed473437ce9dbb37f66f5a"
695
+ },
696
+ {
697
+ "dataPath": "params_shard_22.bin",
698
+ "format": "raw-shard",
699
+ "nbytes": 67108864,
700
+ "records": [
701
+ {
702
+ "name": "model.layers.16.mlp.down_proj.weight",
703
+ "shape": [
704
+ 2048,
705
+ 16384
706
+ ],
707
+ "dtype": "bfloat16",
708
+ "format": "raw",
709
+ "nbytes": 67108864,
710
+ "byteOffset": 0
711
+ }
712
+ ],
713
+ "md5sum": "2218e355df7616289c17f7cf571c21e4"
714
+ },
715
+ {
716
+ "dataPath": "params_shard_23.bin",
717
+ "format": "raw-shard",
718
+ "nbytes": 134217728,
719
+ "records": [
720
+ {
721
+ "name": "model.layers.16.mlp.gate_up_proj.weight",
722
+ "shape": [
723
+ 32768,
724
+ 2048
725
+ ],
726
+ "dtype": "bfloat16",
727
+ "format": "raw",
728
+ "nbytes": 134217728,
729
+ "byteOffset": 0
730
+ }
731
+ ],
732
+ "md5sum": "5ecbb6372848f97fae1bdef305c623d0"
733
+ },
734
+ {
735
+ "dataPath": "params_shard_24.bin",
736
+ "format": "raw-shard",
737
+ "nbytes": 134217728,
738
+ "records": [
739
+ {
740
+ "name": "model.layers.17.mlp.gate_up_proj.weight",
741
+ "shape": [
742
+ 32768,
743
+ 2048
744
+ ],
745
+ "dtype": "bfloat16",
746
+ "format": "raw",
747
+ "nbytes": 134217728,
748
+ "byteOffset": 0
749
+ }
750
+ ],
751
+ "md5sum": "466ae14bce7c0f6efa9c939f132ea082"
752
+ },
753
+ {
754
+ "dataPath": "params_shard_25.bin",
755
+ "format": "raw-shard",
756
+ "nbytes": 27271168,
757
+ "records": [
758
+ {
759
+ "name": "model.layers.15.self_attn.o_proj.weight",
760
+ "shape": [
761
+ 2048,
762
+ 2048
763
+ ],
764
+ "dtype": "bfloat16",
765
+ "format": "raw",
766
+ "nbytes": 8388608,
767
+ "byteOffset": 0
768
+ },
769
+ {
770
+ "name": "model.layers.16.input_layernorm.weight",
771
+ "shape": [
772
+ 2048
773
+ ],
774
+ "dtype": "bfloat16",
775
+ "format": "raw",
776
+ "nbytes": 4096,
777
+ "byteOffset": 8388608
778
+ },
779
+ {
780
+ "name": "model.layers.16.post_attention_layernorm.weight",
781
+ "shape": [
782
+ 2048
783
+ ],
784
+ "dtype": "bfloat16",
785
+ "format": "raw",
786
+ "nbytes": 4096,
787
+ "byteOffset": 8392704
788
+ },
789
+ {
790
+ "name": "model.layers.16.self_attn.qkv_proj.weight",
791
+ "shape": [
792
+ 2560,
793
+ 2048
794
+ ],
795
+ "dtype": "bfloat16",
796
+ "format": "raw",
797
+ "nbytes": 10485760,
798
+ "byteOffset": 8396800
799
+ },
800
+ {
801
+ "name": "model.layers.16.self_attn.o_proj.weight",
802
+ "shape": [
803
+ 2048,
804
+ 2048
805
+ ],
806
+ "dtype": "bfloat16",
807
+ "format": "raw",
808
+ "nbytes": 8388608,
809
+ "byteOffset": 18882560
810
+ }
811
+ ],
812
+ "md5sum": "517de03093360d322e5b3152bd5c1ef0"
813
+ },
814
+ {
815
+ "dataPath": "params_shard_26.bin",
816
+ "format": "raw-shard",
817
+ "nbytes": 67108864,
818
+ "records": [
819
+ {
820
+ "name": "model.layers.2.mlp.down_proj.weight",
821
+ "shape": [
822
+ 2048,
823
+ 16384
824
+ ],
825
+ "dtype": "bfloat16",
826
+ "format": "raw",
827
+ "nbytes": 67108864,
828
+ "byteOffset": 0
829
+ }
830
+ ],
831
+ "md5sum": "7a6c37d60d558c0a9d650a59c4fc5c68"
832
+ },
833
+ {
834
+ "dataPath": "params_shard_27.bin",
835
+ "format": "raw-shard",
836
+ "nbytes": 134217728,
837
+ "records": [
838
+ {
839
+ "name": "model.layers.2.mlp.gate_up_proj.weight",
840
+ "shape": [
841
+ 32768,
842
+ 2048
843
+ ],
844
+ "dtype": "bfloat16",
845
+ "format": "raw",
846
+ "nbytes": 134217728,
847
+ "byteOffset": 0
848
+ }
849
+ ],
850
+ "md5sum": "1f51d337520bbbf5a56530c60e6b4e2b"
851
+ },
852
+ {
853
+ "dataPath": "params_shard_28.bin",
854
+ "format": "raw-shard",
855
+ "nbytes": 29368320,
856
+ "records": [
857
+ {
858
+ "name": "model.layers.17.self_attn.qkv_proj.weight",
859
+ "shape": [
860
+ 2560,
861
+ 2048
862
+ ],
863
+ "dtype": "bfloat16",
864
+ "format": "raw",
865
+ "nbytes": 10485760,
866
+ "byteOffset": 0
867
+ },
868
+ {
869
+ "name": "model.layers.17.self_attn.o_proj.weight",
870
+ "shape": [
871
+ 2048,
872
+ 2048
873
+ ],
874
+ "dtype": "bfloat16",
875
+ "format": "raw",
876
+ "nbytes": 8388608,
877
+ "byteOffset": 10485760
878
+ },
879
+ {
880
+ "name": "model.layers.2.input_layernorm.weight",
881
+ "shape": [
882
+ 2048
883
+ ],
884
+ "dtype": "bfloat16",
885
+ "format": "raw",
886
+ "nbytes": 4096,
887
+ "byteOffset": 18874368
888
+ },
889
+ {
890
+ "name": "model.layers.2.post_attention_layernorm.weight",
891
+ "shape": [
892
+ 2048
893
+ ],
894
+ "dtype": "bfloat16",
895
+ "format": "raw",
896
+ "nbytes": 4096,
897
+ "byteOffset": 18878464
898
+ },
899
+ {
900
+ "name": "model.layers.2.self_attn.qkv_proj.weight",
901
+ "shape": [
902
+ 2560,
903
+ 2048
904
+ ],
905
+ "dtype": "bfloat16",
906
+ "format": "raw",
907
+ "nbytes": 10485760,
908
+ "byteOffset": 18882560
909
+ }
910
+ ],
911
+ "md5sum": "81402425b99d3caf95d4a94ea8616bf0"
912
+ },
913
+ {
914
+ "dataPath": "params_shard_29.bin",
915
+ "format": "raw-shard",
916
+ "nbytes": 67108864,
917
+ "records": [
918
+ {
919
+ "name": "model.layers.3.mlp.down_proj.weight",
920
+ "shape": [
921
+ 2048,
922
+ 16384
923
+ ],
924
+ "dtype": "bfloat16",
925
+ "format": "raw",
926
+ "nbytes": 67108864,
927
+ "byteOffset": 0
928
+ }
929
+ ],
930
+ "md5sum": "c3300e44f9c90f83867efd3d6aef1d8d"
931
+ },
932
+ {
933
+ "dataPath": "params_shard_30.bin",
934
+ "format": "raw-shard",
935
+ "nbytes": 134217728,
936
+ "records": [
937
+ {
938
+ "name": "model.layers.3.mlp.gate_up_proj.weight",
939
+ "shape": [
940
+ 32768,
941
+ 2048
942
+ ],
943
+ "dtype": "bfloat16",
944
+ "format": "raw",
945
+ "nbytes": 134217728,
946
+ "byteOffset": 0
947
+ }
948
+ ],
949
+ "md5sum": "538c26b83e0c2ab4bdf48cd10df6bd53"
950
+ },
951
+ {
952
+ "dataPath": "params_shard_31.bin",
953
+ "format": "raw-shard",
954
+ "nbytes": 67108864,
955
+ "records": [
956
+ {
957
+ "name": "model.layers.4.mlp.down_proj.weight",
958
+ "shape": [
959
+ 2048,
960
+ 16384
961
+ ],
962
+ "dtype": "bfloat16",
963
+ "format": "raw",
964
+ "nbytes": 67108864,
965
+ "byteOffset": 0
966
+ }
967
+ ],
968
+ "md5sum": "88501c646188da831d14676a256ac84f"
969
+ },
970
+ {
971
+ "dataPath": "params_shard_32.bin",
972
+ "format": "raw-shard",
973
+ "nbytes": 134217728,
974
+ "records": [
975
+ {
976
+ "name": "model.layers.4.mlp.gate_up_proj.weight",
977
+ "shape": [
978
+ 32768,
979
+ 2048
980
+ ],
981
+ "dtype": "bfloat16",
982
+ "format": "raw",
983
+ "nbytes": 134217728,
984
+ "byteOffset": 0
985
+ }
986
+ ],
987
+ "md5sum": "074d0db9b1d459438705b37b6670c06b"
988
+ },
989
+ {
990
+ "dataPath": "params_shard_33.bin",
991
+ "format": "raw-shard",
992
+ "nbytes": 27279360,
993
+ "records": [
994
+ {
995
+ "name": "model.layers.2.self_attn.o_proj.weight",
996
+ "shape": [
997
+ 2048,
998
+ 2048
999
+ ],
1000
+ "dtype": "bfloat16",
1001
+ "format": "raw",
1002
+ "nbytes": 8388608,
1003
+ "byteOffset": 0
1004
+ },
1005
+ {
1006
+ "name": "model.layers.3.input_layernorm.weight",
1007
+ "shape": [
1008
+ 2048
1009
+ ],
1010
+ "dtype": "bfloat16",
1011
+ "format": "raw",
1012
+ "nbytes": 4096,
1013
+ "byteOffset": 8388608
1014
+ },
1015
+ {
1016
+ "name": "model.layers.3.post_attention_layernorm.weight",
1017
+ "shape": [
1018
+ 2048
1019
+ ],
1020
+ "dtype": "bfloat16",
1021
+ "format": "raw",
1022
+ "nbytes": 4096,
1023
+ "byteOffset": 8392704
1024
+ },
1025
+ {
1026
+ "name": "model.layers.3.self_attn.qkv_proj.weight",
1027
+ "shape": [
1028
+ 2560,
1029
+ 2048
1030
+ ],
1031
+ "dtype": "bfloat16",
1032
+ "format": "raw",
1033
+ "nbytes": 10485760,
1034
+ "byteOffset": 8396800
1035
+ },
1036
+ {
1037
+ "name": "model.layers.3.self_attn.o_proj.weight",
1038
+ "shape": [
1039
+ 2048,
1040
+ 2048
1041
+ ],
1042
+ "dtype": "bfloat16",
1043
+ "format": "raw",
1044
+ "nbytes": 8388608,
1045
+ "byteOffset": 18882560
1046
+ },
1047
+ {
1048
+ "name": "model.layers.4.input_layernorm.weight",
1049
+ "shape": [
1050
+ 2048
1051
+ ],
1052
+ "dtype": "bfloat16",
1053
+ "format": "raw",
1054
+ "nbytes": 4096,
1055
+ "byteOffset": 27271168
1056
+ },
1057
+ {
1058
+ "name": "model.layers.4.post_attention_layernorm.weight",
1059
+ "shape": [
1060
+ 2048
1061
+ ],
1062
+ "dtype": "bfloat16",
1063
+ "format": "raw",
1064
+ "nbytes": 4096,
1065
+ "byteOffset": 27275264
1066
+ }
1067
+ ],
1068
+ "md5sum": "895ac1ceb3e21746d03374268ff4244e"
1069
+ },
1070
+ {
1071
+ "dataPath": "params_shard_34.bin",
1072
+ "format": "raw-shard",
1073
+ "nbytes": 67108864,
1074
+ "records": [
1075
+ {
1076
+ "name": "model.layers.5.mlp.down_proj.weight",
1077
+ "shape": [
1078
+ 2048,
1079
+ 16384
1080
+ ],
1081
+ "dtype": "bfloat16",
1082
+ "format": "raw",
1083
+ "nbytes": 67108864,
1084
+ "byteOffset": 0
1085
+ }
1086
+ ],
1087
+ "md5sum": "291f8fffa0d9b51370bf90286b0118fc"
1088
+ },
1089
+ {
1090
+ "dataPath": "params_shard_35.bin",
1091
+ "format": "raw-shard",
1092
+ "nbytes": 134217728,
1093
+ "records": [
1094
+ {
1095
+ "name": "model.layers.5.mlp.gate_up_proj.weight",
1096
+ "shape": [
1097
+ 32768,
1098
+ 2048
1099
+ ],
1100
+ "dtype": "bfloat16",
1101
+ "format": "raw",
1102
+ "nbytes": 134217728,
1103
+ "byteOffset": 0
1104
+ }
1105
+ ],
1106
+ "md5sum": "8469971b7c6bca2a6ae1eb38c269a898"
1107
+ },
1108
+ {
1109
+ "dataPath": "params_shard_36.bin",
1110
+ "format": "raw-shard",
1111
+ "nbytes": 29368320,
1112
+ "records": [
1113
+ {
1114
+ "name": "model.layers.4.self_attn.qkv_proj.weight",
1115
+ "shape": [
1116
+ 2560,
1117
+ 2048
1118
+ ],
1119
+ "dtype": "bfloat16",
1120
+ "format": "raw",
1121
+ "nbytes": 10485760,
1122
+ "byteOffset": 0
1123
+ },
1124
+ {
1125
+ "name": "model.layers.4.self_attn.o_proj.weight",
1126
+ "shape": [
1127
+ 2048,
1128
+ 2048
1129
+ ],
1130
+ "dtype": "bfloat16",
1131
+ "format": "raw",
1132
+ "nbytes": 8388608,
1133
+ "byteOffset": 10485760
1134
+ },
1135
+ {
1136
+ "name": "model.layers.5.input_layernorm.weight",
1137
+ "shape": [
1138
+ 2048
1139
+ ],
1140
+ "dtype": "bfloat16",
1141
+ "format": "raw",
1142
+ "nbytes": 4096,
1143
+ "byteOffset": 18874368
1144
+ },
1145
+ {
1146
+ "name": "model.layers.5.post_attention_layernorm.weight",
1147
+ "shape": [
1148
+ 2048
1149
+ ],
1150
+ "dtype": "bfloat16",
1151
+ "format": "raw",
1152
+ "nbytes": 4096,
1153
+ "byteOffset": 18878464
1154
+ },
1155
+ {
1156
+ "name": "model.layers.5.self_attn.qkv_proj.weight",
1157
+ "shape": [
1158
+ 2560,
1159
+ 2048
1160
+ ],
1161
+ "dtype": "bfloat16",
1162
+ "format": "raw",
1163
+ "nbytes": 10485760,
1164
+ "byteOffset": 18882560
1165
+ }
1166
+ ],
1167
+ "md5sum": "a03a45c1809bebe8da1d4c28dd8af86f"
1168
+ },
1169
+ {
1170
+ "dataPath": "params_shard_37.bin",
1171
+ "format": "raw-shard",
1172
+ "nbytes": 67108864,
1173
+ "records": [
1174
+ {
1175
+ "name": "model.layers.6.mlp.down_proj.weight",
1176
+ "shape": [
1177
+ 2048,
1178
+ 16384
1179
+ ],
1180
+ "dtype": "bfloat16",
1181
+ "format": "raw",
1182
+ "nbytes": 67108864,
1183
+ "byteOffset": 0
1184
+ }
1185
+ ],
1186
+ "md5sum": "ede3975fcb55174d920a3f6c5331cbdd"
1187
+ },
1188
+ {
1189
+ "dataPath": "params_shard_38.bin",
1190
+ "format": "raw-shard",
1191
+ "nbytes": 134217728,
1192
+ "records": [
1193
+ {
1194
+ "name": "model.layers.6.mlp.gate_up_proj.weight",
1195
+ "shape": [
1196
+ 32768,
1197
+ 2048
1198
+ ],
1199
+ "dtype": "bfloat16",
1200
+ "format": "raw",
1201
+ "nbytes": 134217728,
1202
+ "byteOffset": 0
1203
+ }
1204
+ ],
1205
+ "md5sum": "81113b1b64021ec2a7d680376b1340c4"
1206
+ },
1207
+ {
1208
+ "dataPath": "params_shard_39.bin",
1209
+ "format": "raw-shard",
1210
+ "nbytes": 67108864,
1211
+ "records": [
1212
+ {
1213
+ "name": "model.layers.7.mlp.down_proj.weight",
1214
+ "shape": [
1215
+ 2048,
1216
+ 16384
1217
+ ],
1218
+ "dtype": "bfloat16",
1219
+ "format": "raw",
1220
+ "nbytes": 67108864,
1221
+ "byteOffset": 0
1222
+ }
1223
+ ],
1224
+ "md5sum": "6b65ffb4a139d0fc34ee600b71a15984"
1225
+ },
1226
+ {
1227
+ "dataPath": "params_shard_40.bin",
1228
+ "format": "raw-shard",
1229
+ "nbytes": 134217728,
1230
+ "records": [
1231
+ {
1232
+ "name": "model.layers.7.mlp.gate_up_proj.weight",
1233
+ "shape": [
1234
+ 32768,
1235
+ 2048
1236
+ ],
1237
+ "dtype": "bfloat16",
1238
+ "format": "raw",
1239
+ "nbytes": 134217728,
1240
+ "byteOffset": 0
1241
+ }
1242
+ ],
1243
+ "md5sum": "7907574e8d6fe13a58bf8b4f2cdf5540"
1244
+ },
1245
+ {
1246
+ "dataPath": "params_shard_41.bin",
1247
+ "format": "raw-shard",
1248
+ "nbytes": 27279360,
1249
+ "records": [
1250
+ {
1251
+ "name": "model.layers.5.self_attn.o_proj.weight",
1252
+ "shape": [
1253
+ 2048,
1254
+ 2048
1255
+ ],
1256
+ "dtype": "bfloat16",
1257
+ "format": "raw",
1258
+ "nbytes": 8388608,
1259
+ "byteOffset": 0
1260
+ },
1261
+ {
1262
+ "name": "model.layers.6.input_layernorm.weight",
1263
+ "shape": [
1264
+ 2048
1265
+ ],
1266
+ "dtype": "bfloat16",
1267
+ "format": "raw",
1268
+ "nbytes": 4096,
1269
+ "byteOffset": 8388608
1270
+ },
1271
+ {
1272
+ "name": "model.layers.6.post_attention_layernorm.weight",
1273
+ "shape": [
1274
+ 2048
1275
+ ],
1276
+ "dtype": "bfloat16",
1277
+ "format": "raw",
1278
+ "nbytes": 4096,
1279
+ "byteOffset": 8392704
1280
+ },
1281
+ {
1282
+ "name": "model.layers.6.self_attn.qkv_proj.weight",
1283
+ "shape": [
1284
+ 2560,
1285
+ 2048
1286
+ ],
1287
+ "dtype": "bfloat16",
1288
+ "format": "raw",
1289
+ "nbytes": 10485760,
1290
+ "byteOffset": 8396800
1291
+ },
1292
+ {
1293
+ "name": "model.layers.6.self_attn.o_proj.weight",
1294
+ "shape": [
1295
+ 2048,
1296
+ 2048
1297
+ ],
1298
+ "dtype": "bfloat16",
1299
+ "format": "raw",
1300
+ "nbytes": 8388608,
1301
+ "byteOffset": 18882560
1302
+ },
1303
+ {
1304
+ "name": "model.layers.7.input_layernorm.weight",
1305
+ "shape": [
1306
+ 2048
1307
+ ],
1308
+ "dtype": "bfloat16",
1309
+ "format": "raw",
1310
+ "nbytes": 4096,
1311
+ "byteOffset": 27271168
1312
+ },
1313
+ {
1314
+ "name": "model.layers.7.post_attention_layernorm.weight",
1315
+ "shape": [
1316
+ 2048
1317
+ ],
1318
+ "dtype": "bfloat16",
1319
+ "format": "raw",
1320
+ "nbytes": 4096,
1321
+ "byteOffset": 27275264
1322
+ }
1323
+ ],
1324
+ "md5sum": "c4d742181e56a57f389cdba6fe1115d4"
1325
+ },
1326
+ {
1327
+ "dataPath": "params_shard_42.bin",
1328
+ "format": "raw-shard",
1329
+ "nbytes": 67108864,
1330
+ "records": [
1331
+ {
1332
+ "name": "model.layers.8.mlp.down_proj.weight",
1333
+ "shape": [
1334
+ 2048,
1335
+ 16384
1336
+ ],
1337
+ "dtype": "bfloat16",
1338
+ "format": "raw",
1339
+ "nbytes": 67108864,
1340
+ "byteOffset": 0
1341
+ }
1342
+ ],
1343
+ "md5sum": "2b3708ef965db08042cfbb26753a31a0"
1344
+ },
1345
+ {
1346
+ "dataPath": "params_shard_43.bin",
1347
+ "format": "raw-shard",
1348
+ "nbytes": 134217728,
1349
+ "records": [
1350
+ {
1351
+ "name": "model.layers.8.mlp.gate_up_proj.weight",
1352
+ "shape": [
1353
+ 32768,
1354
+ 2048
1355
+ ],
1356
+ "dtype": "bfloat16",
1357
+ "format": "raw",
1358
+ "nbytes": 134217728,
1359
+ "byteOffset": 0
1360
+ }
1361
+ ],
1362
+ "md5sum": "396069673d0ae19fb75be4bbc0752a25"
1363
+ },
1364
+ {
1365
+ "dataPath": "params_shard_44.bin",
1366
+ "format": "raw-shard",
1367
+ "nbytes": 29368320,
1368
+ "records": [
1369
+ {
1370
+ "name": "model.layers.7.self_attn.qkv_proj.weight",
1371
+ "shape": [
1372
+ 2560,
1373
+ 2048
1374
+ ],
1375
+ "dtype": "bfloat16",
1376
+ "format": "raw",
1377
+ "nbytes": 10485760,
1378
+ "byteOffset": 0
1379
+ },
1380
+ {
1381
+ "name": "model.layers.7.self_attn.o_proj.weight",
1382
+ "shape": [
1383
+ 2048,
1384
+ 2048
1385
+ ],
1386
+ "dtype": "bfloat16",
1387
+ "format": "raw",
1388
+ "nbytes": 8388608,
1389
+ "byteOffset": 10485760
1390
+ },
1391
+ {
1392
+ "name": "model.layers.8.input_layernorm.weight",
1393
+ "shape": [
1394
+ 2048
1395
+ ],
1396
+ "dtype": "bfloat16",
1397
+ "format": "raw",
1398
+ "nbytes": 4096,
1399
+ "byteOffset": 18874368
1400
+ },
1401
+ {
1402
+ "name": "model.layers.8.post_attention_layernorm.weight",
1403
+ "shape": [
1404
+ 2048
1405
+ ],
1406
+ "dtype": "bfloat16",
1407
+ "format": "raw",
1408
+ "nbytes": 4096,
1409
+ "byteOffset": 18878464
1410
+ },
1411
+ {
1412
+ "name": "model.layers.8.self_attn.qkv_proj.weight",
1413
+ "shape": [
1414
+ 2560,
1415
+ 2048
1416
+ ],
1417
+ "dtype": "bfloat16",
1418
+ "format": "raw",
1419
+ "nbytes": 10485760,
1420
+ "byteOffset": 18882560
1421
+ }
1422
+ ],
1423
+ "md5sum": "a02b4f22cd694b60d34e8889f76be3cb"
1424
+ },
1425
+ {
1426
+ "dataPath": "params_shard_45.bin",
1427
+ "format": "raw-shard",
1428
+ "nbytes": 67108864,
1429
+ "records": [
1430
+ {
1431
+ "name": "model.layers.9.mlp.down_proj.weight",
1432
+ "shape": [
1433
+ 2048,
1434
+ 16384
1435
+ ],
1436
+ "dtype": "bfloat16",
1437
+ "format": "raw",
1438
+ "nbytes": 67108864,
1439
+ "byteOffset": 0
1440
+ }
1441
+ ],
1442
+ "md5sum": "ab38a774e2e25e749bf80c4df50d8253"
1443
+ },
1444
+ {
1445
+ "dataPath": "params_shard_46.bin",
1446
+ "format": "raw-shard",
1447
+ "nbytes": 134217728,
1448
+ "records": [
1449
+ {
1450
+ "name": "model.layers.9.mlp.gate_up_proj.weight",
1451
+ "shape": [
1452
+ 32768,
1453
+ 2048
1454
+ ],
1455
+ "dtype": "bfloat16",
1456
+ "format": "raw",
1457
+ "nbytes": 134217728,
1458
+ "byteOffset": 0
1459
+ }
1460
+ ],
1461
+ "md5sum": "2ee6fc965e7edf0b4354af2d7bee5001"
1462
+ },
1463
+ {
1464
+ "dataPath": "params_shard_47.bin",
1465
+ "format": "raw-shard",
1466
+ "nbytes": 67108864,
1467
+ "records": [
1468
+ {
1469
+ "name": "model.layers.17.mlp.down_proj.weight",
1470
+ "shape": [
1471
+ 2048,
1472
+ 16384
1473
+ ],
1474
+ "dtype": "bfloat16",
1475
+ "format": "raw",
1476
+ "nbytes": 67108864,
1477
+ "byteOffset": 0
1478
+ }
1479
+ ],
1480
+ "md5sum": "e04f27c840be79a43646c85d3bb6e1e6"
1481
+ },
1482
+ {
1483
+ "dataPath": "params_shard_48.bin",
1484
+ "format": "raw-shard",
1485
+ "nbytes": 27283456,
1486
+ "records": [
1487
+ {
1488
+ "name": "model.layers.8.self_attn.o_proj.weight",
1489
+ "shape": [
1490
+ 2048,
1491
+ 2048
1492
+ ],
1493
+ "dtype": "bfloat16",
1494
+ "format": "raw",
1495
+ "nbytes": 8388608,
1496
+ "byteOffset": 0
1497
+ },
1498
+ {
1499
+ "name": "model.layers.9.input_layernorm.weight",
1500
+ "shape": [
1501
+ 2048
1502
+ ],
1503
+ "dtype": "bfloat16",
1504
+ "format": "raw",
1505
+ "nbytes": 4096,
1506
+ "byteOffset": 8388608
1507
+ },
1508
+ {
1509
+ "name": "model.layers.9.post_attention_layernorm.weight",
1510
+ "shape": [
1511
+ 2048
1512
+ ],
1513
+ "dtype": "bfloat16",
1514
+ "format": "raw",
1515
+ "nbytes": 4096,
1516
+ "byteOffset": 8392704
1517
+ },
1518
+ {
1519
+ "name": "model.layers.9.self_attn.qkv_proj.weight",
1520
+ "shape": [
1521
+ 2560,
1522
+ 2048
1523
+ ],
1524
+ "dtype": "bfloat16",
1525
+ "format": "raw",
1526
+ "nbytes": 10485760,
1527
+ "byteOffset": 8396800
1528
+ },
1529
+ {
1530
+ "name": "model.layers.9.self_attn.o_proj.weight",
1531
+ "shape": [
1532
+ 2048,
1533
+ 2048
1534
+ ],
1535
+ "dtype": "bfloat16",
1536
+ "format": "raw",
1537
+ "nbytes": 8388608,
1538
+ "byteOffset": 18882560
1539
+ },
1540
+ {
1541
+ "name": "model.layers.17.input_layernorm.weight",
1542
+ "shape": [
1543
+ 2048
1544
+ ],
1545
+ "dtype": "bfloat16",
1546
+ "format": "raw",
1547
+ "nbytes": 4096,
1548
+ "byteOffset": 27271168
1549
+ },
1550
+ {
1551
+ "name": "model.layers.17.post_attention_layernorm.weight",
1552
+ "shape": [
1553
+ 2048
1554
+ ],
1555
+ "dtype": "bfloat16",
1556
+ "format": "raw",
1557
+ "nbytes": 4096,
1558
+ "byteOffset": 27275264
1559
+ },
1560
+ {
1561
+ "name": "model.norm.weight",
1562
+ "shape": [
1563
+ 2048
1564
+ ],
1565
+ "dtype": "bfloat16",
1566
+ "format": "raw",
1567
+ "nbytes": 4096,
1568
+ "byteOffset": 27279360
1569
+ }
1570
+ ],
1571
+ "md5sum": "a63f5ec6dc518cc6bd4b80ac101aa028"
1572
+ }
1573
+ ]
1574
+ }
ndarray-cache.json ADDED
@@ -0,0 +1,1574 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 110,
4
+ "ParamBytes": 10024689664.0,
5
+ "BitsPerParam": 32.0
6
+ },
7
+ "records": [
8
+ {
9
+ "dataPath": "params_shard_0.bin",
10
+ "format": "raw-shard",
11
+ "nbytes": 1048576000,
12
+ "records": [
13
+ {
14
+ "name": "model.embed_tokens.weight",
15
+ "shape": [
16
+ 256000,
17
+ 2048
18
+ ],
19
+ "dtype": "float32",
20
+ "format": "f32-to-bf16",
21
+ "nbytes": 1048576000,
22
+ "byteOffset": 0
23
+ }
24
+ ],
25
+ "md5sum": "21128c642a829dff80b86de466616d9d"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 67108864,
31
+ "records": [
32
+ {
33
+ "name": "model.layers.0.mlp.down_proj.weight",
34
+ "shape": [
35
+ 2048,
36
+ 16384
37
+ ],
38
+ "dtype": "float32",
39
+ "format": "f32-to-bf16",
40
+ "nbytes": 67108864,
41
+ "byteOffset": 0
42
+ }
43
+ ],
44
+ "md5sum": "e3daa1938df1d775e3448e4fdd33527e"
45
+ },
46
+ {
47
+ "dataPath": "params_shard_2.bin",
48
+ "format": "raw-shard",
49
+ "nbytes": 134217728,
50
+ "records": [
51
+ {
52
+ "name": "model.layers.0.mlp.gate_up_proj.weight",
53
+ "shape": [
54
+ 32768,
55
+ 2048
56
+ ],
57
+ "dtype": "float32",
58
+ "format": "f32-to-bf16",
59
+ "nbytes": 134217728,
60
+ "byteOffset": 0
61
+ }
62
+ ],
63
+ "md5sum": "5ebf170cde97e9025d2c3473b8357c13"
64
+ },
65
+ {
66
+ "dataPath": "params_shard_3.bin",
67
+ "format": "raw-shard",
68
+ "nbytes": 67108864,
69
+ "records": [
70
+ {
71
+ "name": "model.layers.1.mlp.down_proj.weight",
72
+ "shape": [
73
+ 2048,
74
+ 16384
75
+ ],
76
+ "dtype": "float32",
77
+ "format": "f32-to-bf16",
78
+ "nbytes": 67108864,
79
+ "byteOffset": 0
80
+ }
81
+ ],
82
+ "md5sum": "b11672a46d1224e14cd2a5c8590a02ea"
83
+ },
84
+ {
85
+ "dataPath": "params_shard_4.bin",
86
+ "format": "raw-shard",
87
+ "nbytes": 134217728,
88
+ "records": [
89
+ {
90
+ "name": "model.layers.1.mlp.gate_up_proj.weight",
91
+ "shape": [
92
+ 32768,
93
+ 2048
94
+ ],
95
+ "dtype": "float32",
96
+ "format": "f32-to-bf16",
97
+ "nbytes": 134217728,
98
+ "byteOffset": 0
99
+ }
100
+ ],
101
+ "md5sum": "9970e7f9c5a4be1cb1acbb075fe39a60"
102
+ },
103
+ {
104
+ "dataPath": "params_shard_5.bin",
105
+ "format": "raw-shard",
106
+ "nbytes": 29376512,
107
+ "records": [
108
+ {
109
+ "name": "model.layers.0.input_layernorm.weight",
110
+ "shape": [
111
+ 2048
112
+ ],
113
+ "dtype": "float32",
114
+ "format": "f32-to-bf16",
115
+ "nbytes": 4096,
116
+ "byteOffset": 0
117
+ },
118
+ {
119
+ "name": "model.layers.0.post_attention_layernorm.weight",
120
+ "shape": [
121
+ 2048
122
+ ],
123
+ "dtype": "float32",
124
+ "format": "f32-to-bf16",
125
+ "nbytes": 4096,
126
+ "byteOffset": 4096
127
+ },
128
+ {
129
+ "name": "model.layers.0.self_attn.qkv_proj.weight",
130
+ "shape": [
131
+ 2560,
132
+ 2048
133
+ ],
134
+ "dtype": "float32",
135
+ "format": "f32-to-bf16",
136
+ "nbytes": 10485760,
137
+ "byteOffset": 8192
138
+ },
139
+ {
140
+ "name": "model.layers.0.self_attn.o_proj.weight",
141
+ "shape": [
142
+ 2048,
143
+ 2048
144
+ ],
145
+ "dtype": "float32",
146
+ "format": "f32-to-bf16",
147
+ "nbytes": 8388608,
148
+ "byteOffset": 10493952
149
+ },
150
+ {
151
+ "name": "model.layers.1.input_layernorm.weight",
152
+ "shape": [
153
+ 2048
154
+ ],
155
+ "dtype": "float32",
156
+ "format": "f32-to-bf16",
157
+ "nbytes": 4096,
158
+ "byteOffset": 18882560
159
+ },
160
+ {
161
+ "name": "model.layers.1.post_attention_layernorm.weight",
162
+ "shape": [
163
+ 2048
164
+ ],
165
+ "dtype": "float32",
166
+ "format": "f32-to-bf16",
167
+ "nbytes": 4096,
168
+ "byteOffset": 18886656
169
+ },
170
+ {
171
+ "name": "model.layers.1.self_attn.qkv_proj.weight",
172
+ "shape": [
173
+ 2560,
174
+ 2048
175
+ ],
176
+ "dtype": "float32",
177
+ "format": "f32-to-bf16",
178
+ "nbytes": 10485760,
179
+ "byteOffset": 18890752
180
+ }
181
+ ],
182
+ "md5sum": "690bd198a683ab8104715e9fa33e3eaf"
183
+ },
184
+ {
185
+ "dataPath": "params_shard_6.bin",
186
+ "format": "raw-shard",
187
+ "nbytes": 67108864,
188
+ "records": [
189
+ {
190
+ "name": "model.layers.10.mlp.down_proj.weight",
191
+ "shape": [
192
+ 2048,
193
+ 16384
194
+ ],
195
+ "dtype": "float32",
196
+ "format": "f32-to-bf16",
197
+ "nbytes": 67108864,
198
+ "byteOffset": 0
199
+ }
200
+ ],
201
+ "md5sum": "71c4717b75b1547985ea79fb4451908d"
202
+ },
203
+ {
204
+ "dataPath": "params_shard_7.bin",
205
+ "format": "raw-shard",
206
+ "nbytes": 134217728,
207
+ "records": [
208
+ {
209
+ "name": "model.layers.10.mlp.gate_up_proj.weight",
210
+ "shape": [
211
+ 32768,
212
+ 2048
213
+ ],
214
+ "dtype": "float32",
215
+ "format": "f32-to-bf16",
216
+ "nbytes": 134217728,
217
+ "byteOffset": 0
218
+ }
219
+ ],
220
+ "md5sum": "0385b2a5124f03324c0cbbf8f8afb4d5"
221
+ },
222
+ {
223
+ "dataPath": "params_shard_8.bin",
224
+ "format": "raw-shard",
225
+ "nbytes": 67108864,
226
+ "records": [
227
+ {
228
+ "name": "model.layers.11.mlp.down_proj.weight",
229
+ "shape": [
230
+ 2048,
231
+ 16384
232
+ ],
233
+ "dtype": "float32",
234
+ "format": "f32-to-bf16",
235
+ "nbytes": 67108864,
236
+ "byteOffset": 0
237
+ }
238
+ ],
239
+ "md5sum": "9d7bbb224840a3c5f24b33f029fc8d3d"
240
+ },
241
+ {
242
+ "dataPath": "params_shard_9.bin",
243
+ "format": "raw-shard",
244
+ "nbytes": 134217728,
245
+ "records": [
246
+ {
247
+ "name": "model.layers.11.mlp.gate_up_proj.weight",
248
+ "shape": [
249
+ 32768,
250
+ 2048
251
+ ],
252
+ "dtype": "float32",
253
+ "format": "f32-to-bf16",
254
+ "nbytes": 134217728,
255
+ "byteOffset": 0
256
+ }
257
+ ],
258
+ "md5sum": "babd2c92010bfd919b25fbd9daaaa8d9"
259
+ },
260
+ {
261
+ "dataPath": "params_shard_10.bin",
262
+ "format": "raw-shard",
263
+ "nbytes": 27279360,
264
+ "records": [
265
+ {
266
+ "name": "model.layers.1.self_attn.o_proj.weight",
267
+ "shape": [
268
+ 2048,
269
+ 2048
270
+ ],
271
+ "dtype": "float32",
272
+ "format": "f32-to-bf16",
273
+ "nbytes": 8388608,
274
+ "byteOffset": 0
275
+ },
276
+ {
277
+ "name": "model.layers.10.input_layernorm.weight",
278
+ "shape": [
279
+ 2048
280
+ ],
281
+ "dtype": "float32",
282
+ "format": "f32-to-bf16",
283
+ "nbytes": 4096,
284
+ "byteOffset": 8388608
285
+ },
286
+ {
287
+ "name": "model.layers.10.post_attention_layernorm.weight",
288
+ "shape": [
289
+ 2048
290
+ ],
291
+ "dtype": "float32",
292
+ "format": "f32-to-bf16",
293
+ "nbytes": 4096,
294
+ "byteOffset": 8392704
295
+ },
296
+ {
297
+ "name": "model.layers.10.self_attn.qkv_proj.weight",
298
+ "shape": [
299
+ 2560,
300
+ 2048
301
+ ],
302
+ "dtype": "float32",
303
+ "format": "f32-to-bf16",
304
+ "nbytes": 10485760,
305
+ "byteOffset": 8396800
306
+ },
307
+ {
308
+ "name": "model.layers.10.self_attn.o_proj.weight",
309
+ "shape": [
310
+ 2048,
311
+ 2048
312
+ ],
313
+ "dtype": "float32",
314
+ "format": "f32-to-bf16",
315
+ "nbytes": 8388608,
316
+ "byteOffset": 18882560
317
+ },
318
+ {
319
+ "name": "model.layers.11.input_layernorm.weight",
320
+ "shape": [
321
+ 2048
322
+ ],
323
+ "dtype": "float32",
324
+ "format": "f32-to-bf16",
325
+ "nbytes": 4096,
326
+ "byteOffset": 27271168
327
+ },
328
+ {
329
+ "name": "model.layers.11.post_attention_layernorm.weight",
330
+ "shape": [
331
+ 2048
332
+ ],
333
+ "dtype": "float32",
334
+ "format": "f32-to-bf16",
335
+ "nbytes": 4096,
336
+ "byteOffset": 27275264
337
+ }
338
+ ],
339
+ "md5sum": "73e3bd854f6acf1969dae0923e8bb11b"
340
+ },
341
+ {
342
+ "dataPath": "params_shard_11.bin",
343
+ "format": "raw-shard",
344
+ "nbytes": 67108864,
345
+ "records": [
346
+ {
347
+ "name": "model.layers.12.mlp.down_proj.weight",
348
+ "shape": [
349
+ 2048,
350
+ 16384
351
+ ],
352
+ "dtype": "float32",
353
+ "format": "f32-to-bf16",
354
+ "nbytes": 67108864,
355
+ "byteOffset": 0
356
+ }
357
+ ],
358
+ "md5sum": "113f6d9072ff3bd46cb356305c3d6435"
359
+ },
360
+ {
361
+ "dataPath": "params_shard_12.bin",
362
+ "format": "raw-shard",
363
+ "nbytes": 134217728,
364
+ "records": [
365
+ {
366
+ "name": "model.layers.12.mlp.gate_up_proj.weight",
367
+ "shape": [
368
+ 32768,
369
+ 2048
370
+ ],
371
+ "dtype": "float32",
372
+ "format": "f32-to-bf16",
373
+ "nbytes": 134217728,
374
+ "byteOffset": 0
375
+ }
376
+ ],
377
+ "md5sum": "2c179b332604d046c25fdb8fa7e8a38d"
378
+ },
379
+ {
380
+ "dataPath": "params_shard_13.bin",
381
+ "format": "raw-shard",
382
+ "nbytes": 29368320,
383
+ "records": [
384
+ {
385
+ "name": "model.layers.11.self_attn.qkv_proj.weight",
386
+ "shape": [
387
+ 2560,
388
+ 2048
389
+ ],
390
+ "dtype": "float32",
391
+ "format": "f32-to-bf16",
392
+ "nbytes": 10485760,
393
+ "byteOffset": 0
394
+ },
395
+ {
396
+ "name": "model.layers.11.self_attn.o_proj.weight",
397
+ "shape": [
398
+ 2048,
399
+ 2048
400
+ ],
401
+ "dtype": "float32",
402
+ "format": "f32-to-bf16",
403
+ "nbytes": 8388608,
404
+ "byteOffset": 10485760
405
+ },
406
+ {
407
+ "name": "model.layers.12.input_layernorm.weight",
408
+ "shape": [
409
+ 2048
410
+ ],
411
+ "dtype": "float32",
412
+ "format": "f32-to-bf16",
413
+ "nbytes": 4096,
414
+ "byteOffset": 18874368
415
+ },
416
+ {
417
+ "name": "model.layers.12.post_attention_layernorm.weight",
418
+ "shape": [
419
+ 2048
420
+ ],
421
+ "dtype": "float32",
422
+ "format": "f32-to-bf16",
423
+ "nbytes": 4096,
424
+ "byteOffset": 18878464
425
+ },
426
+ {
427
+ "name": "model.layers.12.self_attn.qkv_proj.weight",
428
+ "shape": [
429
+ 2560,
430
+ 2048
431
+ ],
432
+ "dtype": "float32",
433
+ "format": "f32-to-bf16",
434
+ "nbytes": 10485760,
435
+ "byteOffset": 18882560
436
+ }
437
+ ],
438
+ "md5sum": "b155622e091c8a222b1f4de1814cbb49"
439
+ },
440
+ {
441
+ "dataPath": "params_shard_14.bin",
442
+ "format": "raw-shard",
443
+ "nbytes": 67108864,
444
+ "records": [
445
+ {
446
+ "name": "model.layers.13.mlp.down_proj.weight",
447
+ "shape": [
448
+ 2048,
449
+ 16384
450
+ ],
451
+ "dtype": "float32",
452
+ "format": "f32-to-bf16",
453
+ "nbytes": 67108864,
454
+ "byteOffset": 0
455
+ }
456
+ ],
457
+ "md5sum": "fb8e2063f991dd07a06fe38ea7365789"
458
+ },
459
+ {
460
+ "dataPath": "params_shard_15.bin",
461
+ "format": "raw-shard",
462
+ "nbytes": 134217728,
463
+ "records": [
464
+ {
465
+ "name": "model.layers.13.mlp.gate_up_proj.weight",
466
+ "shape": [
467
+ 32768,
468
+ 2048
469
+ ],
470
+ "dtype": "float32",
471
+ "format": "f32-to-bf16",
472
+ "nbytes": 134217728,
473
+ "byteOffset": 0
474
+ }
475
+ ],
476
+ "md5sum": "9d5ceb3198e4924fccc31e3f2421ff14"
477
+ },
478
+ {
479
+ "dataPath": "params_shard_16.bin",
480
+ "format": "raw-shard",
481
+ "nbytes": 67108864,
482
+ "records": [
483
+ {
484
+ "name": "model.layers.14.mlp.down_proj.weight",
485
+ "shape": [
486
+ 2048,
487
+ 16384
488
+ ],
489
+ "dtype": "float32",
490
+ "format": "f32-to-bf16",
491
+ "nbytes": 67108864,
492
+ "byteOffset": 0
493
+ }
494
+ ],
495
+ "md5sum": "9fb12a797891bebed74f0e65eefb538b"
496
+ },
497
+ {
498
+ "dataPath": "params_shard_17.bin",
499
+ "format": "raw-shard",
500
+ "nbytes": 134217728,
501
+ "records": [
502
+ {
503
+ "name": "model.layers.14.mlp.gate_up_proj.weight",
504
+ "shape": [
505
+ 32768,
506
+ 2048
507
+ ],
508
+ "dtype": "float32",
509
+ "format": "f32-to-bf16",
510
+ "nbytes": 134217728,
511
+ "byteOffset": 0
512
+ }
513
+ ],
514
+ "md5sum": "59630bc8e808b139cf23c4a4d048a81f"
515
+ },
516
+ {
517
+ "dataPath": "params_shard_18.bin",
518
+ "format": "raw-shard",
519
+ "nbytes": 27279360,
520
+ "records": [
521
+ {
522
+ "name": "model.layers.12.self_attn.o_proj.weight",
523
+ "shape": [
524
+ 2048,
525
+ 2048
526
+ ],
527
+ "dtype": "float32",
528
+ "format": "f32-to-bf16",
529
+ "nbytes": 8388608,
530
+ "byteOffset": 0
531
+ },
532
+ {
533
+ "name": "model.layers.13.input_layernorm.weight",
534
+ "shape": [
535
+ 2048
536
+ ],
537
+ "dtype": "float32",
538
+ "format": "f32-to-bf16",
539
+ "nbytes": 4096,
540
+ "byteOffset": 8388608
541
+ },
542
+ {
543
+ "name": "model.layers.13.post_attention_layernorm.weight",
544
+ "shape": [
545
+ 2048
546
+ ],
547
+ "dtype": "float32",
548
+ "format": "f32-to-bf16",
549
+ "nbytes": 4096,
550
+ "byteOffset": 8392704
551
+ },
552
+ {
553
+ "name": "model.layers.13.self_attn.qkv_proj.weight",
554
+ "shape": [
555
+ 2560,
556
+ 2048
557
+ ],
558
+ "dtype": "float32",
559
+ "format": "f32-to-bf16",
560
+ "nbytes": 10485760,
561
+ "byteOffset": 8396800
562
+ },
563
+ {
564
+ "name": "model.layers.13.self_attn.o_proj.weight",
565
+ "shape": [
566
+ 2048,
567
+ 2048
568
+ ],
569
+ "dtype": "float32",
570
+ "format": "f32-to-bf16",
571
+ "nbytes": 8388608,
572
+ "byteOffset": 18882560
573
+ },
574
+ {
575
+ "name": "model.layers.14.input_layernorm.weight",
576
+ "shape": [
577
+ 2048
578
+ ],
579
+ "dtype": "float32",
580
+ "format": "f32-to-bf16",
581
+ "nbytes": 4096,
582
+ "byteOffset": 27271168
583
+ },
584
+ {
585
+ "name": "model.layers.14.post_attention_layernorm.weight",
586
+ "shape": [
587
+ 2048
588
+ ],
589
+ "dtype": "float32",
590
+ "format": "f32-to-bf16",
591
+ "nbytes": 4096,
592
+ "byteOffset": 27275264
593
+ }
594
+ ],
595
+ "md5sum": "17759e68d24e2eb936f9129a722522f5"
596
+ },
597
+ {
598
+ "dataPath": "params_shard_19.bin",
599
+ "format": "raw-shard",
600
+ "nbytes": 67108864,
601
+ "records": [
602
+ {
603
+ "name": "model.layers.15.mlp.down_proj.weight",
604
+ "shape": [
605
+ 2048,
606
+ 16384
607
+ ],
608
+ "dtype": "float32",
609
+ "format": "f32-to-bf16",
610
+ "nbytes": 67108864,
611
+ "byteOffset": 0
612
+ }
613
+ ],
614
+ "md5sum": "321ba4846c90abaa26d7179a514625dd"
615
+ },
616
+ {
617
+ "dataPath": "params_shard_20.bin",
618
+ "format": "raw-shard",
619
+ "nbytes": 134217728,
620
+ "records": [
621
+ {
622
+ "name": "model.layers.15.mlp.gate_up_proj.weight",
623
+ "shape": [
624
+ 32768,
625
+ 2048
626
+ ],
627
+ "dtype": "float32",
628
+ "format": "f32-to-bf16",
629
+ "nbytes": 134217728,
630
+ "byteOffset": 0
631
+ }
632
+ ],
633
+ "md5sum": "f8665cbb955f5b9470e4fb4f2bebfd3a"
634
+ },
635
+ {
636
+ "dataPath": "params_shard_21.bin",
637
+ "format": "raw-shard",
638
+ "nbytes": 29368320,
639
+ "records": [
640
+ {
641
+ "name": "model.layers.14.self_attn.qkv_proj.weight",
642
+ "shape": [
643
+ 2560,
644
+ 2048
645
+ ],
646
+ "dtype": "float32",
647
+ "format": "f32-to-bf16",
648
+ "nbytes": 10485760,
649
+ "byteOffset": 0
650
+ },
651
+ {
652
+ "name": "model.layers.14.self_attn.o_proj.weight",
653
+ "shape": [
654
+ 2048,
655
+ 2048
656
+ ],
657
+ "dtype": "float32",
658
+ "format": "f32-to-bf16",
659
+ "nbytes": 8388608,
660
+ "byteOffset": 10485760
661
+ },
662
+ {
663
+ "name": "model.layers.15.input_layernorm.weight",
664
+ "shape": [
665
+ 2048
666
+ ],
667
+ "dtype": "float32",
668
+ "format": "f32-to-bf16",
669
+ "nbytes": 4096,
670
+ "byteOffset": 18874368
671
+ },
672
+ {
673
+ "name": "model.layers.15.post_attention_layernorm.weight",
674
+ "shape": [
675
+ 2048
676
+ ],
677
+ "dtype": "float32",
678
+ "format": "f32-to-bf16",
679
+ "nbytes": 4096,
680
+ "byteOffset": 18878464
681
+ },
682
+ {
683
+ "name": "model.layers.15.self_attn.qkv_proj.weight",
684
+ "shape": [
685
+ 2560,
686
+ 2048
687
+ ],
688
+ "dtype": "float32",
689
+ "format": "f32-to-bf16",
690
+ "nbytes": 10485760,
691
+ "byteOffset": 18882560
692
+ }
693
+ ],
694
+ "md5sum": "1d686c17f8ed473437ce9dbb37f66f5a"
695
+ },
696
+ {
697
+ "dataPath": "params_shard_22.bin",
698
+ "format": "raw-shard",
699
+ "nbytes": 67108864,
700
+ "records": [
701
+ {
702
+ "name": "model.layers.16.mlp.down_proj.weight",
703
+ "shape": [
704
+ 2048,
705
+ 16384
706
+ ],
707
+ "dtype": "float32",
708
+ "format": "f32-to-bf16",
709
+ "nbytes": 67108864,
710
+ "byteOffset": 0
711
+ }
712
+ ],
713
+ "md5sum": "2218e355df7616289c17f7cf571c21e4"
714
+ },
715
+ {
716
+ "dataPath": "params_shard_23.bin",
717
+ "format": "raw-shard",
718
+ "nbytes": 134217728,
719
+ "records": [
720
+ {
721
+ "name": "model.layers.16.mlp.gate_up_proj.weight",
722
+ "shape": [
723
+ 32768,
724
+ 2048
725
+ ],
726
+ "dtype": "float32",
727
+ "format": "f32-to-bf16",
728
+ "nbytes": 134217728,
729
+ "byteOffset": 0
730
+ }
731
+ ],
732
+ "md5sum": "5ecbb6372848f97fae1bdef305c623d0"
733
+ },
734
+ {
735
+ "dataPath": "params_shard_24.bin",
736
+ "format": "raw-shard",
737
+ "nbytes": 134217728,
738
+ "records": [
739
+ {
740
+ "name": "model.layers.17.mlp.gate_up_proj.weight",
741
+ "shape": [
742
+ 32768,
743
+ 2048
744
+ ],
745
+ "dtype": "float32",
746
+ "format": "f32-to-bf16",
747
+ "nbytes": 134217728,
748
+ "byteOffset": 0
749
+ }
750
+ ],
751
+ "md5sum": "466ae14bce7c0f6efa9c939f132ea082"
752
+ },
753
+ {
754
+ "dataPath": "params_shard_25.bin",
755
+ "format": "raw-shard",
756
+ "nbytes": 27271168,
757
+ "records": [
758
+ {
759
+ "name": "model.layers.15.self_attn.o_proj.weight",
760
+ "shape": [
761
+ 2048,
762
+ 2048
763
+ ],
764
+ "dtype": "float32",
765
+ "format": "f32-to-bf16",
766
+ "nbytes": 8388608,
767
+ "byteOffset": 0
768
+ },
769
+ {
770
+ "name": "model.layers.16.input_layernorm.weight",
771
+ "shape": [
772
+ 2048
773
+ ],
774
+ "dtype": "float32",
775
+ "format": "f32-to-bf16",
776
+ "nbytes": 4096,
777
+ "byteOffset": 8388608
778
+ },
779
+ {
780
+ "name": "model.layers.16.post_attention_layernorm.weight",
781
+ "shape": [
782
+ 2048
783
+ ],
784
+ "dtype": "float32",
785
+ "format": "f32-to-bf16",
786
+ "nbytes": 4096,
787
+ "byteOffset": 8392704
788
+ },
789
+ {
790
+ "name": "model.layers.16.self_attn.qkv_proj.weight",
791
+ "shape": [
792
+ 2560,
793
+ 2048
794
+ ],
795
+ "dtype": "float32",
796
+ "format": "f32-to-bf16",
797
+ "nbytes": 10485760,
798
+ "byteOffset": 8396800
799
+ },
800
+ {
801
+ "name": "model.layers.16.self_attn.o_proj.weight",
802
+ "shape": [
803
+ 2048,
804
+ 2048
805
+ ],
806
+ "dtype": "float32",
807
+ "format": "f32-to-bf16",
808
+ "nbytes": 8388608,
809
+ "byteOffset": 18882560
810
+ }
811
+ ],
812
+ "md5sum": "517de03093360d322e5b3152bd5c1ef0"
813
+ },
814
+ {
815
+ "dataPath": "params_shard_26.bin",
816
+ "format": "raw-shard",
817
+ "nbytes": 67108864,
818
+ "records": [
819
+ {
820
+ "name": "model.layers.2.mlp.down_proj.weight",
821
+ "shape": [
822
+ 2048,
823
+ 16384
824
+ ],
825
+ "dtype": "float32",
826
+ "format": "f32-to-bf16",
827
+ "nbytes": 67108864,
828
+ "byteOffset": 0
829
+ }
830
+ ],
831
+ "md5sum": "7a6c37d60d558c0a9d650a59c4fc5c68"
832
+ },
833
+ {
834
+ "dataPath": "params_shard_27.bin",
835
+ "format": "raw-shard",
836
+ "nbytes": 134217728,
837
+ "records": [
838
+ {
839
+ "name": "model.layers.2.mlp.gate_up_proj.weight",
840
+ "shape": [
841
+ 32768,
842
+ 2048
843
+ ],
844
+ "dtype": "float32",
845
+ "format": "f32-to-bf16",
846
+ "nbytes": 134217728,
847
+ "byteOffset": 0
848
+ }
849
+ ],
850
+ "md5sum": "1f51d337520bbbf5a56530c60e6b4e2b"
851
+ },
852
+ {
853
+ "dataPath": "params_shard_28.bin",
854
+ "format": "raw-shard",
855
+ "nbytes": 29368320,
856
+ "records": [
857
+ {
858
+ "name": "model.layers.17.self_attn.qkv_proj.weight",
859
+ "shape": [
860
+ 2560,
861
+ 2048
862
+ ],
863
+ "dtype": "float32",
864
+ "format": "f32-to-bf16",
865
+ "nbytes": 10485760,
866
+ "byteOffset": 0
867
+ },
868
+ {
869
+ "name": "model.layers.17.self_attn.o_proj.weight",
870
+ "shape": [
871
+ 2048,
872
+ 2048
873
+ ],
874
+ "dtype": "float32",
875
+ "format": "f32-to-bf16",
876
+ "nbytes": 8388608,
877
+ "byteOffset": 10485760
878
+ },
879
+ {
880
+ "name": "model.layers.2.input_layernorm.weight",
881
+ "shape": [
882
+ 2048
883
+ ],
884
+ "dtype": "float32",
885
+ "format": "f32-to-bf16",
886
+ "nbytes": 4096,
887
+ "byteOffset": 18874368
888
+ },
889
+ {
890
+ "name": "model.layers.2.post_attention_layernorm.weight",
891
+ "shape": [
892
+ 2048
893
+ ],
894
+ "dtype": "float32",
895
+ "format": "f32-to-bf16",
896
+ "nbytes": 4096,
897
+ "byteOffset": 18878464
898
+ },
899
+ {
900
+ "name": "model.layers.2.self_attn.qkv_proj.weight",
901
+ "shape": [
902
+ 2560,
903
+ 2048
904
+ ],
905
+ "dtype": "float32",
906
+ "format": "f32-to-bf16",
907
+ "nbytes": 10485760,
908
+ "byteOffset": 18882560
909
+ }
910
+ ],
911
+ "md5sum": "81402425b99d3caf95d4a94ea8616bf0"
912
+ },
913
+ {
914
+ "dataPath": "params_shard_29.bin",
915
+ "format": "raw-shard",
916
+ "nbytes": 67108864,
917
+ "records": [
918
+ {
919
+ "name": "model.layers.3.mlp.down_proj.weight",
920
+ "shape": [
921
+ 2048,
922
+ 16384
923
+ ],
924
+ "dtype": "float32",
925
+ "format": "f32-to-bf16",
926
+ "nbytes": 67108864,
927
+ "byteOffset": 0
928
+ }
929
+ ],
930
+ "md5sum": "c3300e44f9c90f83867efd3d6aef1d8d"
931
+ },
932
+ {
933
+ "dataPath": "params_shard_30.bin",
934
+ "format": "raw-shard",
935
+ "nbytes": 134217728,
936
+ "records": [
937
+ {
938
+ "name": "model.layers.3.mlp.gate_up_proj.weight",
939
+ "shape": [
940
+ 32768,
941
+ 2048
942
+ ],
943
+ "dtype": "float32",
944
+ "format": "f32-to-bf16",
945
+ "nbytes": 134217728,
946
+ "byteOffset": 0
947
+ }
948
+ ],
949
+ "md5sum": "538c26b83e0c2ab4bdf48cd10df6bd53"
950
+ },
951
+ {
952
+ "dataPath": "params_shard_31.bin",
953
+ "format": "raw-shard",
954
+ "nbytes": 67108864,
955
+ "records": [
956
+ {
957
+ "name": "model.layers.4.mlp.down_proj.weight",
958
+ "shape": [
959
+ 2048,
960
+ 16384
961
+ ],
962
+ "dtype": "float32",
963
+ "format": "f32-to-bf16",
964
+ "nbytes": 67108864,
965
+ "byteOffset": 0
966
+ }
967
+ ],
968
+ "md5sum": "88501c646188da831d14676a256ac84f"
969
+ },
970
+ {
971
+ "dataPath": "params_shard_32.bin",
972
+ "format": "raw-shard",
973
+ "nbytes": 134217728,
974
+ "records": [
975
+ {
976
+ "name": "model.layers.4.mlp.gate_up_proj.weight",
977
+ "shape": [
978
+ 32768,
979
+ 2048
980
+ ],
981
+ "dtype": "float32",
982
+ "format": "f32-to-bf16",
983
+ "nbytes": 134217728,
984
+ "byteOffset": 0
985
+ }
986
+ ],
987
+ "md5sum": "074d0db9b1d459438705b37b6670c06b"
988
+ },
989
+ {
990
+ "dataPath": "params_shard_33.bin",
991
+ "format": "raw-shard",
992
+ "nbytes": 27279360,
993
+ "records": [
994
+ {
995
+ "name": "model.layers.2.self_attn.o_proj.weight",
996
+ "shape": [
997
+ 2048,
998
+ 2048
999
+ ],
1000
+ "dtype": "float32",
1001
+ "format": "f32-to-bf16",
1002
+ "nbytes": 8388608,
1003
+ "byteOffset": 0
1004
+ },
1005
+ {
1006
+ "name": "model.layers.3.input_layernorm.weight",
1007
+ "shape": [
1008
+ 2048
1009
+ ],
1010
+ "dtype": "float32",
1011
+ "format": "f32-to-bf16",
1012
+ "nbytes": 4096,
1013
+ "byteOffset": 8388608
1014
+ },
1015
+ {
1016
+ "name": "model.layers.3.post_attention_layernorm.weight",
1017
+ "shape": [
1018
+ 2048
1019
+ ],
1020
+ "dtype": "float32",
1021
+ "format": "f32-to-bf16",
1022
+ "nbytes": 4096,
1023
+ "byteOffset": 8392704
1024
+ },
1025
+ {
1026
+ "name": "model.layers.3.self_attn.qkv_proj.weight",
1027
+ "shape": [
1028
+ 2560,
1029
+ 2048
1030
+ ],
1031
+ "dtype": "float32",
1032
+ "format": "f32-to-bf16",
1033
+ "nbytes": 10485760,
1034
+ "byteOffset": 8396800
1035
+ },
1036
+ {
1037
+ "name": "model.layers.3.self_attn.o_proj.weight",
1038
+ "shape": [
1039
+ 2048,
1040
+ 2048
1041
+ ],
1042
+ "dtype": "float32",
1043
+ "format": "f32-to-bf16",
1044
+ "nbytes": 8388608,
1045
+ "byteOffset": 18882560
1046
+ },
1047
+ {
1048
+ "name": "model.layers.4.input_layernorm.weight",
1049
+ "shape": [
1050
+ 2048
1051
+ ],
1052
+ "dtype": "float32",
1053
+ "format": "f32-to-bf16",
1054
+ "nbytes": 4096,
1055
+ "byteOffset": 27271168
1056
+ },
1057
+ {
1058
+ "name": "model.layers.4.post_attention_layernorm.weight",
1059
+ "shape": [
1060
+ 2048
1061
+ ],
1062
+ "dtype": "float32",
1063
+ "format": "f32-to-bf16",
1064
+ "nbytes": 4096,
1065
+ "byteOffset": 27275264
1066
+ }
1067
+ ],
1068
+ "md5sum": "895ac1ceb3e21746d03374268ff4244e"
1069
+ },
1070
+ {
1071
+ "dataPath": "params_shard_34.bin",
1072
+ "format": "raw-shard",
1073
+ "nbytes": 67108864,
1074
+ "records": [
1075
+ {
1076
+ "name": "model.layers.5.mlp.down_proj.weight",
1077
+ "shape": [
1078
+ 2048,
1079
+ 16384
1080
+ ],
1081
+ "dtype": "float32",
1082
+ "format": "f32-to-bf16",
1083
+ "nbytes": 67108864,
1084
+ "byteOffset": 0
1085
+ }
1086
+ ],
1087
+ "md5sum": "291f8fffa0d9b51370bf90286b0118fc"
1088
+ },
1089
+ {
1090
+ "dataPath": "params_shard_35.bin",
1091
+ "format": "raw-shard",
1092
+ "nbytes": 134217728,
1093
+ "records": [
1094
+ {
1095
+ "name": "model.layers.5.mlp.gate_up_proj.weight",
1096
+ "shape": [
1097
+ 32768,
1098
+ 2048
1099
+ ],
1100
+ "dtype": "float32",
1101
+ "format": "f32-to-bf16",
1102
+ "nbytes": 134217728,
1103
+ "byteOffset": 0
1104
+ }
1105
+ ],
1106
+ "md5sum": "8469971b7c6bca2a6ae1eb38c269a898"
1107
+ },
1108
+ {
1109
+ "dataPath": "params_shard_36.bin",
1110
+ "format": "raw-shard",
1111
+ "nbytes": 29368320,
1112
+ "records": [
1113
+ {
1114
+ "name": "model.layers.4.self_attn.qkv_proj.weight",
1115
+ "shape": [
1116
+ 2560,
1117
+ 2048
1118
+ ],
1119
+ "dtype": "float32",
1120
+ "format": "f32-to-bf16",
1121
+ "nbytes": 10485760,
1122
+ "byteOffset": 0
1123
+ },
1124
+ {
1125
+ "name": "model.layers.4.self_attn.o_proj.weight",
1126
+ "shape": [
1127
+ 2048,
1128
+ 2048
1129
+ ],
1130
+ "dtype": "float32",
1131
+ "format": "f32-to-bf16",
1132
+ "nbytes": 8388608,
1133
+ "byteOffset": 10485760
1134
+ },
1135
+ {
1136
+ "name": "model.layers.5.input_layernorm.weight",
1137
+ "shape": [
1138
+ 2048
1139
+ ],
1140
+ "dtype": "float32",
1141
+ "format": "f32-to-bf16",
1142
+ "nbytes": 4096,
1143
+ "byteOffset": 18874368
1144
+ },
1145
+ {
1146
+ "name": "model.layers.5.post_attention_layernorm.weight",
1147
+ "shape": [
1148
+ 2048
1149
+ ],
1150
+ "dtype": "float32",
1151
+ "format": "f32-to-bf16",
1152
+ "nbytes": 4096,
1153
+ "byteOffset": 18878464
1154
+ },
1155
+ {
1156
+ "name": "model.layers.5.self_attn.qkv_proj.weight",
1157
+ "shape": [
1158
+ 2560,
1159
+ 2048
1160
+ ],
1161
+ "dtype": "float32",
1162
+ "format": "f32-to-bf16",
1163
+ "nbytes": 10485760,
1164
+ "byteOffset": 18882560
1165
+ }
1166
+ ],
1167
+ "md5sum": "a03a45c1809bebe8da1d4c28dd8af86f"
1168
+ },
1169
+ {
1170
+ "dataPath": "params_shard_37.bin",
1171
+ "format": "raw-shard",
1172
+ "nbytes": 67108864,
1173
+ "records": [
1174
+ {
1175
+ "name": "model.layers.6.mlp.down_proj.weight",
1176
+ "shape": [
1177
+ 2048,
1178
+ 16384
1179
+ ],
1180
+ "dtype": "float32",
1181
+ "format": "f32-to-bf16",
1182
+ "nbytes": 67108864,
1183
+ "byteOffset": 0
1184
+ }
1185
+ ],
1186
+ "md5sum": "ede3975fcb55174d920a3f6c5331cbdd"
1187
+ },
1188
+ {
1189
+ "dataPath": "params_shard_38.bin",
1190
+ "format": "raw-shard",
1191
+ "nbytes": 134217728,
1192
+ "records": [
1193
+ {
1194
+ "name": "model.layers.6.mlp.gate_up_proj.weight",
1195
+ "shape": [
1196
+ 32768,
1197
+ 2048
1198
+ ],
1199
+ "dtype": "float32",
1200
+ "format": "f32-to-bf16",
1201
+ "nbytes": 134217728,
1202
+ "byteOffset": 0
1203
+ }
1204
+ ],
1205
+ "md5sum": "81113b1b64021ec2a7d680376b1340c4"
1206
+ },
1207
+ {
1208
+ "dataPath": "params_shard_39.bin",
1209
+ "format": "raw-shard",
1210
+ "nbytes": 67108864,
1211
+ "records": [
1212
+ {
1213
+ "name": "model.layers.7.mlp.down_proj.weight",
1214
+ "shape": [
1215
+ 2048,
1216
+ 16384
1217
+ ],
1218
+ "dtype": "float32",
1219
+ "format": "f32-to-bf16",
1220
+ "nbytes": 67108864,
1221
+ "byteOffset": 0
1222
+ }
1223
+ ],
1224
+ "md5sum": "6b65ffb4a139d0fc34ee600b71a15984"
1225
+ },
1226
+ {
1227
+ "dataPath": "params_shard_40.bin",
1228
+ "format": "raw-shard",
1229
+ "nbytes": 134217728,
1230
+ "records": [
1231
+ {
1232
+ "name": "model.layers.7.mlp.gate_up_proj.weight",
1233
+ "shape": [
1234
+ 32768,
1235
+ 2048
1236
+ ],
1237
+ "dtype": "float32",
1238
+ "format": "f32-to-bf16",
1239
+ "nbytes": 134217728,
1240
+ "byteOffset": 0
1241
+ }
1242
+ ],
1243
+ "md5sum": "7907574e8d6fe13a58bf8b4f2cdf5540"
1244
+ },
1245
+ {
1246
+ "dataPath": "params_shard_41.bin",
1247
+ "format": "raw-shard",
1248
+ "nbytes": 27279360,
1249
+ "records": [
1250
+ {
1251
+ "name": "model.layers.5.self_attn.o_proj.weight",
1252
+ "shape": [
1253
+ 2048,
1254
+ 2048
1255
+ ],
1256
+ "dtype": "float32",
1257
+ "format": "f32-to-bf16",
1258
+ "nbytes": 8388608,
1259
+ "byteOffset": 0
1260
+ },
1261
+ {
1262
+ "name": "model.layers.6.input_layernorm.weight",
1263
+ "shape": [
1264
+ 2048
1265
+ ],
1266
+ "dtype": "float32",
1267
+ "format": "f32-to-bf16",
1268
+ "nbytes": 4096,
1269
+ "byteOffset": 8388608
1270
+ },
1271
+ {
1272
+ "name": "model.layers.6.post_attention_layernorm.weight",
1273
+ "shape": [
1274
+ 2048
1275
+ ],
1276
+ "dtype": "float32",
1277
+ "format": "f32-to-bf16",
1278
+ "nbytes": 4096,
1279
+ "byteOffset": 8392704
1280
+ },
1281
+ {
1282
+ "name": "model.layers.6.self_attn.qkv_proj.weight",
1283
+ "shape": [
1284
+ 2560,
1285
+ 2048
1286
+ ],
1287
+ "dtype": "float32",
1288
+ "format": "f32-to-bf16",
1289
+ "nbytes": 10485760,
1290
+ "byteOffset": 8396800
1291
+ },
1292
+ {
1293
+ "name": "model.layers.6.self_attn.o_proj.weight",
1294
+ "shape": [
1295
+ 2048,
1296
+ 2048
1297
+ ],
1298
+ "dtype": "float32",
1299
+ "format": "f32-to-bf16",
1300
+ "nbytes": 8388608,
1301
+ "byteOffset": 18882560
1302
+ },
1303
+ {
1304
+ "name": "model.layers.7.input_layernorm.weight",
1305
+ "shape": [
1306
+ 2048
1307
+ ],
1308
+ "dtype": "float32",
1309
+ "format": "f32-to-bf16",
1310
+ "nbytes": 4096,
1311
+ "byteOffset": 27271168
1312
+ },
1313
+ {
1314
+ "name": "model.layers.7.post_attention_layernorm.weight",
1315
+ "shape": [
1316
+ 2048
1317
+ ],
1318
+ "dtype": "float32",
1319
+ "format": "f32-to-bf16",
1320
+ "nbytes": 4096,
1321
+ "byteOffset": 27275264
1322
+ }
1323
+ ],
1324
+ "md5sum": "c4d742181e56a57f389cdba6fe1115d4"
1325
+ },
1326
+ {
1327
+ "dataPath": "params_shard_42.bin",
1328
+ "format": "raw-shard",
1329
+ "nbytes": 67108864,
1330
+ "records": [
1331
+ {
1332
+ "name": "model.layers.8.mlp.down_proj.weight",
1333
+ "shape": [
1334
+ 2048,
1335
+ 16384
1336
+ ],
1337
+ "dtype": "float32",
1338
+ "format": "f32-to-bf16",
1339
+ "nbytes": 67108864,
1340
+ "byteOffset": 0
1341
+ }
1342
+ ],
1343
+ "md5sum": "2b3708ef965db08042cfbb26753a31a0"
1344
+ },
1345
+ {
1346
+ "dataPath": "params_shard_43.bin",
1347
+ "format": "raw-shard",
1348
+ "nbytes": 134217728,
1349
+ "records": [
1350
+ {
1351
+ "name": "model.layers.8.mlp.gate_up_proj.weight",
1352
+ "shape": [
1353
+ 32768,
1354
+ 2048
1355
+ ],
1356
+ "dtype": "float32",
1357
+ "format": "f32-to-bf16",
1358
+ "nbytes": 134217728,
1359
+ "byteOffset": 0
1360
+ }
1361
+ ],
1362
+ "md5sum": "396069673d0ae19fb75be4bbc0752a25"
1363
+ },
1364
+ {
1365
+ "dataPath": "params_shard_44.bin",
1366
+ "format": "raw-shard",
1367
+ "nbytes": 29368320,
1368
+ "records": [
1369
+ {
1370
+ "name": "model.layers.7.self_attn.qkv_proj.weight",
1371
+ "shape": [
1372
+ 2560,
1373
+ 2048
1374
+ ],
1375
+ "dtype": "float32",
1376
+ "format": "f32-to-bf16",
1377
+ "nbytes": 10485760,
1378
+ "byteOffset": 0
1379
+ },
1380
+ {
1381
+ "name": "model.layers.7.self_attn.o_proj.weight",
1382
+ "shape": [
1383
+ 2048,
1384
+ 2048
1385
+ ],
1386
+ "dtype": "float32",
1387
+ "format": "f32-to-bf16",
1388
+ "nbytes": 8388608,
1389
+ "byteOffset": 10485760
1390
+ },
1391
+ {
1392
+ "name": "model.layers.8.input_layernorm.weight",
1393
+ "shape": [
1394
+ 2048
1395
+ ],
1396
+ "dtype": "float32",
1397
+ "format": "f32-to-bf16",
1398
+ "nbytes": 4096,
1399
+ "byteOffset": 18874368
1400
+ },
1401
+ {
1402
+ "name": "model.layers.8.post_attention_layernorm.weight",
1403
+ "shape": [
1404
+ 2048
1405
+ ],
1406
+ "dtype": "float32",
1407
+ "format": "f32-to-bf16",
1408
+ "nbytes": 4096,
1409
+ "byteOffset": 18878464
1410
+ },
1411
+ {
1412
+ "name": "model.layers.8.self_attn.qkv_proj.weight",
1413
+ "shape": [
1414
+ 2560,
1415
+ 2048
1416
+ ],
1417
+ "dtype": "float32",
1418
+ "format": "f32-to-bf16",
1419
+ "nbytes": 10485760,
1420
+ "byteOffset": 18882560
1421
+ }
1422
+ ],
1423
+ "md5sum": "a02b4f22cd694b60d34e8889f76be3cb"
1424
+ },
1425
+ {
1426
+ "dataPath": "params_shard_45.bin",
1427
+ "format": "raw-shard",
1428
+ "nbytes": 67108864,
1429
+ "records": [
1430
+ {
1431
+ "name": "model.layers.9.mlp.down_proj.weight",
1432
+ "shape": [
1433
+ 2048,
1434
+ 16384
1435
+ ],
1436
+ "dtype": "float32",
1437
+ "format": "f32-to-bf16",
1438
+ "nbytes": 67108864,
1439
+ "byteOffset": 0
1440
+ }
1441
+ ],
1442
+ "md5sum": "ab38a774e2e25e749bf80c4df50d8253"
1443
+ },
1444
+ {
1445
+ "dataPath": "params_shard_46.bin",
1446
+ "format": "raw-shard",
1447
+ "nbytes": 134217728,
1448
+ "records": [
1449
+ {
1450
+ "name": "model.layers.9.mlp.gate_up_proj.weight",
1451
+ "shape": [
1452
+ 32768,
1453
+ 2048
1454
+ ],
1455
+ "dtype": "float32",
1456
+ "format": "f32-to-bf16",
1457
+ "nbytes": 134217728,
1458
+ "byteOffset": 0
1459
+ }
1460
+ ],
1461
+ "md5sum": "2ee6fc965e7edf0b4354af2d7bee5001"
1462
+ },
1463
+ {
1464
+ "dataPath": "params_shard_47.bin",
1465
+ "format": "raw-shard",
1466
+ "nbytes": 67108864,
1467
+ "records": [
1468
+ {
1469
+ "name": "model.layers.17.mlp.down_proj.weight",
1470
+ "shape": [
1471
+ 2048,
1472
+ 16384
1473
+ ],
1474
+ "dtype": "float32",
1475
+ "format": "f32-to-bf16",
1476
+ "nbytes": 67108864,
1477
+ "byteOffset": 0
1478
+ }
1479
+ ],
1480
+ "md5sum": "e04f27c840be79a43646c85d3bb6e1e6"
1481
+ },
1482
+ {
1483
+ "dataPath": "params_shard_48.bin",
1484
+ "format": "raw-shard",
1485
+ "nbytes": 27283456,
1486
+ "records": [
1487
+ {
1488
+ "name": "model.layers.8.self_attn.o_proj.weight",
1489
+ "shape": [
1490
+ 2048,
1491
+ 2048
1492
+ ],
1493
+ "dtype": "float32",
1494
+ "format": "f32-to-bf16",
1495
+ "nbytes": 8388608,
1496
+ "byteOffset": 0
1497
+ },
1498
+ {
1499
+ "name": "model.layers.9.input_layernorm.weight",
1500
+ "shape": [
1501
+ 2048
1502
+ ],
1503
+ "dtype": "float32",
1504
+ "format": "f32-to-bf16",
1505
+ "nbytes": 4096,
1506
+ "byteOffset": 8388608
1507
+ },
1508
+ {
1509
+ "name": "model.layers.9.post_attention_layernorm.weight",
1510
+ "shape": [
1511
+ 2048
1512
+ ],
1513
+ "dtype": "float32",
1514
+ "format": "f32-to-bf16",
1515
+ "nbytes": 4096,
1516
+ "byteOffset": 8392704
1517
+ },
1518
+ {
1519
+ "name": "model.layers.9.self_attn.qkv_proj.weight",
1520
+ "shape": [
1521
+ 2560,
1522
+ 2048
1523
+ ],
1524
+ "dtype": "float32",
1525
+ "format": "f32-to-bf16",
1526
+ "nbytes": 10485760,
1527
+ "byteOffset": 8396800
1528
+ },
1529
+ {
1530
+ "name": "model.layers.9.self_attn.o_proj.weight",
1531
+ "shape": [
1532
+ 2048,
1533
+ 2048
1534
+ ],
1535
+ "dtype": "float32",
1536
+ "format": "f32-to-bf16",
1537
+ "nbytes": 8388608,
1538
+ "byteOffset": 18882560
1539
+ },
1540
+ {
1541
+ "name": "model.layers.17.input_layernorm.weight",
1542
+ "shape": [
1543
+ 2048
1544
+ ],
1545
+ "dtype": "float32",
1546
+ "format": "f32-to-bf16",
1547
+ "nbytes": 4096,
1548
+ "byteOffset": 27271168
1549
+ },
1550
+ {
1551
+ "name": "model.layers.17.post_attention_layernorm.weight",
1552
+ "shape": [
1553
+ 2048
1554
+ ],
1555
+ "dtype": "float32",
1556
+ "format": "f32-to-bf16",
1557
+ "nbytes": 4096,
1558
+ "byteOffset": 27275264
1559
+ },
1560
+ {
1561
+ "name": "model.norm.weight",
1562
+ "shape": [
1563
+ 2048
1564
+ ],
1565
+ "dtype": "float32",
1566
+ "format": "f32-to-bf16",
1567
+ "nbytes": 4096,
1568
+ "byteOffset": 27279360
1569
+ }
1570
+ ],
1571
+ "md5sum": "a63f5ec6dc518cc6bd4b80ac101aa028"
1572
+ }
1573
+ ]
1574
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:447e03cf3694ddbaa3228db3c33207a69d0336039ed82cb8b91731792087e5d2
3
+ size 1048576000
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26441e7afbb07a154fbf159f92a96e1aa6d90fd72ca754c9f116eab781c77242
3
+ size 67108864
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bc0fdded7ebf62bc9373c527bdfccfba54e0cb7fc2956d3ec09f93c1dacf0f1
3
+ size 27279360
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5412e0b2498a85596277d11a773007bf371db8ca9cbe0fe2a0ffaceb8f09939
3
+ size 67108864
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:482813b09613413263b9c5f16ff28e05bb59e9b826adc214ef775be054677d69
3
+ size 134217728
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88b1a8248b4573f8e3aca68fef6d41e07f8ebaaec2cca1ffedcb162769b1c65c
3
+ size 29368320
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:890fe21aad9aad4c1187c129615580b750d1d8a94d38cc467e10d04f45984d24
3
+ size 67108864
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:82d9616de567d0176d1e9ffee5c73a85f755144272fdef14bed6b15f457716d9
3
+ size 134217728
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80bc040cbb1cf9dd0decd7e8fc1bd878e78501c3bc0633aa0bcc2e51a5194d51
3
+ size 67108864
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00246eb05417932fd3defff51fda814c14cda833d3c666f40c3c7606296eacb3
3
+ size 134217728
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:681a32c761851a36339a30c2e7d5bf319d00bb935045a2d2bc7e03e732a8c086
3
+ size 27279360
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef1aca73c0e715ed5a0d31815f1910ce04e4b3a963c839fd3a9c7a3db179e9b3
3
+ size 67108864
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cd65b222a9c525e494646c665ef9b176f5b530dee15f3e767318fba8650d325
3
+ size 134217728
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c45c5fcfedc7c19253291909eca104f13410cc3bb8d191da8e3eae666186fc1
3
+ size 134217728
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eac076b60fac3180cdee49e493937a420bbe9b9b53866f4d997369979c5e60b2
3
+ size 29368320
params_shard_22.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8b0d3a4e463a18d68526afbd313dbc955b1275bdd9b6d27fc51a12fbb2b66b1
3
+ size 67108864
params_shard_23.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a3d0423ba59d52dfcf9ca0bef27393e30a8dd00aa3d3f36f88ab4dbe98ad8e9
3
+ size 134217728
params_shard_24.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22cee6dae3a6aec560edd9bb1de380b4f1ac9105f1e397e87fac1f95f7ed21a9
3
+ size 134217728
params_shard_25.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a93f62d90475bbeed0b210ec0ac51938ada3016e3d28d7ebbc314f2100dd1ac9
3
+ size 27271168
params_shard_26.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:955f39f5be448f871278271f6b086fca71bca4cefc954c6814026df3babcae35
3
+ size 67108864
params_shard_27.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7bd6e728b3f9b3e2b3eb0fb3162b783f96b360b44e50cd94cae1f7ca2536ffd
3
+ size 134217728
params_shard_28.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcf2657eff06297c634c49684185ed501804960e7fbbe6000981f24e35ec8524
3
+ size 29368320
params_shard_29.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e29614f375b3e48b562744732231b57f11bfaf9998771ed0d7a439912b7b7351
3
+ size 67108864
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:868c4288727a2497ddcaf6aa09b69800e18cf779964ee8855ade84c23faa7552
3
+ size 67108864
params_shard_30.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68da507070bada3e67470a8714f29c2001ce2c61719ccfcab8b66d26503d6436
3
+ size 134217728
params_shard_31.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9b376dee4d1c4c217a24cdd0e817179b745a3a32ac50d0b0478e169807008ed
3
+ size 67108864
params_shard_32.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:acce1d34fcd5062393b3320970aacdad8aa6dd1b139b6047a5b8ecec292872a5
3
+ size 134217728
params_shard_33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b4bab597d05d620d5829e94f9d05914773a3f3df55c128ef5830ef0cb23027b
3
+ size 27279360
params_shard_34.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27913d12f755ddbb32eb116c09f643b6edea7e69894a30c0fdbfd86eb1b4c388
3
+ size 67108864
params_shard_35.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14aa93031eb036e78b4936ac988123742fc79471114e9d8f7332a4da06d84c51
3
+ size 134217728
params_shard_36.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58c8627b735cc97216d330b57e804ce46d0cf14aa4f5977b35e8a6ada2c87d54
3
+ size 29368320
params_shard_37.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ee3e532e8a7d329d86f73fc4ede3cf7777748aededb6645537e38710c5ffe8d
3
+ size 67108864
params_shard_38.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4017f187713cf5eedcd60b39323c37b24fb0e3a4cd9f136bb5b4f2e1f8b7714f
3
+ size 134217728
params_shard_39.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f97eaf4de2a7055ca7fcd973b39f5cddfda1fb655354c5f27c1be159e5839c4
3
+ size 67108864
params_shard_4.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc74e137cc14ce40cee5bcd449f42b96e87005b0e6a5f2944f2f45df303339e8
3
+ size 134217728
params_shard_40.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0eee01b0cd2caf46187ffc3c5b39df19118a8f3bedecbff1625609b929e0df8a
3
+ size 134217728
params_shard_41.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d621c313343e30b8e8b3bea72424c14bbb3ef596ab2c83ad2f73152b2e71f54
3
+ size 27279360
params_shard_42.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7b08f2aea1ce1d883bae8f8cafbc852e3d16a5d930fd154eebe4c3d652abed6
3
+ size 67108864
params_shard_43.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3a1c7404100c3b64b4f317f91f0f22290bada47082ca507b068ff1923c262bf9
3
+ size 134217728
params_shard_44.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07336e596c3f7fb7f069315d78eac56a30d62e22272e6308d3f16770418fee3d
3
+ size 29368320
params_shard_45.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e62495cd8a2a2a8a60486e5843a51bf416cb4aaca5e0d476b7d028f3cad0f4f
3
+ size 67108864
params_shard_46.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67ccce1615198b1b7e8839c5ad33d11604a170ff66e1e81abbe1df77bdcbc564
3
+ size 134217728
params_shard_47.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9985c210d8360ab962c68727716c385d395a7466396448c3a6353c524289b1c1
3
+ size 67108864
params_shard_48.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03531febebfcf6f7f0113bbfab50ef47a9fe4fb5944669c4bf3456c9146d4a56
3
+ size 27283456
params_shard_5.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8ecd0681032c3726b1a88381743209199d34b5fdb759c329929a3f8b1cc60f4
3
+ size 29376512