Hugging Face's logo

nota-ai
/
cpt-lora_st-vicuna-v1.3-1.5b-ppl-q4f16_0-MLC

Model card Files Community
illian01 commited on
Commit
1168a19
·
1 Parent(s): 47755cb

Add weights and config

Browse files
Files changed (29) hide show
  1. mlc-chat-config.json +77 -0
  2. ndarray-cache.json +903 -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 +35 -0
mlc-chat-config.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "0.1.0",
3
+ "model_type": "llama",
4
+ "quantization": "q4f16_0",
5
+ "model_config": {
6
+ "hidden_size": 4096,
7
+ "intermediate_size": 11008,
8
+ "num_attention_heads": 32,
9
+ "num_hidden_layers": 6,
10
+ "rms_norm_eps": 1e-06,
11
+ "vocab_size": 32000,
12
+ "position_embedding_base": 10000,
13
+ "context_window_size": 2048,
14
+ "prefill_chunk_size": 2048,
15
+ "num_key_value_heads": 32,
16
+ "head_dim": 128,
17
+ "tensor_parallel_shards": 1,
18
+ "max_batch_size": 80
19
+ },
20
+ "vocab_size": 32000,
21
+ "context_window_size": 2048,
22
+ "sliding_window_size": -1,
23
+ "prefill_chunk_size": 2048,
24
+ "attention_sink_size": -1,
25
+ "tensor_parallel_shards": 1,
26
+ "mean_gen_len": 128,
27
+ "max_gen_len": 512,
28
+ "shift_fill_factor": 0.3,
29
+ "temperature": 0.6,
30
+ "presence_penalty": 0.0,
31
+ "frequency_penalty": 0.0,
32
+ "repetition_penalty": 1.0,
33
+ "top_p": 0.9,
34
+ "tokenizer_files": [
35
+ "tokenizer.model",
36
+ "tokenizer_config.json",
37
+ "tokenizer.json"
38
+ ],
39
+ "tokenizer_info": {
40
+ "token_postproc_method": "byte_fallback",
41
+ "prepend_space_in_encode": true,
42
+ "strip_space_in_decode": true
43
+ },
44
+ "conv_template": {
45
+ "name": "st-llm",
46
+ "system_template": "Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.\n\n### Instruction:\n{MessagePlaceholders.SYSTEM.value}</s>",
47
+ "system_message": "You are a helpful, respectful and honest assistant.",
48
+ "system_prefix_token_ids": null,
49
+ "add_role_after_system_message": true,
50
+ "roles": {
51
+ "user": "### Input:",
52
+ "assistant": "### Response:"
53
+ },
54
+ "role_templates": {
55
+ "user": "{user_message}",
56
+ "assistant": "{assistant_message}",
57
+ "tool": "{tool_message}"
58
+ },
59
+ "messages": [],
60
+ "seps": [
61
+ "</s>"
62
+ ],
63
+ "role_content_sep": "\n",
64
+ "role_empty_sep": "\n",
65
+ "stop_str": [
66
+ "</s>"
67
+ ],
68
+ "stop_token_ids": [
69
+ 2
70
+ ],
71
+ "function_string": "",
72
+ "use_function_calling": false
73
+ },
74
+ "pad_token_id": 0,
75
+ "bos_token_id": 1,
76
+ "eos_token_id": 2
77
+ }
ndarray-cache.json ADDED
@@ -0,0 +1,903 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "ParamSize": 65,
4
+ "ParamBytes": 830578688.0,
5
+ "BitsPerParam": 4.500414746671623
6
+ },
7
+ "records": [
8
+ {
9
+ "dataPath": "params_shard_0.bin",
10
+ "format": "raw-shard",
11
+ "nbytes": 65536000,
12
+ "records": [
13
+ {
14
+ "name": "model.embed_tokens.q_weight",
15
+ "shape": [
16
+ 32000,
17
+ 512
18
+ ],
19
+ "dtype": "uint32",
20
+ "format": "f32-to-bf16",
21
+ "nbytes": 65536000,
22
+ "byteOffset": 0
23
+ }
24
+ ],
25
+ "md5sum": "34491ecb22b5e7f4ca56d82527c438f9"
26
+ },
27
+ {
28
+ "dataPath": "params_shard_1.bin",
29
+ "format": "raw-shard",
30
+ "nbytes": 33357824,
31
+ "records": [
32
+ {
33
+ "name": "model.embed_tokens.q_scale",
34
+ "shape": [
35
+ 32000,
36
+ 128
37
+ ],
38
+ "dtype": "float16",
39
+ "format": "f32-to-bf16",
40
+ "nbytes": 8192000,
41
+ "byteOffset": 0
42
+ },
43
+ {
44
+ "name": "model.layers.0.self_attn.qkv_proj.q_weight",
45
+ "shape": [
46
+ 512,
47
+ 12288
48
+ ],
49
+ "dtype": "uint32",
50
+ "format": "f32-to-bf16",
51
+ "nbytes": 25165824,
52
+ "byteOffset": 8192000
53
+ }
54
+ ],
55
+ "md5sum": "661089d83913d30e7575835461db0abf"
56
+ },
57
+ {
58
+ "dataPath": "params_shard_2.bin",
59
+ "format": "raw-shard",
60
+ "nbytes": 45088768,
61
+ "records": [
62
+ {
63
+ "name": "model.layers.0.mlp.gate_up_proj.q_weight",
64
+ "shape": [
65
+ 512,
66
+ 22016
67
+ ],
68
+ "dtype": "uint32",
69
+ "format": "f32-to-bf16",
70
+ "nbytes": 45088768,
71
+ "byteOffset": 0
72
+ }
73
+ ],
74
+ "md5sum": "78b992a9d971e604e70b413c233a60c8"
75
+ },
76
+ {
77
+ "dataPath": "params_shard_3.bin",
78
+ "format": "raw-shard",
79
+ "nbytes": 22544384,
80
+ "records": [
81
+ {
82
+ "name": "model.layers.0.mlp.down_proj.q_weight",
83
+ "shape": [
84
+ 1376,
85
+ 4096
86
+ ],
87
+ "dtype": "uint32",
88
+ "format": "f32-to-bf16",
89
+ "nbytes": 22544384,
90
+ "byteOffset": 0
91
+ }
92
+ ],
93
+ "md5sum": "2120fd58f3fa655d99cc9fe71bfbee19"
94
+ },
95
+ {
96
+ "dataPath": "params_shard_4.bin",
97
+ "format": "raw-shard",
98
+ "nbytes": 25165824,
99
+ "records": [
100
+ {
101
+ "name": "model.layers.1.self_attn.qkv_proj.q_weight",
102
+ "shape": [
103
+ 512,
104
+ 12288
105
+ ],
106
+ "dtype": "uint32",
107
+ "format": "f32-to-bf16",
108
+ "nbytes": 25165824,
109
+ "byteOffset": 0
110
+ }
111
+ ],
112
+ "md5sum": "c15d92613464fb1db85970571ee48e90"
113
+ },
114
+ {
115
+ "dataPath": "params_shard_5.bin",
116
+ "format": "raw-shard",
117
+ "nbytes": 32587776,
118
+ "records": [
119
+ {
120
+ "name": "model.layers.0.self_attn.qkv_proj.q_scale",
121
+ "shape": [
122
+ 128,
123
+ 12288
124
+ ],
125
+ "dtype": "float16",
126
+ "format": "f32-to-bf16",
127
+ "nbytes": 3145728,
128
+ "byteOffset": 0
129
+ },
130
+ {
131
+ "name": "model.layers.0.self_attn.o_proj.q_weight",
132
+ "shape": [
133
+ 512,
134
+ 4096
135
+ ],
136
+ "dtype": "uint32",
137
+ "format": "f32-to-bf16",
138
+ "nbytes": 8388608,
139
+ "byteOffset": 3145728
140
+ },
141
+ {
142
+ "name": "model.layers.0.self_attn.o_proj.q_scale",
143
+ "shape": [
144
+ 128,
145
+ 4096
146
+ ],
147
+ "dtype": "float16",
148
+ "format": "f32-to-bf16",
149
+ "nbytes": 1048576,
150
+ "byteOffset": 11534336
151
+ },
152
+ {
153
+ "name": "model.layers.0.mlp.gate_up_proj.q_scale",
154
+ "shape": [
155
+ 128,
156
+ 22016
157
+ ],
158
+ "dtype": "float16",
159
+ "format": "f32-to-bf16",
160
+ "nbytes": 5636096,
161
+ "byteOffset": 12582912
162
+ },
163
+ {
164
+ "name": "model.layers.0.mlp.down_proj.q_scale",
165
+ "shape": [
166
+ 344,
167
+ 4096
168
+ ],
169
+ "dtype": "float16",
170
+ "format": "f32-to-bf16",
171
+ "nbytes": 2818048,
172
+ "byteOffset": 18219008
173
+ },
174
+ {
175
+ "name": "model.layers.0.input_layernorm.weight",
176
+ "shape": [
177
+ 4096
178
+ ],
179
+ "dtype": "float16",
180
+ "format": "f32-to-bf16",
181
+ "nbytes": 8192,
182
+ "byteOffset": 21037056
183
+ },
184
+ {
185
+ "name": "model.layers.0.post_attention_layernorm.weight",
186
+ "shape": [
187
+ 4096
188
+ ],
189
+ "dtype": "float16",
190
+ "format": "f32-to-bf16",
191
+ "nbytes": 8192,
192
+ "byteOffset": 21045248
193
+ },
194
+ {
195
+ "name": "model.layers.1.self_attn.qkv_proj.q_scale",
196
+ "shape": [
197
+ 128,
198
+ 12288
199
+ ],
200
+ "dtype": "float16",
201
+ "format": "f32-to-bf16",
202
+ "nbytes": 3145728,
203
+ "byteOffset": 21053440
204
+ },
205
+ {
206
+ "name": "model.layers.1.self_attn.o_proj.q_weight",
207
+ "shape": [
208
+ 512,
209
+ 4096
210
+ ],
211
+ "dtype": "uint32",
212
+ "format": "f32-to-bf16",
213
+ "nbytes": 8388608,
214
+ "byteOffset": 24199168
215
+ }
216
+ ],
217
+ "md5sum": "dfe551efa483fbf60228e22987a26a32"
218
+ },
219
+ {
220
+ "dataPath": "params_shard_6.bin",
221
+ "format": "raw-shard",
222
+ "nbytes": 45088768,
223
+ "records": [
224
+ {
225
+ "name": "model.layers.1.mlp.gate_up_proj.q_weight",
226
+ "shape": [
227
+ 512,
228
+ 22016
229
+ ],
230
+ "dtype": "uint32",
231
+ "format": "f32-to-bf16",
232
+ "nbytes": 45088768,
233
+ "byteOffset": 0
234
+ }
235
+ ],
236
+ "md5sum": "c68ba999a632e735721f525e36cb4188"
237
+ },
238
+ {
239
+ "dataPath": "params_shard_7.bin",
240
+ "format": "raw-shard",
241
+ "nbytes": 25165824,
242
+ "records": [
243
+ {
244
+ "name": "model.layers.2.self_attn.qkv_proj.q_weight",
245
+ "shape": [
246
+ 512,
247
+ 12288
248
+ ],
249
+ "dtype": "uint32",
250
+ "format": "f32-to-bf16",
251
+ "nbytes": 25165824,
252
+ "byteOffset": 0
253
+ }
254
+ ],
255
+ "md5sum": "567649c94813883b48ed1dfcc8ee6998"
256
+ },
257
+ {
258
+ "dataPath": "params_shard_8.bin",
259
+ "format": "raw-shard",
260
+ "nbytes": 32063488,
261
+ "records": [
262
+ {
263
+ "name": "model.layers.1.self_attn.o_proj.q_scale",
264
+ "shape": [
265
+ 128,
266
+ 4096
267
+ ],
268
+ "dtype": "float16",
269
+ "format": "f32-to-bf16",
270
+ "nbytes": 1048576,
271
+ "byteOffset": 0
272
+ },
273
+ {
274
+ "name": "model.layers.1.mlp.gate_up_proj.q_scale",
275
+ "shape": [
276
+ 128,
277
+ 22016
278
+ ],
279
+ "dtype": "float16",
280
+ "format": "f32-to-bf16",
281
+ "nbytes": 5636096,
282
+ "byteOffset": 1048576
283
+ },
284
+ {
285
+ "name": "model.layers.1.mlp.down_proj.q_weight",
286
+ "shape": [
287
+ 1376,
288
+ 4096
289
+ ],
290
+ "dtype": "uint32",
291
+ "format": "f32-to-bf16",
292
+ "nbytes": 22544384,
293
+ "byteOffset": 6684672
294
+ },
295
+ {
296
+ "name": "model.layers.1.mlp.down_proj.q_scale",
297
+ "shape": [
298
+ 344,
299
+ 4096
300
+ ],
301
+ "dtype": "float16",
302
+ "format": "f32-to-bf16",
303
+ "nbytes": 2818048,
304
+ "byteOffset": 29229056
305
+ },
306
+ {
307
+ "name": "model.layers.1.input_layernorm.weight",
308
+ "shape": [
309
+ 4096
310
+ ],
311
+ "dtype": "float16",
312
+ "format": "f32-to-bf16",
313
+ "nbytes": 8192,
314
+ "byteOffset": 32047104
315
+ },
316
+ {
317
+ "name": "model.layers.1.post_attention_layernorm.weight",
318
+ "shape": [
319
+ 4096
320
+ ],
321
+ "dtype": "float16",
322
+ "format": "f32-to-bf16",
323
+ "nbytes": 8192,
324
+ "byteOffset": 32055296
325
+ }
326
+ ],
327
+ "md5sum": "573dff167a868366a699177dbfbc44c2"
328
+ },
329
+ {
330
+ "dataPath": "params_shard_9.bin",
331
+ "format": "raw-shard",
332
+ "nbytes": 45088768,
333
+ "records": [
334
+ {
335
+ "name": "model.layers.2.mlp.gate_up_proj.q_weight",
336
+ "shape": [
337
+ 512,
338
+ 22016
339
+ ],
340
+ "dtype": "uint32",
341
+ "format": "f32-to-bf16",
342
+ "nbytes": 45088768,
343
+ "byteOffset": 0
344
+ }
345
+ ],
346
+ "md5sum": "d369ce6ca2712ad0265c77bc31aca5c9"
347
+ },
348
+ {
349
+ "dataPath": "params_shard_10.bin",
350
+ "format": "raw-shard",
351
+ "nbytes": 22544384,
352
+ "records": [
353
+ {
354
+ "name": "model.layers.2.mlp.down_proj.q_weight",
355
+ "shape": [
356
+ 1376,
357
+ 4096
358
+ ],
359
+ "dtype": "uint32",
360
+ "format": "f32-to-bf16",
361
+ "nbytes": 22544384,
362
+ "byteOffset": 0
363
+ }
364
+ ],
365
+ "md5sum": "c642cf007bf1c783ce4318c03dd865f5"
366
+ },
367
+ {
368
+ "dataPath": "params_shard_11.bin",
369
+ "format": "raw-shard",
370
+ "nbytes": 25165824,
371
+ "records": [
372
+ {
373
+ "name": "model.layers.3.self_attn.qkv_proj.q_weight",
374
+ "shape": [
375
+ 512,
376
+ 12288
377
+ ],
378
+ "dtype": "uint32",
379
+ "format": "f32-to-bf16",
380
+ "nbytes": 25165824,
381
+ "byteOffset": 0
382
+ }
383
+ ],
384
+ "md5sum": "c677dbc5d6763d852ccc339f40d70a46"
385
+ },
386
+ {
387
+ "dataPath": "params_shard_12.bin",
388
+ "format": "raw-shard",
389
+ "nbytes": 32587776,
390
+ "records": [
391
+ {
392
+ "name": "model.layers.2.self_attn.qkv_proj.q_scale",
393
+ "shape": [
394
+ 128,
395
+ 12288
396
+ ],
397
+ "dtype": "float16",
398
+ "format": "f32-to-bf16",
399
+ "nbytes": 3145728,
400
+ "byteOffset": 0
401
+ },
402
+ {
403
+ "name": "model.layers.2.self_attn.o_proj.q_weight",
404
+ "shape": [
405
+ 512,
406
+ 4096
407
+ ],
408
+ "dtype": "uint32",
409
+ "format": "f32-to-bf16",
410
+ "nbytes": 8388608,
411
+ "byteOffset": 3145728
412
+ },
413
+ {
414
+ "name": "model.layers.2.self_attn.o_proj.q_scale",
415
+ "shape": [
416
+ 128,
417
+ 4096
418
+ ],
419
+ "dtype": "float16",
420
+ "format": "f32-to-bf16",
421
+ "nbytes": 1048576,
422
+ "byteOffset": 11534336
423
+ },
424
+ {
425
+ "name": "model.layers.2.mlp.gate_up_proj.q_scale",
426
+ "shape": [
427
+ 128,
428
+ 22016
429
+ ],
430
+ "dtype": "float16",
431
+ "format": "f32-to-bf16",
432
+ "nbytes": 5636096,
433
+ "byteOffset": 12582912
434
+ },
435
+ {
436
+ "name": "model.layers.2.mlp.down_proj.q_scale",
437
+ "shape": [
438
+ 344,
439
+ 4096
440
+ ],
441
+ "dtype": "float16",
442
+ "format": "f32-to-bf16",
443
+ "nbytes": 2818048,
444
+ "byteOffset": 18219008
445
+ },
446
+ {
447
+ "name": "model.layers.2.input_layernorm.weight",
448
+ "shape": [
449
+ 4096
450
+ ],
451
+ "dtype": "float16",
452
+ "format": "f32-to-bf16",
453
+ "nbytes": 8192,
454
+ "byteOffset": 21037056
455
+ },
456
+ {
457
+ "name": "model.layers.2.post_attention_layernorm.weight",
458
+ "shape": [
459
+ 4096
460
+ ],
461
+ "dtype": "float16",
462
+ "format": "f32-to-bf16",
463
+ "nbytes": 8192,
464
+ "byteOffset": 21045248
465
+ },
466
+ {
467
+ "name": "model.layers.3.self_attn.qkv_proj.q_scale",
468
+ "shape": [
469
+ 128,
470
+ 12288
471
+ ],
472
+ "dtype": "float16",
473
+ "format": "f32-to-bf16",
474
+ "nbytes": 3145728,
475
+ "byteOffset": 21053440
476
+ },
477
+ {
478
+ "name": "model.layers.3.self_attn.o_proj.q_weight",
479
+ "shape": [
480
+ 512,
481
+ 4096
482
+ ],
483
+ "dtype": "uint32",
484
+ "format": "f32-to-bf16",
485
+ "nbytes": 8388608,
486
+ "byteOffset": 24199168
487
+ }
488
+ ],
489
+ "md5sum": "2fe4e2d31ded29ecbc97713c15cf3065"
490
+ },
491
+ {
492
+ "dataPath": "params_shard_13.bin",
493
+ "format": "raw-shard",
494
+ "nbytes": 45088768,
495
+ "records": [
496
+ {
497
+ "name": "model.layers.3.mlp.gate_up_proj.q_weight",
498
+ "shape": [
499
+ 512,
500
+ 22016
501
+ ],
502
+ "dtype": "uint32",
503
+ "format": "f32-to-bf16",
504
+ "nbytes": 45088768,
505
+ "byteOffset": 0
506
+ }
507
+ ],
508
+ "md5sum": "b245482ff81bc17b05d1deaba369fddc"
509
+ },
510
+ {
511
+ "dataPath": "params_shard_14.bin",
512
+ "format": "raw-shard",
513
+ "nbytes": 25165824,
514
+ "records": [
515
+ {
516
+ "name": "model.layers.4.self_attn.qkv_proj.q_weight",
517
+ "shape": [
518
+ 512,
519
+ 12288
520
+ ],
521
+ "dtype": "uint32",
522
+ "format": "f32-to-bf16",
523
+ "nbytes": 25165824,
524
+ "byteOffset": 0
525
+ }
526
+ ],
527
+ "md5sum": "3c69628ac1609e53340a59ba0d171bf9"
528
+ },
529
+ {
530
+ "dataPath": "params_shard_15.bin",
531
+ "format": "raw-shard",
532
+ "nbytes": 32063488,
533
+ "records": [
534
+ {
535
+ "name": "model.layers.3.self_attn.o_proj.q_scale",
536
+ "shape": [
537
+ 128,
538
+ 4096
539
+ ],
540
+ "dtype": "float16",
541
+ "format": "f32-to-bf16",
542
+ "nbytes": 1048576,
543
+ "byteOffset": 0
544
+ },
545
+ {
546
+ "name": "model.layers.3.mlp.gate_up_proj.q_scale",
547
+ "shape": [
548
+ 128,
549
+ 22016
550
+ ],
551
+ "dtype": "float16",
552
+ "format": "f32-to-bf16",
553
+ "nbytes": 5636096,
554
+ "byteOffset": 1048576
555
+ },
556
+ {
557
+ "name": "model.layers.3.mlp.down_proj.q_weight",
558
+ "shape": [
559
+ 1376,
560
+ 4096
561
+ ],
562
+ "dtype": "uint32",
563
+ "format": "f32-to-bf16",
564
+ "nbytes": 22544384,
565
+ "byteOffset": 6684672
566
+ },
567
+ {
568
+ "name": "model.layers.3.mlp.down_proj.q_scale",
569
+ "shape": [
570
+ 344,
571
+ 4096
572
+ ],
573
+ "dtype": "float16",
574
+ "format": "f32-to-bf16",
575
+ "nbytes": 2818048,
576
+ "byteOffset": 29229056
577
+ },
578
+ {
579
+ "name": "model.layers.3.input_layernorm.weight",
580
+ "shape": [
581
+ 4096
582
+ ],
583
+ "dtype": "float16",
584
+ "format": "f32-to-bf16",
585
+ "nbytes": 8192,
586
+ "byteOffset": 32047104
587
+ },
588
+ {
589
+ "name": "model.layers.3.post_attention_layernorm.weight",
590
+ "shape": [
591
+ 4096
592
+ ],
593
+ "dtype": "float16",
594
+ "format": "f32-to-bf16",
595
+ "nbytes": 8192,
596
+ "byteOffset": 32055296
597
+ }
598
+ ],
599
+ "md5sum": "756331ddf99ea961d49b8df7b6008ed1"
600
+ },
601
+ {
602
+ "dataPath": "params_shard_16.bin",
603
+ "format": "raw-shard",
604
+ "nbytes": 45088768,
605
+ "records": [
606
+ {
607
+ "name": "model.layers.4.mlp.gate_up_proj.q_weight",
608
+ "shape": [
609
+ 512,
610
+ 22016
611
+ ],
612
+ "dtype": "uint32",
613
+ "format": "f32-to-bf16",
614
+ "nbytes": 45088768,
615
+ "byteOffset": 0
616
+ }
617
+ ],
618
+ "md5sum": "faaae6a9aea2564ef9c18b82b6cc119c"
619
+ },
620
+ {
621
+ "dataPath": "params_shard_17.bin",
622
+ "format": "raw-shard",
623
+ "nbytes": 22544384,
624
+ "records": [
625
+ {
626
+ "name": "model.layers.4.mlp.down_proj.q_weight",
627
+ "shape": [
628
+ 1376,
629
+ 4096
630
+ ],
631
+ "dtype": "uint32",
632
+ "format": "f32-to-bf16",
633
+ "nbytes": 22544384,
634
+ "byteOffset": 0
635
+ }
636
+ ],
637
+ "md5sum": "325c866363e7b74b4ab16daf82c0c166"
638
+ },
639
+ {
640
+ "dataPath": "params_shard_18.bin",
641
+ "format": "raw-shard",
642
+ "nbytes": 25165824,
643
+ "records": [
644
+ {
645
+ "name": "model.layers.5.self_attn.qkv_proj.q_weight",
646
+ "shape": [
647
+ 512,
648
+ 12288
649
+ ],
650
+ "dtype": "uint32",
651
+ "format": "f32-to-bf16",
652
+ "nbytes": 25165824,
653
+ "byteOffset": 0
654
+ }
655
+ ],
656
+ "md5sum": "be9918024a9f24380a415b53edf2dd9f"
657
+ },
658
+ {
659
+ "dataPath": "params_shard_19.bin",
660
+ "format": "raw-shard",
661
+ "nbytes": 32587776,
662
+ "records": [
663
+ {
664
+ "name": "model.layers.4.self_attn.qkv_proj.q_scale",
665
+ "shape": [
666
+ 128,
667
+ 12288
668
+ ],
669
+ "dtype": "float16",
670
+ "format": "f32-to-bf16",
671
+ "nbytes": 3145728,
672
+ "byteOffset": 0
673
+ },
674
+ {
675
+ "name": "model.layers.4.self_attn.o_proj.q_weight",
676
+ "shape": [
677
+ 512,
678
+ 4096
679
+ ],
680
+ "dtype": "uint32",
681
+ "format": "f32-to-bf16",
682
+ "nbytes": 8388608,
683
+ "byteOffset": 3145728
684
+ },
685
+ {
686
+ "name": "model.layers.4.self_attn.o_proj.q_scale",
687
+ "shape": [
688
+ 128,
689
+ 4096
690
+ ],
691
+ "dtype": "float16",
692
+ "format": "f32-to-bf16",
693
+ "nbytes": 1048576,
694
+ "byteOffset": 11534336
695
+ },
696
+ {
697
+ "name": "model.layers.4.mlp.gate_up_proj.q_scale",
698
+ "shape": [
699
+ 128,
700
+ 22016
701
+ ],
702
+ "dtype": "float16",
703
+ "format": "f32-to-bf16",
704
+ "nbytes": 5636096,
705
+ "byteOffset": 12582912
706
+ },
707
+ {
708
+ "name": "model.layers.4.mlp.down_proj.q_scale",
709
+ "shape": [
710
+ 344,
711
+ 4096
712
+ ],
713
+ "dtype": "float16",
714
+ "format": "f32-to-bf16",
715
+ "nbytes": 2818048,
716
+ "byteOffset": 18219008
717
+ },
718
+ {
719
+ "name": "model.layers.4.input_layernorm.weight",
720
+ "shape": [
721
+ 4096
722
+ ],
723
+ "dtype": "float16",
724
+ "format": "f32-to-bf16",
725
+ "nbytes": 8192,
726
+ "byteOffset": 21037056
727
+ },
728
+ {
729
+ "name": "model.layers.4.post_attention_layernorm.weight",
730
+ "shape": [
731
+ 4096
732
+ ],
733
+ "dtype": "float16",
734
+ "format": "f32-to-bf16",
735
+ "nbytes": 8192,
736
+ "byteOffset": 21045248
737
+ },
738
+ {
739
+ "name": "model.layers.5.self_attn.qkv_proj.q_scale",
740
+ "shape": [
741
+ 128,
742
+ 12288
743
+ ],
744
+ "dtype": "float16",
745
+ "format": "f32-to-bf16",
746
+ "nbytes": 3145728,
747
+ "byteOffset": 21053440
748
+ },
749
+ {
750
+ "name": "model.layers.5.self_attn.o_proj.q_weight",
751
+ "shape": [
752
+ 512,
753
+ 4096
754
+ ],
755
+ "dtype": "uint32",
756
+ "format": "f32-to-bf16",
757
+ "nbytes": 8388608,
758
+ "byteOffset": 24199168
759
+ }
760
+ ],
761
+ "md5sum": "44c124c1dc4bc71dab03063e8902598f"
762
+ },
763
+ {
764
+ "dataPath": "params_shard_20.bin",
765
+ "format": "raw-shard",
766
+ "nbytes": 45088768,
767
+ "records": [
768
+ {
769
+ "name": "model.layers.5.mlp.gate_up_proj.q_weight",
770
+ "shape": [
771
+ 512,
772
+ 22016
773
+ ],
774
+ "dtype": "uint32",
775
+ "format": "f32-to-bf16",
776
+ "nbytes": 45088768,
777
+ "byteOffset": 0
778
+ }
779
+ ],
780
+ "md5sum": "2704865ad9cc478fcb1d00bea211f60b"
781
+ },
782
+ {
783
+ "dataPath": "params_shard_21.bin",
784
+ "format": "raw-shard",
785
+ "nbytes": 65536000,
786
+ "records": [
787
+ {
788
+ "name": "lm_head.q_weight",
789
+ "shape": [
790
+ 512,
791
+ 32000
792
+ ],
793
+ "dtype": "uint32",
794
+ "format": "f32-to-bf16",
795
+ "nbytes": 65536000,
796
+ "byteOffset": 0
797
+ }
798
+ ],
799
+ "md5sum": "3ab5243697098da96f215d9c7fe2a7ca"
800
+ },
801
+ {
802
+ "dataPath": "params_shard_22.bin",
803
+ "format": "raw-shard",
804
+ "nbytes": 32071680,
805
+ "records": [
806
+ {
807
+ "name": "model.layers.5.self_attn.o_proj.q_scale",
808
+ "shape": [
809
+ 128,
810
+ 4096
811
+ ],
812
+ "dtype": "float16",
813
+ "format": "f32-to-bf16",
814
+ "nbytes": 1048576,
815
+ "byteOffset": 0
816
+ },
817
+ {
818
+ "name": "model.layers.5.mlp.gate_up_proj.q_scale",
819
+ "shape": [
820
+ 128,
821
+ 22016
822
+ ],
823
+ "dtype": "float16",
824
+ "format": "f32-to-bf16",
825
+ "nbytes": 5636096,
826
+ "byteOffset": 1048576
827
+ },
828
+ {
829
+ "name": "model.layers.5.mlp.down_proj.q_weight",
830
+ "shape": [
831
+ 1376,
832
+ 4096
833
+ ],
834
+ "dtype": "uint32",
835
+ "format": "f32-to-bf16",
836
+ "nbytes": 22544384,
837
+ "byteOffset": 6684672
838
+ },
839
+ {
840
+ "name": "model.layers.5.mlp.down_proj.q_scale",
841
+ "shape": [
842
+ 344,
843
+ 4096
844
+ ],
845
+ "dtype": "float16",
846
+ "format": "f32-to-bf16",
847
+ "nbytes": 2818048,
848
+ "byteOffset": 29229056
849
+ },
850
+ {
851
+ "name": "model.layers.5.input_layernorm.weight",
852
+ "shape": [
853
+ 4096
854
+ ],
855
+ "dtype": "float16",
856
+ "format": "f32-to-bf16",
857
+ "nbytes": 8192,
858
+ "byteOffset": 32047104
859
+ },
860
+ {
861
+ "name": "model.layers.5.post_attention_layernorm.weight",
862
+ "shape": [
863
+ 4096
864
+ ],
865
+ "dtype": "float16",
866
+ "format": "f32-to-bf16",
867
+ "nbytes": 8192,
868
+ "byteOffset": 32055296
869
+ },
870
+ {
871
+ "name": "model.norm.weight",
872
+ "shape": [
873
+ 4096
874
+ ],
875
+ "dtype": "float16",
876
+ "format": "f32-to-bf16",
877
+ "nbytes": 8192,
878
+ "byteOffset": 32063488
879
+ }
880
+ ],
881
+ "md5sum": "ca0268efd419df23435db72a5aa6133b"
882
+ },
883
+ {
884
+ "dataPath": "params_shard_23.bin",
885
+ "format": "raw-shard",
886
+ "nbytes": 8192000,
887
+ "records": [
888
+ {
889
+ "name": "lm_head.q_scale",
890
+ "shape": [
891
+ 128,
892
+ 32000
893
+ ],
894
+ "dtype": "float16",
895
+ "format": "f32-to-bf16",
896
+ "nbytes": 8192000,
897
+ "byteOffset": 0
898
+ }
899
+ ],
900
+ "md5sum": "4c61f062381dc9b41123ce288cc6a887"
901
+ }
902
+ ]
903
+ }
params_shard_0.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a766fa5b3b9ed5715cc9a106a9478447d61af9b3110874b6121d6cf07b48ac6f
3
+ size 65536000
params_shard_1.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c8a829ded1fa4143d44553efd1de6dd6ccb16a4eb3670400940ebbfb950ab7a
3
+ size 33357824
params_shard_10.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:860f798ba01afbadeaf84572812e2934f236d5da0a8dd98b5223a9044dfb5f73
3
+ size 22544384
params_shard_11.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a750565abffee0b8cb90e656d39c930a75614e292a2cf1235f6907a17aec251d
3
+ size 25165824
params_shard_12.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51d64239db28072e47bb286665ebd6ab85fc765173db8cfdd799cfedaf9e1379
3
+ size 32587776
params_shard_13.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e8804d51060552fefc6013b0b4e652d6a3461ef40fe1e62bb3050b2a3a2d936
3
+ size 45088768
params_shard_14.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:522d43f25215d2bf86ed7d0f99adc0093c9c7ebee846651c4da5fe78d02e11d0
3
+ size 25165824
params_shard_15.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf98e27b3aa7ff1406cdb9c2021714a96cc3c93b5ad103afe70a3639d75af72e
3
+ size 32063488
params_shard_16.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8789a02ade9bf2ecf122fbde883ea271ca468e483c512416867e32bd612638f3
3
+ size 45088768
params_shard_17.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d086eb233e77b877b1e5c5a2734ef3056114ab9a15dd66ca8b0f2d5e5c449cad
3
+ size 22544384
params_shard_18.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89adc67b3d9e73211ebc00aab857d71309f1a16c359a8f8c99b6f6a4f81d2276
3
+ size 25165824
params_shard_19.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:834a75dddd5b69cb3694ffe5a35a50da15ec568450c30bab85bb1e73ebd76dad
3
+ size 32587776
params_shard_2.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:64a8feebb1ba2712b4ebaf0b75d9cf69d1d948a323d5c35d72508ca8d5d544c7
3
+ size 45088768
params_shard_20.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f42476be6571deb6e8b35ecec28e6613647fd96deb39849958fdb5f5919a841f
3
+ size 45088768
params_shard_21.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ee49dfed4a6787435c0425af2fde4b2f9ece24585366a24930963d8cc9a0e5f
3
+ size 65536000
params_shard_22.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cba0e369fc1def7df3646b772c102f85a0503ea499bf4572553ed4079cb1ac34
3
+ size 32071680
params_shard_23.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6ae01c2d825b8d595f783e5b9d8891b14ef3130b227818f560ac9fe4c58d38a
3
+ size 8192000
params_shard_3.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:966c800d3f02485839b80baf195ecc87ffae37f94a38f9524a9925c2cbd1ecbd
3
+ size 22544384
params_shard_4.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1a2d392a6d5a378e80e4b8484b5290db4bb399f4ffcc7f8c07498687f26b200
3
+ size 25165824
params_shard_5.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad393502e93e2571f94530eac38a2c3e6ec40f9f905c775ea9a73f941cf3c6bf
3
+ size 32587776
params_shard_6.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9812df18711bf64739a436040cc749f48b33b1ae839d7fb8441be97f6b22e2ed
3
+ size 45088768
params_shard_7.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6184301e856579b324649d278cbcd358b0b0ebfc8df7a6f49ae1dadcae475104
3
+ size 25165824
params_shard_8.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9fd38b3d6d3978a9d507dacec0c16145265c925cd8b227aea77ce2c670081f0
3
+ size 32063488
params_shard_9.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b83a48d5076324826afd793aee6a4cb25bbd8c75030fbbd56df316015f39fbf
3
+ size 45088768
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 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "bos_token": {
5
+ "__type": "AddedToken",
6
+ "content": "<s>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "clean_up_tokenization_spaces": false,
13
+ "eos_token": {
14
+ "__type": "AddedToken",
15
+ "content": "</s>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "legacy": true,
22
+ "model_max_length": 2048,
23
+ "pad_token": null,
24
+ "padding_side": "right",
25
+ "sp_model_kwargs": {},
26
+ "tokenizer_class": "LlamaTokenizer",
27
+ "unk_token": {
28
+ "__type": "AddedToken",
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false
34
+ }
35
+ }