Danilosistu commited on
Commit
e7f7d99
·
verified ·
1 Parent(s): 7051413

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "OpenMythosForCausalLM"
4
+ ],
5
+ "bias": false,
6
+ "block_size": 512,
7
+ "bos_token_id": 1,
8
+ "dtype": "float32",
9
+ "effective_expected_depth": 10,
10
+ "eos_token_id": 2,
11
+ "head_dim": 64,
12
+ "init_values": {
13
+ "embed_scale": 16.0,
14
+ "embedding": 0.03952847075210474,
15
+ "out_proj": 0.008838834764831844,
16
+ "std": 0.03952847075210474
17
+ },
18
+ "injection_type": "lti",
19
+ "intermediate_size": 1024,
20
+ "kv_lora_rank": 512,
21
+ "mean_backprop_depth": 2,
22
+ "mean_recurrence": 4,
23
+ "model_type": "open_mythos",
24
+ "moe_intermediate_size": 1024,
25
+ "moe_top_k": 2,
26
+ "n_embd": 256,
27
+ "n_heads": 4,
28
+ "n_kv_heads": 4,
29
+ "n_layers": 4,
30
+ "n_layers_in_coda": 1,
31
+ "n_layers_in_prelude": 1,
32
+ "n_layers_in_recurrent_block": 2,
33
+ "n_routed_experts": 8,
34
+ "n_shared_experts": 2,
35
+ "norm_eps": 1e-06,
36
+ "num_key_value_heads": 4,
37
+ "pad_token_id": 0,
38
+ "q_lora_rank": 1536,
39
+ "qk_bias": true,
40
+ "rope_base": 50000.0,
41
+ "rope_head_dim": 64,
42
+ "state_init": "like-init",
43
+ "tie_embeddings": true,
44
+ "tie_word_embeddings": true,
45
+ "transformers_version": "5.12.1",
46
+ "use_cache": false,
47
+ "use_mla": false,
48
+ "use_moe": false,
49
+ "vocab_size": 50257
50
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "output_attentions": false,
6
+ "output_hidden_states": false,
7
+ "pad_token_id": 0,
8
+ "transformers_version": "5.12.1"
9
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17fec733bba76bbea9654a739252c82e91a429976b0d8e990c8f1e13c9cc0d9b
3
+ size 68667688
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75df1a6f6eced2ac9d338654b82a1802da341d3750c6e59f0aaeea66bd1255ca
3
+ size 137102027
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2458afedab118e85bc7f6d6d4d0840693061e372c5d0bdb87ff860941c24bf5
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fce4703bf6c409f68165faf4b50b370c99acf3d793e9a8199a3a663742db568
3
+ size 1465
trainer_state.json ADDED
@@ -0,0 +1,741 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 5000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0002,
14
+ "grad_norm": 5.551233768463135,
15
+ "learning_rate": 0.0,
16
+ "loss": 43.93932342529297,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.01,
21
+ "grad_norm": 6.35080099105835,
22
+ "learning_rate": 2.94e-05,
23
+ "loss": 43.64981764190051,
24
+ "step": 50
25
+ },
26
+ {
27
+ "epoch": 0.02,
28
+ "grad_norm": 5.6910929679870605,
29
+ "learning_rate": 5.94e-05,
30
+ "loss": 40.47484375,
31
+ "step": 100
32
+ },
33
+ {
34
+ "epoch": 0.03,
35
+ "grad_norm": 5.369699954986572,
36
+ "learning_rate": 8.939999999999999e-05,
37
+ "loss": 37.00744873046875,
38
+ "step": 150
39
+ },
40
+ {
41
+ "epoch": 0.04,
42
+ "grad_norm": 4.0027079582214355,
43
+ "learning_rate": 0.0001194,
44
+ "loss": 33.13272216796875,
45
+ "step": 200
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "grad_norm": 3.570080041885376,
50
+ "learning_rate": 0.0001494,
51
+ "loss": 30.27530517578125,
52
+ "step": 250
53
+ },
54
+ {
55
+ "epoch": 0.06,
56
+ "grad_norm": 3.417503595352173,
57
+ "learning_rate": 0.00017939999999999997,
58
+ "loss": 29.3436669921875,
59
+ "step": 300
60
+ },
61
+ {
62
+ "epoch": 0.07,
63
+ "grad_norm": 3.3700530529022217,
64
+ "learning_rate": 0.00020939999999999997,
65
+ "loss": 28.82450927734375,
66
+ "step": 350
67
+ },
68
+ {
69
+ "epoch": 0.08,
70
+ "grad_norm": 3.179579496383667,
71
+ "learning_rate": 0.0002394,
72
+ "loss": 28.15291015625,
73
+ "step": 400
74
+ },
75
+ {
76
+ "epoch": 0.09,
77
+ "grad_norm": 3.0590453147888184,
78
+ "learning_rate": 0.0002694,
79
+ "loss": 27.42884521484375,
80
+ "step": 450
81
+ },
82
+ {
83
+ "epoch": 0.1,
84
+ "grad_norm": 2.864997625350952,
85
+ "learning_rate": 0.00029939999999999996,
86
+ "loss": 27.338408203125,
87
+ "step": 500
88
+ },
89
+ {
90
+ "epoch": 0.11,
91
+ "grad_norm": 2.719647169113159,
92
+ "learning_rate": 0.00029991224218749483,
93
+ "loss": 27.45952880859375,
94
+ "step": 550
95
+ },
96
+ {
97
+ "epoch": 0.12,
98
+ "grad_norm": 3.106184720993042,
99
+ "learning_rate": 0.00029964187595418377,
100
+ "loss": 26.478095703125,
101
+ "step": 600
102
+ },
103
+ {
104
+ "epoch": 0.13,
105
+ "grad_norm": 2.9786314964294434,
106
+ "learning_rate": 0.0002991891941455865,
107
+ "loss": 26.3969140625,
108
+ "step": 650
109
+ },
110
+ {
111
+ "epoch": 0.14,
112
+ "grad_norm": 3.4186649322509766,
113
+ "learning_rate": 0.0002985547482847564,
114
+ "loss": 26.35082763671875,
115
+ "step": 700
116
+ },
117
+ {
118
+ "epoch": 0.15,
119
+ "grad_norm": 3.016622304916382,
120
+ "learning_rate": 0.00029773931134624614,
121
+ "loss": 26.2290625,
122
+ "step": 750
123
+ },
124
+ {
125
+ "epoch": 0.16,
126
+ "grad_norm": 2.8012444972991943,
127
+ "learning_rate": 0.0002967438768143568,
128
+ "loss": 25.8839404296875,
129
+ "step": 800
130
+ },
131
+ {
132
+ "epoch": 0.17,
133
+ "grad_norm": 3.0996336936950684,
134
+ "learning_rate": 0.0002955696574727306,
135
+ "loss": 25.84023681640625,
136
+ "step": 850
137
+ },
138
+ {
139
+ "epoch": 0.18,
140
+ "grad_norm": 3.161088705062866,
141
+ "learning_rate": 0.0002942180839267606,
142
+ "loss": 25.78875732421875,
143
+ "step": 900
144
+ },
145
+ {
146
+ "epoch": 0.19,
147
+ "grad_norm": 3.0394628047943115,
148
+ "learning_rate": 0.00029269080286061866,
149
+ "loss": 25.57990966796875,
150
+ "step": 950
151
+ },
152
+ {
153
+ "epoch": 0.2,
154
+ "grad_norm": 3.1691832542419434,
155
+ "learning_rate": 0.0002909896750310243,
156
+ "loss": 25.3314111328125,
157
+ "step": 1000
158
+ },
159
+ {
160
+ "epoch": 0.21,
161
+ "grad_norm": 3.3732755184173584,
162
+ "learning_rate": 0.0002891167730001992,
163
+ "loss": 25.2071630859375,
164
+ "step": 1050
165
+ },
166
+ {
167
+ "epoch": 0.22,
168
+ "grad_norm": 3.613858461380005,
169
+ "learning_rate": 0.00028707437861076943,
170
+ "loss": 25.27132080078125,
171
+ "step": 1100
172
+ },
173
+ {
174
+ "epoch": 0.23,
175
+ "grad_norm": 3.145481824874878,
176
+ "learning_rate": 0.0002848649802056918,
177
+ "loss": 25.17646240234375,
178
+ "step": 1150
179
+ },
180
+ {
181
+ "epoch": 0.24,
182
+ "grad_norm": 3.1994168758392334,
183
+ "learning_rate": 0.00028249126959659126,
184
+ "loss": 24.8016015625,
185
+ "step": 1200
186
+ },
187
+ {
188
+ "epoch": 0.25,
189
+ "grad_norm": 2.9414961338043213,
190
+ "learning_rate": 0.00027995613878420276,
191
+ "loss": 24.93425537109375,
192
+ "step": 1250
193
+ },
194
+ {
195
+ "epoch": 0.26,
196
+ "grad_norm": 3.3210513591766357,
197
+ "learning_rate": 0.00027726267643491434,
198
+ "loss": 24.86907958984375,
199
+ "step": 1300
200
+ },
201
+ {
202
+ "epoch": 0.27,
203
+ "grad_norm": 3.235185384750366,
204
+ "learning_rate": 0.00027441416411770245,
205
+ "loss": 24.840205078125,
206
+ "step": 1350
207
+ },
208
+ {
209
+ "epoch": 0.28,
210
+ "grad_norm": 3.130532741546631,
211
+ "learning_rate": 0.00027141407230604596,
212
+ "loss": 24.85203857421875,
213
+ "step": 1400
214
+ },
215
+ {
216
+ "epoch": 0.29,
217
+ "grad_norm": 3.515564203262329,
218
+ "learning_rate": 0.0002682660561496887,
219
+ "loss": 24.42603515625,
220
+ "step": 1450
221
+ },
222
+ {
223
+ "epoch": 0.3,
224
+ "grad_norm": 3.041907787322998,
225
+ "learning_rate": 0.00026497395102140247,
226
+ "loss": 24.52408203125,
227
+ "step": 1500
228
+ },
229
+ {
230
+ "epoch": 0.31,
231
+ "grad_norm": 3.3526618480682373,
232
+ "learning_rate": 0.0002615417678441761,
233
+ "loss": 24.54887939453125,
234
+ "step": 1550
235
+ },
236
+ {
237
+ "epoch": 0.32,
238
+ "grad_norm": 3.3341870307922363,
239
+ "learning_rate": 0.0002579736882045238,
240
+ "loss": 24.48134521484375,
241
+ "step": 1600
242
+ },
243
+ {
244
+ "epoch": 0.33,
245
+ "grad_norm": 3.107917070388794,
246
+ "learning_rate": 0.00025427405925786595,
247
+ "loss": 24.417197265625,
248
+ "step": 1650
249
+ },
250
+ {
251
+ "epoch": 0.34,
252
+ "grad_norm": 3.5469655990600586,
253
+ "learning_rate": 0.0002504473884321899,
254
+ "loss": 24.3250341796875,
255
+ "step": 1700
256
+ },
257
+ {
258
+ "epoch": 0.35,
259
+ "grad_norm": 3.412889003753662,
260
+ "learning_rate": 0.0002464983379364433,
261
+ "loss": 24.1046923828125,
262
+ "step": 1750
263
+ },
264
+ {
265
+ "epoch": 0.36,
266
+ "grad_norm": 3.5492677688598633,
267
+ "learning_rate": 0.00024243171908035053,
268
+ "loss": 24.25704345703125,
269
+ "step": 1800
270
+ },
271
+ {
272
+ "epoch": 0.37,
273
+ "grad_norm": 3.6415436267852783,
274
+ "learning_rate": 0.00023825248641257327,
275
+ "loss": 23.98925048828125,
276
+ "step": 1850
277
+ },
278
+ {
279
+ "epoch": 0.38,
280
+ "grad_norm": 3.313443899154663,
281
+ "learning_rate": 0.00023396573168435567,
282
+ "loss": 24.17134521484375,
283
+ "step": 1900
284
+ },
285
+ {
286
+ "epoch": 0.39,
287
+ "grad_norm": 3.796315908432007,
288
+ "learning_rate": 0.00022957667764601008,
289
+ "loss": 23.89406005859375,
290
+ "step": 1950
291
+ },
292
+ {
293
+ "epoch": 0.4,
294
+ "grad_norm": 3.2706825733184814,
295
+ "learning_rate": 0.00022509067168380038,
296
+ "loss": 24.1194677734375,
297
+ "step": 2000
298
+ },
299
+ {
300
+ "epoch": 0.41,
301
+ "grad_norm": 3.441154718399048,
302
+ "learning_rate": 0.00022051317930497523,
303
+ "loss": 24.1999072265625,
304
+ "step": 2050
305
+ },
306
+ {
307
+ "epoch": 0.42,
308
+ "grad_norm": 3.518808364868164,
309
+ "learning_rate": 0.0002158497774788896,
310
+ "loss": 23.89898193359375,
311
+ "step": 2100
312
+ },
313
+ {
314
+ "epoch": 0.43,
315
+ "grad_norm": 3.7033307552337646,
316
+ "learning_rate": 0.0002111061478423265,
317
+ "loss": 23.790986328125,
318
+ "step": 2150
319
+ },
320
+ {
321
+ "epoch": 0.44,
322
+ "grad_norm": 3.673150062561035,
323
+ "learning_rate": 0.00020628806977729812,
324
+ "loss": 24.0228662109375,
325
+ "step": 2200
326
+ },
327
+ {
328
+ "epoch": 0.45,
329
+ "grad_norm": 3.573422908782959,
330
+ "learning_rate": 0.0002014014133697585,
331
+ "loss": 23.67351318359375,
332
+ "step": 2250
333
+ },
334
+ {
335
+ "epoch": 0.46,
336
+ "grad_norm": 4.668125629425049,
337
+ "learning_rate": 0.0001964521322578086,
338
+ "loss": 23.75174560546875,
339
+ "step": 2300
340
+ },
341
+ {
342
+ "epoch": 0.47,
343
+ "grad_norm": 4.469000339508057,
344
+ "learning_rate": 0.00019144625637810503,
345
+ "loss": 23.79510986328125,
346
+ "step": 2350
347
+ },
348
+ {
349
+ "epoch": 0.48,
350
+ "grad_norm": 4.140970230102539,
351
+ "learning_rate": 0.0001863898846193111,
352
+ "loss": 23.70107421875,
353
+ "step": 2400
354
+ },
355
+ {
356
+ "epoch": 0.49,
357
+ "grad_norm": 4.114079475402832,
358
+ "learning_rate": 0.00018128917739154056,
359
+ "loss": 23.66251708984375,
360
+ "step": 2450
361
+ },
362
+ {
363
+ "epoch": 0.5,
364
+ "grad_norm": 4.908941268920898,
365
+ "learning_rate": 0.0001761503491208467,
366
+ "loss": 23.6627685546875,
367
+ "step": 2500
368
+ },
369
+ {
370
+ "epoch": 0.51,
371
+ "grad_norm": 3.531893491744995,
372
+ "learning_rate": 0.00017097966067790163,
373
+ "loss": 23.67588623046875,
374
+ "step": 2550
375
+ },
376
+ {
377
+ "epoch": 0.52,
378
+ "grad_norm": 3.9994258880615234,
379
+ "learning_rate": 0.00016578341175008948,
380
+ "loss": 23.6676025390625,
381
+ "step": 2600
382
+ },
383
+ {
384
+ "epoch": 0.53,
385
+ "grad_norm": 4.651762962341309,
386
+ "learning_rate": 0.00016056793316630803,
387
+ "loss": 23.414052734375,
388
+ "step": 2650
389
+ },
390
+ {
391
+ "epoch": 0.54,
392
+ "grad_norm": 3.7902722358703613,
393
+ "learning_rate": 0.00015533957918382874,
394
+ "loss": 23.56423828125,
395
+ "step": 2700
396
+ },
397
+ {
398
+ "epoch": 0.55,
399
+ "grad_norm": 4.4413838386535645,
400
+ "learning_rate": 0.0001501047197466131,
401
+ "loss": 23.55817626953125,
402
+ "step": 2750
403
+ },
404
+ {
405
+ "epoch": 0.56,
406
+ "grad_norm": 3.6125080585479736,
407
+ "learning_rate": 0.0001448697327245171,
408
+ "loss": 23.47991455078125,
409
+ "step": 2800
410
+ },
411
+ {
412
+ "epoch": 0.57,
413
+ "grad_norm": 4.151065826416016,
414
+ "learning_rate": 0.0001396409961428391,
415
+ "loss": 23.5330859375,
416
+ "step": 2850
417
+ },
418
+ {
419
+ "epoch": 0.58,
420
+ "grad_norm": 5.125370025634766,
421
+ "learning_rate": 0.0001344248804116789,
422
+ "loss": 23.56696533203125,
423
+ "step": 2900
424
+ },
425
+ {
426
+ "epoch": 0.59,
427
+ "grad_norm": 3.613190174102783,
428
+ "learning_rate": 0.00012922774056457376,
429
+ "loss": 23.46764404296875,
430
+ "step": 2950
431
+ },
432
+ {
433
+ "epoch": 0.6,
434
+ "grad_norm": 4.484012126922607,
435
+ "learning_rate": 0.00012405590851586945,
436
+ "loss": 23.2342529296875,
437
+ "step": 3000
438
+ },
439
+ {
440
+ "epoch": 0.61,
441
+ "grad_norm": 4.2972092628479,
442
+ "learning_rate": 0.00011891568534625763,
443
+ "loss": 23.21994140625,
444
+ "step": 3050
445
+ },
446
+ {
447
+ "epoch": 0.62,
448
+ "grad_norm": 4.617368698120117,
449
+ "learning_rate": 0.00011381333362587972,
450
+ "loss": 23.39466552734375,
451
+ "step": 3100
452
+ },
453
+ {
454
+ "epoch": 0.63,
455
+ "grad_norm": 4.377896785736084,
456
+ "learning_rate": 0.0001087550697843501,
457
+ "loss": 23.0036181640625,
458
+ "step": 3150
459
+ },
460
+ {
461
+ "epoch": 0.64,
462
+ "grad_norm": 3.8239970207214355,
463
+ "learning_rate": 0.00010374705653699379,
464
+ "loss": 23.21521484375,
465
+ "step": 3200
466
+ },
467
+ {
468
+ "epoch": 0.65,
469
+ "grad_norm": 3.8129734992980957,
470
+ "learning_rate": 9.879539537652734e-05,
471
+ "loss": 23.19203125,
472
+ "step": 3250
473
+ },
474
+ {
475
+ "epoch": 0.66,
476
+ "grad_norm": 4.068327903747559,
477
+ "learning_rate": 9.390611913932994e-05,
478
+ "loss": 22.90923583984375,
479
+ "step": 3300
480
+ },
481
+ {
482
+ "epoch": 0.67,
483
+ "grad_norm": 4.683315277099609,
484
+ "learning_rate": 8.90851846553613e-05,
485
+ "loss": 23.061630859375,
486
+ "step": 3350
487
+ },
488
+ {
489
+ "epoch": 0.68,
490
+ "grad_norm": 4.174139976501465,
491
+ "learning_rate": 8.433846549068207e-05,
492
+ "loss": 23.2664208984375,
493
+ "step": 3400
494
+ },
495
+ {
496
+ "epoch": 0.69,
497
+ "grad_norm": 4.326903820037842,
498
+ "learning_rate": 7.967174479141851e-05,
499
+ "loss": 22.98652099609375,
500
+ "step": 3450
501
+ },
502
+ {
503
+ "epoch": 0.7,
504
+ "grad_norm": 4.383793354034424,
505
+ "learning_rate": 7.509070823788932e-05,
506
+ "loss": 23.09397216796875,
507
+ "step": 3500
508
+ },
509
+ {
510
+ "epoch": 0.71,
511
+ "grad_norm": 4.890023231506348,
512
+ "learning_rate": 7.060093711748041e-05,
513
+ "loss": 23.15705078125,
514
+ "step": 3550
515
+ },
516
+ {
517
+ "epoch": 0.72,
518
+ "grad_norm": 4.316667556762695,
519
+ "learning_rate": 6.620790152470581e-05,
520
+ "loss": 23.2762353515625,
521
+ "step": 3600
522
+ },
523
+ {
524
+ "epoch": 0.73,
525
+ "grad_norm": 4.178554534912109,
526
+ "learning_rate": 6.191695369674005e-05,
527
+ "loss": 23.15247314453125,
528
+ "step": 3650
529
+ },
530
+ {
531
+ "epoch": 0.74,
532
+ "grad_norm": 4.168890476226807,
533
+ "learning_rate": 5.773332149254167e-05,
534
+ "loss": 22.8707421875,
535
+ "step": 3700
536
+ },
537
+ {
538
+ "epoch": 0.75,
539
+ "grad_norm": 4.778957843780518,
540
+ "learning_rate": 5.3662102023512335e-05,
541
+ "loss": 23.31576904296875,
542
+ "step": 3750
543
+ },
544
+ {
545
+ "epoch": 0.76,
546
+ "grad_norm": 4.6559576988220215,
547
+ "learning_rate": 4.970825544345178e-05,
548
+ "loss": 22.9227734375,
549
+ "step": 3800
550
+ },
551
+ {
552
+ "epoch": 0.77,
553
+ "grad_norm": 4.201073169708252,
554
+ "learning_rate": 4.5876598905374424e-05,
555
+ "loss": 23.15135498046875,
556
+ "step": 3850
557
+ },
558
+ {
559
+ "epoch": 0.78,
560
+ "grad_norm": 4.3280181884765625,
561
+ "learning_rate": 4.217180069255052e-05,
562
+ "loss": 22.96529052734375,
563
+ "step": 3900
564
+ },
565
+ {
566
+ "epoch": 0.79,
567
+ "grad_norm": 4.0987229347229,
568
+ "learning_rate": 3.859837453092184e-05,
569
+ "loss": 23.18628173828125,
570
+ "step": 3950
571
+ },
572
+ {
573
+ "epoch": 0.8,
574
+ "grad_norm": 4.1396098136901855,
575
+ "learning_rate": 3.5160674089822415e-05,
576
+ "loss": 23.08689453125,
577
+ "step": 4000
578
+ },
579
+ {
580
+ "epoch": 0.81,
581
+ "grad_norm": 4.06009578704834,
582
+ "learning_rate": 3.186288767770243e-05,
583
+ "loss": 23.24049072265625,
584
+ "step": 4050
585
+ },
586
+ {
587
+ "epoch": 0.82,
588
+ "grad_norm": 4.152278423309326,
589
+ "learning_rate": 2.8709033139320016e-05,
590
+ "loss": 23.12670654296875,
591
+ "step": 4100
592
+ },
593
+ {
594
+ "epoch": 0.83,
595
+ "grad_norm": 4.040670394897461,
596
+ "learning_rate": 2.570295296061618e-05,
597
+ "loss": 22.70718505859375,
598
+ "step": 4150
599
+ },
600
+ {
601
+ "epoch": 0.84,
602
+ "grad_norm": 4.150204181671143,
603
+ "learning_rate": 2.2848309587237545e-05,
604
+ "loss": 22.81280517578125,
605
+ "step": 4200
606
+ },
607
+ {
608
+ "epoch": 0.85,
609
+ "grad_norm": 4.331852436065674,
610
+ "learning_rate": 2.014858096241036e-05,
611
+ "loss": 23.07447021484375,
612
+ "step": 4250
613
+ },
614
+ {
615
+ "epoch": 0.86,
616
+ "grad_norm": 4.494900226593018,
617
+ "learning_rate": 1.7607056289602783e-05,
618
+ "loss": 23.06335205078125,
619
+ "step": 4300
620
+ },
621
+ {
622
+ "epoch": 0.87,
623
+ "grad_norm": 3.983609199523926,
624
+ "learning_rate": 1.5226832025136698e-05,
625
+ "loss": 22.91209228515625,
626
+ "step": 4350
627
+ },
628
+ {
629
+ "epoch": 0.88,
630
+ "grad_norm": 4.992405891418457,
631
+ "learning_rate": 1.3010808105632909e-05,
632
+ "loss": 23.0981298828125,
633
+ "step": 4400
634
+ },
635
+ {
636
+ "epoch": 0.89,
637
+ "grad_norm": 3.8591504096984863,
638
+ "learning_rate": 1.0961684414885064e-05,
639
+ "loss": 23.1151025390625,
640
+ "step": 4450
641
+ },
642
+ {
643
+ "epoch": 0.9,
644
+ "grad_norm": 3.8510167598724365,
645
+ "learning_rate": 9.081957494466985e-06,
646
+ "loss": 23.03157470703125,
647
+ "step": 4500
648
+ },
649
+ {
650
+ "epoch": 0.91,
651
+ "grad_norm": 4.121877193450928,
652
+ "learning_rate": 7.3739175020815e-06,
653
+ "loss": 22.97037353515625,
654
+ "step": 4550
655
+ },
656
+ {
657
+ "epoch": 0.92,
658
+ "grad_norm": 4.034373760223389,
659
+ "learning_rate": 5.839645421355926e-06,
660
+ "loss": 22.92754150390625,
661
+ "step": 4600
662
+ },
663
+ {
664
+ "epoch": 0.93,
665
+ "grad_norm": 4.7229461669921875,
666
+ "learning_rate": 4.48101052648413e-06,
667
+ "loss": 23.0431982421875,
668
+ "step": 4650
669
+ },
670
+ {
671
+ "epoch": 0.94,
672
+ "grad_norm": 3.8502602577209473,
673
+ "learning_rate": 3.299668104803843e-06,
674
+ "loss": 22.8787841796875,
675
+ "step": 4700
676
+ },
677
+ {
678
+ "epoch": 0.95,
679
+ "grad_norm": 4.068970203399658,
680
+ "learning_rate": 2.297057440084049e-06,
681
+ "loss": 23.1052392578125,
682
+ "step": 4750
683
+ },
684
+ {
685
+ "epoch": 0.96,
686
+ "grad_norm": 4.614439964294434,
687
+ "learning_rate": 1.4744000589793581e-06,
688
+ "loss": 22.9152294921875,
689
+ "step": 4800
690
+ },
691
+ {
692
+ "epoch": 0.97,
693
+ "grad_norm": 4.608243465423584,
694
+ "learning_rate": 8.326982427880103e-07,
695
+ "loss": 22.91056884765625,
696
+ "step": 4850
697
+ },
698
+ {
699
+ "epoch": 0.98,
700
+ "grad_norm": 4.816669940948486,
701
+ "learning_rate": 3.727338063264418e-07,
702
+ "loss": 22.81651611328125,
703
+ "step": 4900
704
+ },
705
+ {
706
+ "epoch": 0.99,
707
+ "grad_norm": 3.9090330600738525,
708
+ "learning_rate": 9.50671454083729e-08,
709
+ "loss": 22.8877197265625,
710
+ "step": 4950
711
+ },
712
+ {
713
+ "epoch": 1.0,
714
+ "grad_norm": 3.918071985244751,
715
+ "learning_rate": 3.655408888914024e-11,
716
+ "loss": 22.89197265625,
717
+ "step": 5000
718
+ }
719
+ ],
720
+ "logging_steps": 50,
721
+ "max_steps": 5000,
722
+ "num_input_tokens_seen": 0,
723
+ "num_train_epochs": 9223372036854775807,
724
+ "save_steps": 1000,
725
+ "stateful_callbacks": {
726
+ "TrainerControl": {
727
+ "args": {
728
+ "should_epoch_stop": false,
729
+ "should_evaluate": false,
730
+ "should_log": false,
731
+ "should_save": true,
732
+ "should_training_stop": true
733
+ },
734
+ "attributes": {}
735
+ }
736
+ },
737
+ "total_flos": 262180700160000.0,
738
+ "train_batch_size": 1,
739
+ "trial_name": null,
740
+ "trial_params": null
741
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ab8084dab1a441e9076b5583da563e59f901d2007d8e44f895cbc4ca56bc90a
3
+ size 5265