Text Generation
Adapters
Safetensors
Turkish
llama
burak commited on
Commit
9d2138c
·
verified ·
1 Parent(s): b3f049e

Upload folder using huggingface_hub

Browse files
checkpoint-1500/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "dtype": "float32",
9
+ "eos_token_id": 0,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2816,
15
+ "max_position_embeddings": 1024,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 22,
20
+ "num_key_value_heads": 16,
21
+ "pad_token_id": null,
22
+ "pretraining_tp": 1,
23
+ "rms_norm_eps": 1e-06,
24
+ "rope_parameters": {
25
+ "rope_theta": 10000.0,
26
+ "rope_type": "default"
27
+ },
28
+ "tie_word_embeddings": false,
29
+ "transformers_version": "5.1.0",
30
+ "use_cache": false,
31
+ "vocab_size": 32002
32
+ }
checkpoint-1500/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 0,
5
+ "output_attentions": false,
6
+ "output_hidden_states": false,
7
+ "transformers_version": "5.1.0",
8
+ "use_cache": false
9
+ }
checkpoint-1500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92e71dd50419c7610c05b20c82f9c8e2406ffc43e5742a0b8f6fe2cf5b0338a1
3
+ size 1392732464
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:61afb01fe17939bc5b93a247fcdb5bb1b40c5abf62eedd925b9b2b88e077e8ca
3
+ size 2785591947
checkpoint-1500/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5101a4e3f9bef652df36b6cdfe84db2d22c939a4c41f5f878b26d4c05dbde68a
3
+ size 14917
checkpoint-1500/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c2be6540f91a23cc821ef44837aff7fdec0c3227e3354b58fa742ee652715b3
3
+ size 14917
checkpoint-1500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca372268f4fa9335030c0cb7aedb6cdba75f457da50e7a4034abb1a2d0843689
3
+ size 1383
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8486c06adb38c266ede7bcff4417549f208b7fd28fafdcdfcc051b6ff3bdb4f3
3
+ size 1465
checkpoint-1500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<|endoftext|>",
4
+ "eos_token": "<|endoftext|>",
5
+ "extra_special_tokens": [
6
+ "<|im_start|>",
7
+ "<|im_end|>"
8
+ ],
9
+ "is_local": false,
10
+ "model_max_length": 1000000000000000019884624838656,
11
+ "pad_token": "<|pad|>",
12
+ "tokenizer_class": "TokenizersBackend"
13
+ }
checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,559 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.96,
6
+ "eval_steps": 500,
7
+ "global_step": 1500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0128,
14
+ "grad_norm": 1.6443285942077637,
15
+ "learning_rate": 1.2127659574468084e-05,
16
+ "loss": 4.572507858276367,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.0256,
21
+ "grad_norm": 1.156632661819458,
22
+ "learning_rate": 2.4893617021276595e-05,
23
+ "loss": 3.922951507568359,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.0384,
28
+ "grad_norm": 1.1741294860839844,
29
+ "learning_rate": 2.9995362299296104e-05,
30
+ "loss": 3.7445632934570314,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.0512,
35
+ "grad_norm": 1.1301523447036743,
36
+ "learning_rate": 2.996703117966497e-05,
37
+ "loss": 3.595957946777344,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.064,
42
+ "grad_norm": 1.1449520587921143,
43
+ "learning_rate": 2.9912994040750924e-05,
44
+ "loss": 3.5559593200683595,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.0768,
49
+ "grad_norm": 1.24018394947052,
50
+ "learning_rate": 2.9833343691857188e-05,
51
+ "loss": 3.438861846923828,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.0896,
56
+ "grad_norm": 1.0804022550582886,
57
+ "learning_rate": 2.972821693322021e-05,
58
+ "loss": 3.4017875671386717,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.1024,
63
+ "grad_norm": 1.0477685928344727,
64
+ "learning_rate": 2.9597794321054012e-05,
65
+ "loss": 3.282175064086914,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.1152,
70
+ "grad_norm": 1.1796107292175293,
71
+ "learning_rate": 2.9442299857443096e-05,
72
+ "loss": 3.298114776611328,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.128,
77
+ "grad_norm": 1.053169846534729,
78
+ "learning_rate": 2.9262000605616772e-05,
79
+ "loss": 3.186061477661133,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.1408,
84
+ "grad_norm": 1.1376056671142578,
85
+ "learning_rate": 2.905720623126549e-05,
86
+ "loss": 3.1561294555664063,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.1536,
91
+ "grad_norm": 1.128514289855957,
92
+ "learning_rate": 2.8828268470687036e-05,
93
+ "loss": 3.115443801879883,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1664,
98
+ "grad_norm": 1.0387954711914062,
99
+ "learning_rate": 2.85755805266761e-05,
100
+ "loss": 3.107771873474121,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.1792,
105
+ "grad_norm": 1.0744367837905884,
106
+ "learning_rate": 2.82995763931947e-05,
107
+ "loss": 3.107578086853027,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.192,
112
+ "grad_norm": 1.0310322046279907,
113
+ "learning_rate": 2.8000730109983406e-05,
114
+ "loss": 3.0090654373168944,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.2048,
119
+ "grad_norm": 1.0029125213623047,
120
+ "learning_rate": 2.7679554948393527e-05,
121
+ "loss": 3.0286556243896485,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.2176,
126
+ "grad_norm": 0.9451630115509033,
127
+ "learning_rate": 2.7336602529838696e-05,
128
+ "loss": 3.0148574829101564,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.2304,
133
+ "grad_norm": 0.9913703203201294,
134
+ "learning_rate": 2.6972461878379766e-05,
135
+ "loss": 3.0343597412109373,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.2432,
140
+ "grad_norm": 0.9380889534950256,
141
+ "learning_rate": 2.6587758409070383e-05,
142
+ "loss": 3.044027328491211,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.256,
147
+ "grad_norm": 0.9598795175552368,
148
+ "learning_rate": 2.618315285380063e-05,
149
+ "loss": 2.9781471252441407,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.2688,
154
+ "grad_norm": 1.0529745817184448,
155
+ "learning_rate": 2.5759340126483685e-05,
156
+ "loss": 3.013719177246094,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.2816,
161
+ "grad_norm": 1.0288140773773193,
162
+ "learning_rate": 2.531704812953453e-05,
163
+ "loss": 2.964109420776367,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2944,
168
+ "grad_norm": 0.9722886085510254,
169
+ "learning_rate": 2.4857036503690573e-05,
170
+ "loss": 2.9432010650634766,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.3072,
175
+ "grad_norm": 1.0502101182937622,
176
+ "learning_rate": 2.4380095323321438e-05,
177
+ "loss": 2.959673309326172,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.32,
182
+ "grad_norm": 0.9286924004554749,
183
+ "learning_rate": 2.388704373946865e-05,
184
+ "loss": 2.8894092559814455,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.3328,
189
+ "grad_norm": 0.9448413252830505,
190
+ "learning_rate": 2.3378728572945893e-05,
191
+ "loss": 2.9437007904052734,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.3456,
196
+ "grad_norm": 0.9626290202140808,
197
+ "learning_rate": 2.285602285991616e-05,
198
+ "loss": 2.91724853515625,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3584,
203
+ "grad_norm": 0.9125235080718994,
204
+ "learning_rate": 2.231982435244377e-05,
205
+ "loss": 2.887943649291992,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3712,
210
+ "grad_norm": 0.9829279184341431,
211
+ "learning_rate": 2.1771053976596635e-05,
212
+ "loss": 2.87862548828125,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.384,
217
+ "grad_norm": 0.9327661395072937,
218
+ "learning_rate": 2.1210654250746914e-05,
219
+ "loss": 2.913781929016113,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3968,
224
+ "grad_norm": 0.9741169810295105,
225
+ "learning_rate": 2.0639587666786676e-05,
226
+ "loss": 2.9079227447509766,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.4096,
231
+ "grad_norm": 0.9748414158821106,
232
+ "learning_rate": 2.0058835037038875e-05,
233
+ "loss": 2.925547409057617,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.4224,
238
+ "grad_norm": 1.0026845932006836,
239
+ "learning_rate": 1.946939380970281e-05,
240
+ "loss": 2.8957664489746096,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.4352,
245
+ "grad_norm": 1.0120295286178589,
246
+ "learning_rate": 1.887227635572727e-05,
247
+ "loss": 2.813709259033203,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.448,
252
+ "grad_norm": 0.9583562612533569,
253
+ "learning_rate": 1.8268508230053774e-05,
254
+ "loss": 2.8637973785400392,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.4608,
259
+ "grad_norm": 1.0258755683898926,
260
+ "learning_rate": 1.7659126410216118e-05,
261
+ "loss": 2.8567317962646483,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.4736,
266
+ "grad_norm": 0.999519407749176,
267
+ "learning_rate": 1.7045177515321544e-05,
268
+ "loss": 2.85314826965332,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.4864,
273
+ "grad_norm": 1.0208712816238403,
274
+ "learning_rate": 1.642771600847242e-05,
275
+ "loss": 2.796009826660156,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.4992,
280
+ "grad_norm": 0.9479166269302368,
281
+ "learning_rate": 1.5807802385715758e-05,
282
+ "loss": 2.8422531127929687,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.512,
287
+ "grad_norm": 1.0228406190872192,
288
+ "learning_rate": 1.5186501354631101e-05,
289
+ "loss": 2.8279146194458007,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.5248,
294
+ "grad_norm": 1.0206751823425293,
295
+ "learning_rate": 1.456488000568508e-05,
296
+ "loss": 2.8559734344482424,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.5376,
301
+ "grad_norm": 1.082664966583252,
302
+ "learning_rate": 1.3944005979493312e-05,
303
+ "loss": 2.797427177429199,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.5504,
308
+ "grad_norm": 1.0255582332611084,
309
+ "learning_rate": 1.3324945633137446e-05,
310
+ "loss": 2.8546092987060545,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.5632,
315
+ "grad_norm": 1.0040563344955444,
316
+ "learning_rate": 1.270876220868664e-05,
317
+ "loss": 2.8626388549804687,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.576,
322
+ "grad_norm": 0.9735684990882874,
323
+ "learning_rate": 1.209651400706915e-05,
324
+ "loss": 2.832137680053711,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.5888,
329
+ "grad_norm": 1.0309313535690308,
330
+ "learning_rate": 1.148925257043037e-05,
331
+ "loss": 2.7832464218139648,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.6016,
336
+ "grad_norm": 0.9424602389335632,
337
+ "learning_rate": 1.0888020876099103e-05,
338
+ "loss": 2.7802665710449217,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.6144,
343
+ "grad_norm": 1.0042119026184082,
344
+ "learning_rate": 1.0293851545264042e-05,
345
+ "loss": 2.814408874511719,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.6272,
350
+ "grad_norm": 1.0467122793197632,
351
+ "learning_rate": 9.707765069437001e-06,
352
+ "loss": 2.7961553573608398,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.64,
357
+ "grad_norm": 0.9580904245376587,
358
+ "learning_rate": 9.130768057748953e-06,
359
+ "loss": 2.7776145935058594,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.6528,
364
+ "grad_norm": 1.0763484239578247,
365
+ "learning_rate": 8.563851508089224e-06,
366
+ "loss": 2.785654067993164,
367
+ "step": 1020
368
+ },
369
+ {
370
+ "epoch": 0.6656,
371
+ "grad_norm": 0.9613010883331299,
372
+ "learning_rate": 8.007989105057156e-06,
373
+ "loss": 2.8012434005737306,
374
+ "step": 1040
375
+ },
376
+ {
377
+ "epoch": 0.6784,
378
+ "grad_norm": 1.0293607711791992,
379
+ "learning_rate": 7.464135547649457e-06,
380
+ "loss": 2.8345165252685547,
381
+ "step": 1060
382
+ },
383
+ {
384
+ "epoch": 0.6912,
385
+ "grad_norm": 1.1513640880584717,
386
+ "learning_rate": 6.933224909555556e-06,
387
+ "loss": 2.7340419769287108,
388
+ "step": 1080
389
+ },
390
+ {
391
+ "epoch": 0.704,
392
+ "grad_norm": 0.9785656929016113,
393
+ "learning_rate": 6.4161690348771e-06,
394
+ "loss": 2.7901845932006837,
395
+ "step": 1100
396
+ },
397
+ {
398
+ "epoch": 0.7168,
399
+ "grad_norm": 1.0474238395690918,
400
+ "learning_rate": 5.913855972026982e-06,
401
+ "loss": 2.7927894592285156,
402
+ "step": 1120
403
+ },
404
+ {
405
+ "epoch": 0.7296,
406
+ "grad_norm": 1.0646806955337524,
407
+ "learning_rate": 5.42714844849769e-06,
408
+ "loss": 2.7932479858398436,
409
+ "step": 1140
410
+ },
411
+ {
412
+ "epoch": 0.7424,
413
+ "grad_norm": 0.9252598285675049,
414
+ "learning_rate": 4.9568823891185985e-06,
415
+ "loss": 2.7005069732666014,
416
+ "step": 1160
417
+ },
418
+ {
419
+ "epoch": 0.7552,
420
+ "grad_norm": 0.9208779335021973,
421
+ "learning_rate": 4.50386548034705e-06,
422
+ "loss": 2.7948162078857424,
423
+ "step": 1180
424
+ },
425
+ {
426
+ "epoch": 0.768,
427
+ "grad_norm": 0.988008439540863,
428
+ "learning_rate": 4.0688757830591534e-06,
429
+ "loss": 2.753126525878906,
430
+ "step": 1200
431
+ },
432
+ {
433
+ "epoch": 0.7808,
434
+ "grad_norm": 1.1100376844406128,
435
+ "learning_rate": 3.652660396222766e-06,
436
+ "loss": 2.7805652618408203,
437
+ "step": 1220
438
+ },
439
+ {
440
+ "epoch": 0.7936,
441
+ "grad_norm": 0.9880450367927551,
442
+ "learning_rate": 3.255934173747897e-06,
443
+ "loss": 2.7673030853271485,
444
+ "step": 1240
445
+ },
446
+ {
447
+ "epoch": 0.8064,
448
+ "grad_norm": 1.0234110355377197,
449
+ "learning_rate": 2.8793784967182452e-06,
450
+ "loss": 2.744676971435547,
451
+ "step": 1260
452
+ },
453
+ {
454
+ "epoch": 0.8192,
455
+ "grad_norm": 0.8995652198791504,
456
+ "learning_rate": 2.5236401031126716e-06,
457
+ "loss": 2.743621826171875,
458
+ "step": 1280
459
+ },
460
+ {
461
+ "epoch": 0.832,
462
+ "grad_norm": 0.9905585646629333,
463
+ "learning_rate": 2.1893299770265585e-06,
464
+ "loss": 2.7316511154174803,
465
+ "step": 1300
466
+ },
467
+ {
468
+ "epoch": 0.8448,
469
+ "grad_norm": 0.9099847078323364,
470
+ "learning_rate": 1.8770222993007425e-06,
471
+ "loss": 2.7857704162597656,
472
+ "step": 1320
473
+ },
474
+ {
475
+ "epoch": 0.8576,
476
+ "grad_norm": 1.0823032855987549,
477
+ "learning_rate": 1.587253461360445e-06,
478
+ "loss": 2.746516799926758,
479
+ "step": 1340
480
+ },
481
+ {
482
+ "epoch": 0.8704,
483
+ "grad_norm": 0.9655073285102844,
484
+ "learning_rate": 1.3205211439578658e-06,
485
+ "loss": 2.71276912689209,
486
+ "step": 1360
487
+ },
488
+ {
489
+ "epoch": 0.8832,
490
+ "grad_norm": 1.0540977716445923,
491
+ "learning_rate": 1.0772834624007266e-06,
492
+ "loss": 2.7274484634399414,
493
+ "step": 1380
494
+ },
495
+ {
496
+ "epoch": 0.896,
497
+ "grad_norm": 0.9392573237419128,
498
+ "learning_rate": 8.579581797348379e-07,
499
+ "loss": 2.7726985931396486,
500
+ "step": 1400
501
+ },
502
+ {
503
+ "epoch": 0.9088,
504
+ "grad_norm": 0.9293719530105591,
505
+ "learning_rate": 6.629219892320898e-07,
506
+ "loss": 2.750566864013672,
507
+ "step": 1420
508
+ },
509
+ {
510
+ "epoch": 0.9216,
511
+ "grad_norm": 0.9862186908721924,
512
+ "learning_rate": 4.925098674161299e-07,
513
+ "loss": 2.7667617797851562,
514
+ "step": 1440
515
+ },
516
+ {
517
+ "epoch": 0.9344,
518
+ "grad_norm": 0.9760239124298096,
519
+ "learning_rate": 3.4701449873698774e-07,
520
+ "loss": 2.8068206787109373,
521
+ "step": 1460
522
+ },
523
+ {
524
+ "epoch": 0.9472,
525
+ "grad_norm": 1.0637469291687012,
526
+ "learning_rate": 2.2668577288273707e-07,
527
+ "loss": 2.7901378631591798,
528
+ "step": 1480
529
+ },
530
+ {
531
+ "epoch": 0.96,
532
+ "grad_norm": 0.9915500283241272,
533
+ "learning_rate": 1.3173035559155888e-07,
534
+ "loss": 2.7704195022583007,
535
+ "step": 1500
536
+ }
537
+ ],
538
+ "logging_steps": 20,
539
+ "max_steps": 1563,
540
+ "num_input_tokens_seen": 0,
541
+ "num_train_epochs": 1,
542
+ "save_steps": 500,
543
+ "stateful_callbacks": {
544
+ "TrainerControl": {
545
+ "args": {
546
+ "should_epoch_stop": false,
547
+ "should_evaluate": false,
548
+ "should_log": false,
549
+ "should_save": true,
550
+ "should_training_stop": false
551
+ },
552
+ "attributes": {}
553
+ }
554
+ },
555
+ "total_flos": 9.301741535232e+16,
556
+ "train_batch_size": 16,
557
+ "trial_name": null,
558
+ "trial_params": null
559
+ }
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc3223a9d0c4357d0ca82ae03ede3de1f8ac8a2833cb38a3f794c4f4ccb4297b
3
+ size 5201
checkpoint-1563/config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "dtype": "float32",
9
+ "eos_token_id": 0,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2816,
15
+ "max_position_embeddings": 1024,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 22,
20
+ "num_key_value_heads": 16,
21
+ "pad_token_id": null,
22
+ "pretraining_tp": 1,
23
+ "rms_norm_eps": 1e-06,
24
+ "rope_parameters": {
25
+ "rope_theta": 10000.0,
26
+ "rope_type": "default"
27
+ },
28
+ "tie_word_embeddings": false,
29
+ "transformers_version": "5.1.0",
30
+ "use_cache": false,
31
+ "vocab_size": 32002
32
+ }
checkpoint-1563/generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 0,
5
+ "output_attentions": false,
6
+ "output_hidden_states": false,
7
+ "transformers_version": "5.1.0",
8
+ "use_cache": false
9
+ }
checkpoint-1563/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:915f67d2b65cf24ec31619f12fa672984645ff57e18497bb79bc69f485a2a9e4
3
+ size 1392732464
checkpoint-1563/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5213a3460cc2dc0ade2588096a274af3b603eb7c4ec45239d2e0a70cd67eb6b3
3
+ size 2785591947
checkpoint-1563/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5101a4e3f9bef652df36b6cdfe84db2d22c939a4c41f5f878b26d4c05dbde68a
3
+ size 14917
checkpoint-1563/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c2be6540f91a23cc821ef44837aff7fdec0c3227e3354b58fa742ee652715b3
3
+ size 14917
checkpoint-1563/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e701467b512373db873a416bbcad04794b20cfccdbc879516e937be92100c103
3
+ size 1383
checkpoint-1563/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0cdd84bbef3447401e92e5a4362a3eb418017286f0f2f153a1221e5bb930bdc
3
+ size 1465
checkpoint-1563/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1563/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<|endoftext|>",
4
+ "eos_token": "<|endoftext|>",
5
+ "extra_special_tokens": [
6
+ "<|im_start|>",
7
+ "<|im_end|>"
8
+ ],
9
+ "is_local": false,
10
+ "model_max_length": 1000000000000000019884624838656,
11
+ "pad_token": "<|pad|>",
12
+ "tokenizer_class": "TokenizersBackend"
13
+ }
checkpoint-1563/trainer_state.json ADDED
@@ -0,0 +1,580 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1563,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0128,
14
+ "grad_norm": 1.6443285942077637,
15
+ "learning_rate": 1.2127659574468084e-05,
16
+ "loss": 4.572507858276367,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.0256,
21
+ "grad_norm": 1.156632661819458,
22
+ "learning_rate": 2.4893617021276595e-05,
23
+ "loss": 3.922951507568359,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.0384,
28
+ "grad_norm": 1.1741294860839844,
29
+ "learning_rate": 2.9995362299296104e-05,
30
+ "loss": 3.7445632934570314,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.0512,
35
+ "grad_norm": 1.1301523447036743,
36
+ "learning_rate": 2.996703117966497e-05,
37
+ "loss": 3.595957946777344,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.064,
42
+ "grad_norm": 1.1449520587921143,
43
+ "learning_rate": 2.9912994040750924e-05,
44
+ "loss": 3.5559593200683595,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.0768,
49
+ "grad_norm": 1.24018394947052,
50
+ "learning_rate": 2.9833343691857188e-05,
51
+ "loss": 3.438861846923828,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.0896,
56
+ "grad_norm": 1.0804022550582886,
57
+ "learning_rate": 2.972821693322021e-05,
58
+ "loss": 3.4017875671386717,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.1024,
63
+ "grad_norm": 1.0477685928344727,
64
+ "learning_rate": 2.9597794321054012e-05,
65
+ "loss": 3.282175064086914,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.1152,
70
+ "grad_norm": 1.1796107292175293,
71
+ "learning_rate": 2.9442299857443096e-05,
72
+ "loss": 3.298114776611328,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.128,
77
+ "grad_norm": 1.053169846534729,
78
+ "learning_rate": 2.9262000605616772e-05,
79
+ "loss": 3.186061477661133,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.1408,
84
+ "grad_norm": 1.1376056671142578,
85
+ "learning_rate": 2.905720623126549e-05,
86
+ "loss": 3.1561294555664063,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.1536,
91
+ "grad_norm": 1.128514289855957,
92
+ "learning_rate": 2.8828268470687036e-05,
93
+ "loss": 3.115443801879883,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1664,
98
+ "grad_norm": 1.0387954711914062,
99
+ "learning_rate": 2.85755805266761e-05,
100
+ "loss": 3.107771873474121,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.1792,
105
+ "grad_norm": 1.0744367837905884,
106
+ "learning_rate": 2.82995763931947e-05,
107
+ "loss": 3.107578086853027,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.192,
112
+ "grad_norm": 1.0310322046279907,
113
+ "learning_rate": 2.8000730109983406e-05,
114
+ "loss": 3.0090654373168944,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.2048,
119
+ "grad_norm": 1.0029125213623047,
120
+ "learning_rate": 2.7679554948393527e-05,
121
+ "loss": 3.0286556243896485,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.2176,
126
+ "grad_norm": 0.9451630115509033,
127
+ "learning_rate": 2.7336602529838696e-05,
128
+ "loss": 3.0148574829101564,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.2304,
133
+ "grad_norm": 0.9913703203201294,
134
+ "learning_rate": 2.6972461878379766e-05,
135
+ "loss": 3.0343597412109373,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.2432,
140
+ "grad_norm": 0.9380889534950256,
141
+ "learning_rate": 2.6587758409070383e-05,
142
+ "loss": 3.044027328491211,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.256,
147
+ "grad_norm": 0.9598795175552368,
148
+ "learning_rate": 2.618315285380063e-05,
149
+ "loss": 2.9781471252441407,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.2688,
154
+ "grad_norm": 1.0529745817184448,
155
+ "learning_rate": 2.5759340126483685e-05,
156
+ "loss": 3.013719177246094,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.2816,
161
+ "grad_norm": 1.0288140773773193,
162
+ "learning_rate": 2.531704812953453e-05,
163
+ "loss": 2.964109420776367,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2944,
168
+ "grad_norm": 0.9722886085510254,
169
+ "learning_rate": 2.4857036503690573e-05,
170
+ "loss": 2.9432010650634766,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.3072,
175
+ "grad_norm": 1.0502101182937622,
176
+ "learning_rate": 2.4380095323321438e-05,
177
+ "loss": 2.959673309326172,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.32,
182
+ "grad_norm": 0.9286924004554749,
183
+ "learning_rate": 2.388704373946865e-05,
184
+ "loss": 2.8894092559814455,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.3328,
189
+ "grad_norm": 0.9448413252830505,
190
+ "learning_rate": 2.3378728572945893e-05,
191
+ "loss": 2.9437007904052734,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.3456,
196
+ "grad_norm": 0.9626290202140808,
197
+ "learning_rate": 2.285602285991616e-05,
198
+ "loss": 2.91724853515625,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3584,
203
+ "grad_norm": 0.9125235080718994,
204
+ "learning_rate": 2.231982435244377e-05,
205
+ "loss": 2.887943649291992,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3712,
210
+ "grad_norm": 0.9829279184341431,
211
+ "learning_rate": 2.1771053976596635e-05,
212
+ "loss": 2.87862548828125,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.384,
217
+ "grad_norm": 0.9327661395072937,
218
+ "learning_rate": 2.1210654250746914e-05,
219
+ "loss": 2.913781929016113,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3968,
224
+ "grad_norm": 0.9741169810295105,
225
+ "learning_rate": 2.0639587666786676e-05,
226
+ "loss": 2.9079227447509766,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.4096,
231
+ "grad_norm": 0.9748414158821106,
232
+ "learning_rate": 2.0058835037038875e-05,
233
+ "loss": 2.925547409057617,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.4224,
238
+ "grad_norm": 1.0026845932006836,
239
+ "learning_rate": 1.946939380970281e-05,
240
+ "loss": 2.8957664489746096,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.4352,
245
+ "grad_norm": 1.0120295286178589,
246
+ "learning_rate": 1.887227635572727e-05,
247
+ "loss": 2.813709259033203,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.448,
252
+ "grad_norm": 0.9583562612533569,
253
+ "learning_rate": 1.8268508230053774e-05,
254
+ "loss": 2.8637973785400392,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.4608,
259
+ "grad_norm": 1.0258755683898926,
260
+ "learning_rate": 1.7659126410216118e-05,
261
+ "loss": 2.8567317962646483,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.4736,
266
+ "grad_norm": 0.999519407749176,
267
+ "learning_rate": 1.7045177515321544e-05,
268
+ "loss": 2.85314826965332,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.4864,
273
+ "grad_norm": 1.0208712816238403,
274
+ "learning_rate": 1.642771600847242e-05,
275
+ "loss": 2.796009826660156,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.4992,
280
+ "grad_norm": 0.9479166269302368,
281
+ "learning_rate": 1.5807802385715758e-05,
282
+ "loss": 2.8422531127929687,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.512,
287
+ "grad_norm": 1.0228406190872192,
288
+ "learning_rate": 1.5186501354631101e-05,
289
+ "loss": 2.8279146194458007,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.5248,
294
+ "grad_norm": 1.0206751823425293,
295
+ "learning_rate": 1.456488000568508e-05,
296
+ "loss": 2.8559734344482424,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.5376,
301
+ "grad_norm": 1.082664966583252,
302
+ "learning_rate": 1.3944005979493312e-05,
303
+ "loss": 2.797427177429199,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.5504,
308
+ "grad_norm": 1.0255582332611084,
309
+ "learning_rate": 1.3324945633137446e-05,
310
+ "loss": 2.8546092987060545,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.5632,
315
+ "grad_norm": 1.0040563344955444,
316
+ "learning_rate": 1.270876220868664e-05,
317
+ "loss": 2.8626388549804687,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.576,
322
+ "grad_norm": 0.9735684990882874,
323
+ "learning_rate": 1.209651400706915e-05,
324
+ "loss": 2.832137680053711,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.5888,
329
+ "grad_norm": 1.0309313535690308,
330
+ "learning_rate": 1.148925257043037e-05,
331
+ "loss": 2.7832464218139648,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.6016,
336
+ "grad_norm": 0.9424602389335632,
337
+ "learning_rate": 1.0888020876099103e-05,
338
+ "loss": 2.7802665710449217,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.6144,
343
+ "grad_norm": 1.0042119026184082,
344
+ "learning_rate": 1.0293851545264042e-05,
345
+ "loss": 2.814408874511719,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.6272,
350
+ "grad_norm": 1.0467122793197632,
351
+ "learning_rate": 9.707765069437001e-06,
352
+ "loss": 2.7961553573608398,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.64,
357
+ "grad_norm": 0.9580904245376587,
358
+ "learning_rate": 9.130768057748953e-06,
359
+ "loss": 2.7776145935058594,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.6528,
364
+ "grad_norm": 1.0763484239578247,
365
+ "learning_rate": 8.563851508089224e-06,
366
+ "loss": 2.785654067993164,
367
+ "step": 1020
368
+ },
369
+ {
370
+ "epoch": 0.6656,
371
+ "grad_norm": 0.9613010883331299,
372
+ "learning_rate": 8.007989105057156e-06,
373
+ "loss": 2.8012434005737306,
374
+ "step": 1040
375
+ },
376
+ {
377
+ "epoch": 0.6784,
378
+ "grad_norm": 1.0293607711791992,
379
+ "learning_rate": 7.464135547649457e-06,
380
+ "loss": 2.8345165252685547,
381
+ "step": 1060
382
+ },
383
+ {
384
+ "epoch": 0.6912,
385
+ "grad_norm": 1.1513640880584717,
386
+ "learning_rate": 6.933224909555556e-06,
387
+ "loss": 2.7340419769287108,
388
+ "step": 1080
389
+ },
390
+ {
391
+ "epoch": 0.704,
392
+ "grad_norm": 0.9785656929016113,
393
+ "learning_rate": 6.4161690348771e-06,
394
+ "loss": 2.7901845932006837,
395
+ "step": 1100
396
+ },
397
+ {
398
+ "epoch": 0.7168,
399
+ "grad_norm": 1.0474238395690918,
400
+ "learning_rate": 5.913855972026982e-06,
401
+ "loss": 2.7927894592285156,
402
+ "step": 1120
403
+ },
404
+ {
405
+ "epoch": 0.7296,
406
+ "grad_norm": 1.0646806955337524,
407
+ "learning_rate": 5.42714844849769e-06,
408
+ "loss": 2.7932479858398436,
409
+ "step": 1140
410
+ },
411
+ {
412
+ "epoch": 0.7424,
413
+ "grad_norm": 0.9252598285675049,
414
+ "learning_rate": 4.9568823891185985e-06,
415
+ "loss": 2.7005069732666014,
416
+ "step": 1160
417
+ },
418
+ {
419
+ "epoch": 0.7552,
420
+ "grad_norm": 0.9208779335021973,
421
+ "learning_rate": 4.50386548034705e-06,
422
+ "loss": 2.7948162078857424,
423
+ "step": 1180
424
+ },
425
+ {
426
+ "epoch": 0.768,
427
+ "grad_norm": 0.988008439540863,
428
+ "learning_rate": 4.0688757830591534e-06,
429
+ "loss": 2.753126525878906,
430
+ "step": 1200
431
+ },
432
+ {
433
+ "epoch": 0.7808,
434
+ "grad_norm": 1.1100376844406128,
435
+ "learning_rate": 3.652660396222766e-06,
436
+ "loss": 2.7805652618408203,
437
+ "step": 1220
438
+ },
439
+ {
440
+ "epoch": 0.7936,
441
+ "grad_norm": 0.9880450367927551,
442
+ "learning_rate": 3.255934173747897e-06,
443
+ "loss": 2.7673030853271485,
444
+ "step": 1240
445
+ },
446
+ {
447
+ "epoch": 0.8064,
448
+ "grad_norm": 1.0234110355377197,
449
+ "learning_rate": 2.8793784967182452e-06,
450
+ "loss": 2.744676971435547,
451
+ "step": 1260
452
+ },
453
+ {
454
+ "epoch": 0.8192,
455
+ "grad_norm": 0.8995652198791504,
456
+ "learning_rate": 2.5236401031126716e-06,
457
+ "loss": 2.743621826171875,
458
+ "step": 1280
459
+ },
460
+ {
461
+ "epoch": 0.832,
462
+ "grad_norm": 0.9905585646629333,
463
+ "learning_rate": 2.1893299770265585e-06,
464
+ "loss": 2.7316511154174803,
465
+ "step": 1300
466
+ },
467
+ {
468
+ "epoch": 0.8448,
469
+ "grad_norm": 0.9099847078323364,
470
+ "learning_rate": 1.8770222993007425e-06,
471
+ "loss": 2.7857704162597656,
472
+ "step": 1320
473
+ },
474
+ {
475
+ "epoch": 0.8576,
476
+ "grad_norm": 1.0823032855987549,
477
+ "learning_rate": 1.587253461360445e-06,
478
+ "loss": 2.746516799926758,
479
+ "step": 1340
480
+ },
481
+ {
482
+ "epoch": 0.8704,
483
+ "grad_norm": 0.9655073285102844,
484
+ "learning_rate": 1.3205211439578658e-06,
485
+ "loss": 2.71276912689209,
486
+ "step": 1360
487
+ },
488
+ {
489
+ "epoch": 0.8832,
490
+ "grad_norm": 1.0540977716445923,
491
+ "learning_rate": 1.0772834624007266e-06,
492
+ "loss": 2.7274484634399414,
493
+ "step": 1380
494
+ },
495
+ {
496
+ "epoch": 0.896,
497
+ "grad_norm": 0.9392573237419128,
498
+ "learning_rate": 8.579581797348379e-07,
499
+ "loss": 2.7726985931396486,
500
+ "step": 1400
501
+ },
502
+ {
503
+ "epoch": 0.9088,
504
+ "grad_norm": 0.9293719530105591,
505
+ "learning_rate": 6.629219892320898e-07,
506
+ "loss": 2.750566864013672,
507
+ "step": 1420
508
+ },
509
+ {
510
+ "epoch": 0.9216,
511
+ "grad_norm": 0.9862186908721924,
512
+ "learning_rate": 4.925098674161299e-07,
513
+ "loss": 2.7667617797851562,
514
+ "step": 1440
515
+ },
516
+ {
517
+ "epoch": 0.9344,
518
+ "grad_norm": 0.9760239124298096,
519
+ "learning_rate": 3.4701449873698774e-07,
520
+ "loss": 2.8068206787109373,
521
+ "step": 1460
522
+ },
523
+ {
524
+ "epoch": 0.9472,
525
+ "grad_norm": 1.0637469291687012,
526
+ "learning_rate": 2.2668577288273707e-07,
527
+ "loss": 2.7901378631591798,
528
+ "step": 1480
529
+ },
530
+ {
531
+ "epoch": 0.96,
532
+ "grad_norm": 0.9915500283241272,
533
+ "learning_rate": 1.3173035559155888e-07,
534
+ "loss": 2.7704195022583007,
535
+ "step": 1500
536
+ },
537
+ {
538
+ "epoch": 0.9728,
539
+ "grad_norm": 0.9312824010848999,
540
+ "learning_rate": 6.231133370135422e-08,
541
+ "loss": 2.763196563720703,
542
+ "step": 1520
543
+ },
544
+ {
545
+ "epoch": 0.9856,
546
+ "grad_norm": 0.951641321182251,
547
+ "learning_rate": 1.8547935046548392e-08,
548
+ "loss": 2.789643096923828,
549
+ "step": 1540
550
+ },
551
+ {
552
+ "epoch": 0.9984,
553
+ "grad_norm": 0.9073544144630432,
554
+ "learning_rate": 5.153236831406672e-10,
555
+ "loss": 2.784554672241211,
556
+ "step": 1560
557
+ }
558
+ ],
559
+ "logging_steps": 20,
560
+ "max_steps": 1563,
561
+ "num_input_tokens_seen": 0,
562
+ "num_train_epochs": 1,
563
+ "save_steps": 500,
564
+ "stateful_callbacks": {
565
+ "TrainerControl": {
566
+ "args": {
567
+ "should_epoch_stop": false,
568
+ "should_evaluate": false,
569
+ "should_log": false,
570
+ "should_save": true,
571
+ "should_training_stop": true
572
+ },
573
+ "attributes": {}
574
+ }
575
+ },
576
+ "total_flos": 9.6893140992e+16,
577
+ "train_batch_size": 16,
578
+ "trial_name": null,
579
+ "trial_params": null
580
+ }
checkpoint-1563/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc3223a9d0c4357d0ca82ae03ede3de1f8ac8a2833cb38a3f794c4f4ccb4297b
3
+ size 5201
config.json ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "dtype": "float32",
9
+ "eos_token_id": 0,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2816,
15
+ "max_position_embeddings": 1024,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 22,
20
+ "num_key_value_heads": 16,
21
+ "pad_token_id": null,
22
+ "pretraining_tp": 1,
23
+ "rms_norm_eps": 1e-06,
24
+ "rope_parameters": {
25
+ "rope_theta": 10000.0,
26
+ "rope_type": "default"
27
+ },
28
+ "tie_word_embeddings": false,
29
+ "transformers_version": "5.1.0",
30
+ "use_cache": false,
31
+ "vocab_size": 32002
32
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 0,
5
+ "output_attentions": false,
6
+ "output_hidden_states": false,
7
+ "transformers_version": "5.1.0",
8
+ "use_cache": false
9
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:915f67d2b65cf24ec31619f12fa672984645ff57e18497bb79bc69f485a2a9e4
3
+ size 1392732464
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<|endoftext|>",
4
+ "eos_token": "<|endoftext|>",
5
+ "extra_special_tokens": [
6
+ "<|im_start|>",
7
+ "<|im_end|>"
8
+ ],
9
+ "is_local": false,
10
+ "model_max_length": 1000000000000000019884624838656,
11
+ "pad_token": "<|pad|>",
12
+ "tokenizer_class": "TokenizersBackend"
13
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc3223a9d0c4357d0ca82ae03ede3de1f8ac8a2833cb38a3f794c4f4ccb4297b
3
+ size 5201