CodeIsAbstract commited on
Commit
71b227a
·
verified ·
1 Parent(s): 67370d1

Training in progress, step 1000, checkpoint

Browse files
last-checkpoint/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "HybridTimeScaleLM"
4
+ ],
5
+ "bos_token_id": 1,
6
+ "chunk_size": 128,
7
+ "dropout": 0.05,
8
+ "dtype": "float32",
9
+ "eos_token_id": 2,
10
+ "latent_dim": 768,
11
+ "layer_types": [
12
+ "linear",
13
+ "linear",
14
+ "linear",
15
+ "softmax",
16
+ "linear",
17
+ "linear",
18
+ "linear",
19
+ "softmax",
20
+ "linear",
21
+ "linear",
22
+ "linear",
23
+ "softmax"
24
+ ],
25
+ "model_type": "hybrid_fourier_lm",
26
+ "num_layers": 12,
27
+ "num_modes": 64,
28
+ "pad_token_id": 2,
29
+ "tie_word_embeddings": true,
30
+ "time_scale": 128.0,
31
+ "transformers_version": "5.14.1",
32
+ "use_cache": false,
33
+ "vocab_size": 32768
34
+ }
last-checkpoint/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": 2,
8
+ "transformers_version": "5.14.1"
9
+ }
last-checkpoint/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62dc08c3b04819417d33e643fa282f64d9da56cda9541aa632f613efde78508e
3
+ size 570000656
last-checkpoint/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b003b26b6d552a0254dfd899e2d049539bbb1845d678ea4a3ed7d216fc3e1639
3
+ size 1622091
last-checkpoint/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9005df1db3604719ef2b1f952bcbe6e5795d91eb0bc40df5034fd8db2705365e
3
+ size 14469
last-checkpoint/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d608683ffb8e84cf2d336157b612a438455b21c1600a21c17c850a2632c0fedc
3
+ size 14469
last-checkpoint/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8042083e4cce93a1000bc15c672459f6a53221e9e39b96887f2517bac83f10ef
3
+ size 14469
last-checkpoint/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e58d6d1599b3abb1a8cdce2f14fd9c1c4addaa2af7376feba879df9dae37078b
3
+ size 14469
last-checkpoint/rng_state_4.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:407b22baf13dac6bcd1015b1df28b2dd3a3c97fbede2f76998def32b5e905567
3
+ size 14469
last-checkpoint/rng_state_5.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:071373895eb00df0c26cc9e3d11eb09f6b8857b9b793650d36134b166517b58a
3
+ size 14469
last-checkpoint/rng_state_6.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35d119cff383a6da3658fd22867e623b9344c15dbe5f28858bcd429d9f0c2ca3
3
+ size 14469
last-checkpoint/rng_state_7.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba1de290ddd3fc1c66aa674896665299ba7c226ec725195539df4670b870604d
3
+ size 14469
last-checkpoint/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d6700ad120881919d7f723021a36d18ff10cbbe0ccfe22139444743a9c8d3b3
3
+ size 1465
last-checkpoint/trainer_state.json ADDED
@@ -0,0 +1,814 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.09090909090909091,
6
+ "eval_steps": 100,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": false,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0009090909090909091,
14
+ "grad_norm": 10715246100480.0,
15
+ "learning_rate": 9.999983482695944e-06,
16
+ "loss": 31.467498779296875,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.0018181818181818182,
21
+ "grad_norm": 3264906526720.0,
22
+ "learning_rate": 9.999926385982521e-06,
23
+ "loss": 30.928512573242188,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.0027272727272727275,
28
+ "grad_norm": 3908894195712.0,
29
+ "learning_rate": 9.999828506408004e-06,
30
+ "loss": 31.09224853515625,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.0036363636363636364,
35
+ "grad_norm": 2899376340992.0,
36
+ "learning_rate": 9.999689844770768e-06,
37
+ "loss": 31.473455810546874,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.004545454545454545,
42
+ "grad_norm": 3217774870528.0,
43
+ "learning_rate": 9.999510402201835e-06,
44
+ "loss": 31.383303833007812,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.005454545454545455,
49
+ "grad_norm": 10107091943424.0,
50
+ "learning_rate": 9.99929018016486e-06,
51
+ "loss": 31.083233642578126,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.006363636363636364,
56
+ "grad_norm": 15255724883968.0,
57
+ "learning_rate": 9.99902918045613e-06,
58
+ "loss": 31.615689086914063,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.007272727272727273,
63
+ "grad_norm": 6402773876736.0,
64
+ "learning_rate": 9.998727405204536e-06,
65
+ "loss": 31.038418579101563,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.008181818181818182,
70
+ "grad_norm": 1864420425728.0,
71
+ "learning_rate": 9.998384856871566e-06,
72
+ "loss": 31.079879760742188,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.00909090909090909,
77
+ "grad_norm": 2461987504128.0,
78
+ "learning_rate": 9.998001538251283e-06,
79
+ "loss": 31.453466796875,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.00909090909090909,
84
+ "eval_loss": 30.88099479675293,
85
+ "eval_runtime": 22.3501,
86
+ "eval_samples_per_second": 0.179,
87
+ "eval_steps_per_second": 0.045,
88
+ "step": 100
89
+ },
90
+ {
91
+ "epoch": 0.01,
92
+ "grad_norm": 9235392888832.0,
93
+ "learning_rate": 9.9975774524703e-06,
94
+ "loss": 31.271917724609374,
95
+ "step": 110
96
+ },
97
+ {
98
+ "epoch": 0.01090909090909091,
99
+ "grad_norm": 4316258631680.0,
100
+ "learning_rate": 9.997112602987752e-06,
101
+ "loss": 31.086871337890624,
102
+ "step": 120
103
+ },
104
+ {
105
+ "epoch": 0.011818181818181818,
106
+ "grad_norm": 18270237229056.0,
107
+ "learning_rate": 9.996606993595271e-06,
108
+ "loss": 30.72734375,
109
+ "step": 130
110
+ },
111
+ {
112
+ "epoch": 0.012727272727272728,
113
+ "grad_norm": 12576711770112.0,
114
+ "learning_rate": 9.996060628416964e-06,
115
+ "loss": 31.115863037109374,
116
+ "step": 140
117
+ },
118
+ {
119
+ "epoch": 0.013636363636363636,
120
+ "grad_norm": 10765503299584.0,
121
+ "learning_rate": 9.995473511909356e-06,
122
+ "loss": 31.426641845703124,
123
+ "step": 150
124
+ },
125
+ {
126
+ "epoch": 0.014545454545454545,
127
+ "grad_norm": 9155954868224.0,
128
+ "learning_rate": 9.994845648861383e-06,
129
+ "loss": 30.834494018554686,
130
+ "step": 160
131
+ },
132
+ {
133
+ "epoch": 0.015454545454545455,
134
+ "grad_norm": 22049049280512.0,
135
+ "learning_rate": 9.994177044394324e-06,
136
+ "loss": 30.765631103515624,
137
+ "step": 170
138
+ },
139
+ {
140
+ "epoch": 0.016363636363636365,
141
+ "grad_norm": 5104832872448.0,
142
+ "learning_rate": 9.993467703961786e-06,
143
+ "loss": 31.10051574707031,
144
+ "step": 180
145
+ },
146
+ {
147
+ "epoch": 0.017272727272727273,
148
+ "grad_norm": 17401550733312.0,
149
+ "learning_rate": 9.992717633349635e-06,
150
+ "loss": 31.037506103515625,
151
+ "step": 190
152
+ },
153
+ {
154
+ "epoch": 0.01818181818181818,
155
+ "grad_norm": 13520005496832.0,
156
+ "learning_rate": 9.991926838675972e-06,
157
+ "loss": 30.930630493164063,
158
+ "step": 200
159
+ },
160
+ {
161
+ "epoch": 0.01818181818181818,
162
+ "eval_loss": 30.778047561645508,
163
+ "eval_runtime": 2.2346,
164
+ "eval_samples_per_second": 1.79,
165
+ "eval_steps_per_second": 0.448,
166
+ "step": 200
167
+ },
168
+ {
169
+ "epoch": 0.019090909090909092,
170
+ "grad_norm": 19465198632960.0,
171
+ "learning_rate": 9.991095326391061e-06,
172
+ "loss": 31.265313720703126,
173
+ "step": 210
174
+ },
175
+ {
176
+ "epoch": 0.02,
177
+ "grad_norm": 11458692775936.0,
178
+ "learning_rate": 9.990223103277292e-06,
179
+ "loss": 31.001461791992188,
180
+ "step": 220
181
+ },
182
+ {
183
+ "epoch": 0.02090909090909091,
184
+ "grad_norm": 5719059857408.0,
185
+ "learning_rate": 9.989310176449123e-06,
186
+ "loss": 31.132208251953124,
187
+ "step": 230
188
+ },
189
+ {
190
+ "epoch": 0.02181818181818182,
191
+ "grad_norm": 26260075970560.0,
192
+ "learning_rate": 9.988356553353018e-06,
193
+ "loss": 30.999411010742186,
194
+ "step": 240
195
+ },
196
+ {
197
+ "epoch": 0.022727272727272728,
198
+ "grad_norm": 8768481918976.0,
199
+ "learning_rate": 9.987362241767384e-06,
200
+ "loss": 30.9473388671875,
201
+ "step": 250
202
+ },
203
+ {
204
+ "epoch": 0.023636363636363636,
205
+ "grad_norm": 21412534288384.0,
206
+ "learning_rate": 9.986327249802517e-06,
207
+ "loss": 30.593316650390626,
208
+ "step": 260
209
+ },
210
+ {
211
+ "epoch": 0.024545454545454544,
212
+ "grad_norm": 20514749808640.0,
213
+ "learning_rate": 9.985251585900526e-06,
214
+ "loss": 30.888565063476562,
215
+ "step": 270
216
+ },
217
+ {
218
+ "epoch": 0.025454545454545455,
219
+ "grad_norm": 11260780347392.0,
220
+ "learning_rate": 9.984135258835273e-06,
221
+ "loss": 31.34383544921875,
222
+ "step": 280
223
+ },
224
+ {
225
+ "epoch": 0.026363636363636363,
226
+ "grad_norm": 28134447513600.0,
227
+ "learning_rate": 9.982978277712294e-06,
228
+ "loss": 31.72291259765625,
229
+ "step": 290
230
+ },
231
+ {
232
+ "epoch": 0.02727272727272727,
233
+ "grad_norm": 16376609636352.0,
234
+ "learning_rate": 9.981780651968724e-06,
235
+ "loss": 31.0668212890625,
236
+ "step": 300
237
+ },
238
+ {
239
+ "epoch": 0.02727272727272727,
240
+ "eval_loss": 30.80950355529785,
241
+ "eval_runtime": 2.2475,
242
+ "eval_samples_per_second": 1.78,
243
+ "eval_steps_per_second": 0.445,
244
+ "step": 300
245
+ },
246
+ {
247
+ "epoch": 0.028181818181818183,
248
+ "grad_norm": 26542667202560.0,
249
+ "learning_rate": 9.980542391373234e-06,
250
+ "loss": 31.477923583984374,
251
+ "step": 310
252
+ },
253
+ {
254
+ "epoch": 0.02909090909090909,
255
+ "grad_norm": 26602693984256.0,
256
+ "learning_rate": 9.97926350602593e-06,
257
+ "loss": 30.95055236816406,
258
+ "step": 320
259
+ },
260
+ {
261
+ "epoch": 0.03,
262
+ "grad_norm": 37361534631936.0,
263
+ "learning_rate": 9.977944006358287e-06,
264
+ "loss": 30.81255187988281,
265
+ "step": 330
266
+ },
267
+ {
268
+ "epoch": 0.03090909090909091,
269
+ "grad_norm": 11495527153664.0,
270
+ "learning_rate": 9.97658390313306e-06,
271
+ "loss": 30.550958251953126,
272
+ "step": 340
273
+ },
274
+ {
275
+ "epoch": 0.031818181818181815,
276
+ "grad_norm": 21247079481344.0,
277
+ "learning_rate": 9.97518320744419e-06,
278
+ "loss": 29.98631896972656,
279
+ "step": 350
280
+ },
281
+ {
282
+ "epoch": 0.03272727272727273,
283
+ "grad_norm": 4213213233152.0,
284
+ "learning_rate": 9.973741930716722e-06,
285
+ "loss": 30.632174682617187,
286
+ "step": 360
287
+ },
288
+ {
289
+ "epoch": 0.03363636363636364,
290
+ "grad_norm": 23871176572928.0,
291
+ "learning_rate": 9.972260084706707e-06,
292
+ "loss": 30.851519775390624,
293
+ "step": 370
294
+ },
295
+ {
296
+ "epoch": 0.034545454545454546,
297
+ "grad_norm": 5361785896960.0,
298
+ "learning_rate": 9.970737681501105e-06,
299
+ "loss": 30.885147094726562,
300
+ "step": 380
301
+ },
302
+ {
303
+ "epoch": 0.035454545454545454,
304
+ "grad_norm": 26303981944832.0,
305
+ "learning_rate": 9.969174733517693e-06,
306
+ "loss": 30.719467163085938,
307
+ "step": 390
308
+ },
309
+ {
310
+ "epoch": 0.03636363636363636,
311
+ "grad_norm": 25164339216384.0,
312
+ "learning_rate": 9.967571253504957e-06,
313
+ "loss": 30.72432861328125,
314
+ "step": 400
315
+ },
316
+ {
317
+ "epoch": 0.03636363636363636,
318
+ "eval_loss": 30.80552864074707,
319
+ "eval_runtime": 2.2454,
320
+ "eval_samples_per_second": 1.781,
321
+ "eval_steps_per_second": 0.445,
322
+ "step": 400
323
+ },
324
+ {
325
+ "epoch": 0.03727272727272727,
326
+ "grad_norm": 10448714858496.0,
327
+ "learning_rate": 9.965927254541987e-06,
328
+ "loss": 30.711407470703126,
329
+ "step": 410
330
+ },
331
+ {
332
+ "epoch": 0.038181818181818185,
333
+ "grad_norm": 9058416328704.0,
334
+ "learning_rate": 9.96424275003838e-06,
335
+ "loss": 31.141534423828126,
336
+ "step": 420
337
+ },
338
+ {
339
+ "epoch": 0.03909090909090909,
340
+ "grad_norm": 4919203987456.0,
341
+ "learning_rate": 9.96251775373412e-06,
342
+ "loss": 31.216400146484375,
343
+ "step": 430
344
+ },
345
+ {
346
+ "epoch": 0.04,
347
+ "grad_norm": 38016089325568.0,
348
+ "learning_rate": 9.96075227969947e-06,
349
+ "loss": 31.38250732421875,
350
+ "step": 440
351
+ },
352
+ {
353
+ "epoch": 0.04090909090909091,
354
+ "grad_norm": 5236365721600.0,
355
+ "learning_rate": 9.958946342334859e-06,
356
+ "loss": 31.027285766601562,
357
+ "step": 450
358
+ },
359
+ {
360
+ "epoch": 0.04181818181818182,
361
+ "grad_norm": 35290588643328.0,
362
+ "learning_rate": 9.957099956370761e-06,
363
+ "loss": 30.80054626464844,
364
+ "step": 460
365
+ },
366
+ {
367
+ "epoch": 0.042727272727272725,
368
+ "grad_norm": 20989702307840.0,
369
+ "learning_rate": 9.955213136867581e-06,
370
+ "loss": 31.026223754882814,
371
+ "step": 470
372
+ },
373
+ {
374
+ "epoch": 0.04363636363636364,
375
+ "grad_norm": 12943608512512.0,
376
+ "learning_rate": 9.953285899215526e-06,
377
+ "loss": 30.8896240234375,
378
+ "step": 480
379
+ },
380
+ {
381
+ "epoch": 0.04454545454545455,
382
+ "grad_norm": 13833738387456.0,
383
+ "learning_rate": 9.951318259134475e-06,
384
+ "loss": 31.069573974609376,
385
+ "step": 490
386
+ },
387
+ {
388
+ "epoch": 0.045454545454545456,
389
+ "grad_norm": 6541920960512.0,
390
+ "learning_rate": 9.949310232673868e-06,
391
+ "loss": 31.00003662109375,
392
+ "step": 500
393
+ },
394
+ {
395
+ "epoch": 0.045454545454545456,
396
+ "eval_loss": 30.771488189697266,
397
+ "eval_runtime": 2.2598,
398
+ "eval_samples_per_second": 1.77,
399
+ "eval_steps_per_second": 0.443,
400
+ "step": 500
401
+ },
402
+ {
403
+ "epoch": 0.046363636363636364,
404
+ "grad_norm": 11619662823424.0,
405
+ "learning_rate": 9.947261836212556e-06,
406
+ "loss": 30.898284912109375,
407
+ "step": 510
408
+ },
409
+ {
410
+ "epoch": 0.04727272727272727,
411
+ "grad_norm": 5862945456128.0,
412
+ "learning_rate": 9.945173086458681e-06,
413
+ "loss": 31.47274169921875,
414
+ "step": 520
415
+ },
416
+ {
417
+ "epoch": 0.04818181818181818,
418
+ "grad_norm": 14174934532096.0,
419
+ "learning_rate": 9.94304400044953e-06,
420
+ "loss": 30.892333984375,
421
+ "step": 530
422
+ },
423
+ {
424
+ "epoch": 0.04909090909090909,
425
+ "grad_norm": 15664712515584.0,
426
+ "learning_rate": 9.940874595551403e-06,
427
+ "loss": 31.075680541992188,
428
+ "step": 540
429
+ },
430
+ {
431
+ "epoch": 0.05,
432
+ "grad_norm": 21449087647744.0,
433
+ "learning_rate": 9.938664889459469e-06,
434
+ "loss": 30.584109497070312,
435
+ "step": 550
436
+ },
437
+ {
438
+ "epoch": 0.05090909090909091,
439
+ "grad_norm": 19566610612224.0,
440
+ "learning_rate": 9.93641490019762e-06,
441
+ "loss": 31.092770385742188,
442
+ "step": 560
443
+ },
444
+ {
445
+ "epoch": 0.05181818181818182,
446
+ "grad_norm": 5505416167424.0,
447
+ "learning_rate": 9.934124646118326e-06,
448
+ "loss": 31.263589477539064,
449
+ "step": 570
450
+ },
451
+ {
452
+ "epoch": 0.05272727272727273,
453
+ "grad_norm": 6437640077312.0,
454
+ "learning_rate": 9.931794145902486e-06,
455
+ "loss": 30.765380859375,
456
+ "step": 580
457
+ },
458
+ {
459
+ "epoch": 0.053636363636363635,
460
+ "grad_norm": 50019268820992.0,
461
+ "learning_rate": 9.929423418559272e-06,
462
+ "loss": 30.655712890625,
463
+ "step": 590
464
+ },
465
+ {
466
+ "epoch": 0.05454545454545454,
467
+ "grad_norm": 4587999723520.0,
468
+ "learning_rate": 9.927012483425977e-06,
469
+ "loss": 31.214239501953124,
470
+ "step": 600
471
+ },
472
+ {
473
+ "epoch": 0.05454545454545454,
474
+ "eval_loss": 30.60542106628418,
475
+ "eval_runtime": 2.2299,
476
+ "eval_samples_per_second": 1.794,
477
+ "eval_steps_per_second": 0.448,
478
+ "step": 600
479
+ },
480
+ {
481
+ "epoch": 0.05545454545454546,
482
+ "grad_norm": 11168276021248.0,
483
+ "learning_rate": 9.924561360167858e-06,
484
+ "loss": 30.37421875,
485
+ "step": 610
486
+ },
487
+ {
488
+ "epoch": 0.056363636363636366,
489
+ "grad_norm": 9794649849856.0,
490
+ "learning_rate": 9.922070068777973e-06,
491
+ "loss": 30.828488159179688,
492
+ "step": 620
493
+ },
494
+ {
495
+ "epoch": 0.057272727272727274,
496
+ "grad_norm": 8697803702272.0,
497
+ "learning_rate": 9.91953862957702e-06,
498
+ "loss": 30.759384155273438,
499
+ "step": 630
500
+ },
501
+ {
502
+ "epoch": 0.05818181818181818,
503
+ "grad_norm": 16887538778112.0,
504
+ "learning_rate": 9.916967063213169e-06,
505
+ "loss": 31.13458251953125,
506
+ "step": 640
507
+ },
508
+ {
509
+ "epoch": 0.05909090909090909,
510
+ "grad_norm": 20750322892800.0,
511
+ "learning_rate": 9.914355390661897e-06,
512
+ "loss": 31.220294189453124,
513
+ "step": 650
514
+ },
515
+ {
516
+ "epoch": 0.06,
517
+ "grad_norm": 14343894728704.0,
518
+ "learning_rate": 9.91170363322581e-06,
519
+ "loss": 30.375115966796876,
520
+ "step": 660
521
+ },
522
+ {
523
+ "epoch": 0.060909090909090906,
524
+ "grad_norm": 7196735700992.0,
525
+ "learning_rate": 9.90901181253448e-06,
526
+ "loss": 30.902432250976563,
527
+ "step": 670
528
+ },
529
+ {
530
+ "epoch": 0.06181818181818182,
531
+ "grad_norm": 8269929644032.0,
532
+ "learning_rate": 9.906279950544259e-06,
533
+ "loss": 30.382269287109374,
534
+ "step": 680
535
+ },
536
+ {
537
+ "epoch": 0.06272727272727273,
538
+ "grad_norm": 8951715332096.0,
539
+ "learning_rate": 9.903508069538108e-06,
540
+ "loss": 30.57989501953125,
541
+ "step": 690
542
+ },
543
+ {
544
+ "epoch": 0.06363636363636363,
545
+ "grad_norm": 16277289566208.0,
546
+ "learning_rate": 9.900696192125403e-06,
547
+ "loss": 30.9401123046875,
548
+ "step": 700
549
+ },
550
+ {
551
+ "epoch": 0.06363636363636363,
552
+ "eval_loss": 30.497390747070312,
553
+ "eval_runtime": 2.2282,
554
+ "eval_samples_per_second": 1.795,
555
+ "eval_steps_per_second": 0.449,
556
+ "step": 700
557
+ },
558
+ {
559
+ "epoch": 0.06454545454545454,
560
+ "grad_norm": 6656336855040.0,
561
+ "learning_rate": 9.897844341241765e-06,
562
+ "loss": 30.22789611816406,
563
+ "step": 710
564
+ },
565
+ {
566
+ "epoch": 0.06545454545454546,
567
+ "grad_norm": 11471946776576.0,
568
+ "learning_rate": 9.894952540148863e-06,
569
+ "loss": 30.879425048828125,
570
+ "step": 720
571
+ },
572
+ {
573
+ "epoch": 0.06636363636363636,
574
+ "grad_norm": 13703840792576.0,
575
+ "learning_rate": 9.892020812434229e-06,
576
+ "loss": 30.433935546875,
577
+ "step": 730
578
+ },
579
+ {
580
+ "epoch": 0.06727272727272728,
581
+ "grad_norm": 7095926128640.0,
582
+ "learning_rate": 9.889049182011065e-06,
583
+ "loss": 30.42056884765625,
584
+ "step": 740
585
+ },
586
+ {
587
+ "epoch": 0.06818181818181818,
588
+ "grad_norm": 18210455814144.0,
589
+ "learning_rate": 9.886037673118048e-06,
590
+ "loss": 30.620001220703124,
591
+ "step": 750
592
+ },
593
+ {
594
+ "epoch": 0.06909090909090909,
595
+ "grad_norm": 6186389209088.0,
596
+ "learning_rate": 9.882986310319124e-06,
597
+ "loss": 30.185427856445312,
598
+ "step": 760
599
+ },
600
+ {
601
+ "epoch": 0.07,
602
+ "grad_norm": 7534207827968.0,
603
+ "learning_rate": 9.879895118503324e-06,
604
+ "loss": 30.567578125,
605
+ "step": 770
606
+ },
607
+ {
608
+ "epoch": 0.07090909090909091,
609
+ "grad_norm": 9775391703040.0,
610
+ "learning_rate": 9.876764122884548e-06,
611
+ "loss": 30.495364379882812,
612
+ "step": 780
613
+ },
614
+ {
615
+ "epoch": 0.07181818181818182,
616
+ "grad_norm": 23249035460608.0,
617
+ "learning_rate": 9.873593349001365e-06,
618
+ "loss": 30.3556640625,
619
+ "step": 790
620
+ },
621
+ {
622
+ "epoch": 0.07272727272727272,
623
+ "grad_norm": 9671637204992.0,
624
+ "learning_rate": 9.870382822716798e-06,
625
+ "loss": 30.30170593261719,
626
+ "step": 800
627
+ },
628
+ {
629
+ "epoch": 0.07272727272727272,
630
+ "eval_loss": 30.311595916748047,
631
+ "eval_runtime": 2.2334,
632
+ "eval_samples_per_second": 1.791,
633
+ "eval_steps_per_second": 0.448,
634
+ "step": 800
635
+ },
636
+ {
637
+ "epoch": 0.07363636363636364,
638
+ "grad_norm": 12116743421952.0,
639
+ "learning_rate": 9.867132570218128e-06,
640
+ "loss": 30.769448852539064,
641
+ "step": 810
642
+ },
643
+ {
644
+ "epoch": 0.07454545454545454,
645
+ "grad_norm": 15943338033152.0,
646
+ "learning_rate": 9.863842618016658e-06,
647
+ "loss": 30.824539184570312,
648
+ "step": 820
649
+ },
650
+ {
651
+ "epoch": 0.07545454545454545,
652
+ "grad_norm": 12775806992384.0,
653
+ "learning_rate": 9.860512992947521e-06,
654
+ "loss": 30.45408935546875,
655
+ "step": 830
656
+ },
657
+ {
658
+ "epoch": 0.07636363636363637,
659
+ "grad_norm": 6045823926272.0,
660
+ "learning_rate": 9.857143722169442e-06,
661
+ "loss": 29.990557861328124,
662
+ "step": 840
663
+ },
664
+ {
665
+ "epoch": 0.07727272727272727,
666
+ "grad_norm": 5246831034368.0,
667
+ "learning_rate": 9.853734833164526e-06,
668
+ "loss": 30.83751220703125,
669
+ "step": 850
670
+ },
671
+ {
672
+ "epoch": 0.07818181818181819,
673
+ "grad_norm": 4444648374272.0,
674
+ "learning_rate": 9.850286353738033e-06,
675
+ "loss": 30.621414184570312,
676
+ "step": 860
677
+ },
678
+ {
679
+ "epoch": 0.07909090909090909,
680
+ "grad_norm": 3663882878976.0,
681
+ "learning_rate": 9.846798312018147e-06,
682
+ "loss": 30.290618896484375,
683
+ "step": 870
684
+ },
685
+ {
686
+ "epoch": 0.08,
687
+ "grad_norm": 10184265039872.0,
688
+ "learning_rate": 9.843270736455753e-06,
689
+ "loss": 30.944476318359374,
690
+ "step": 880
691
+ },
692
+ {
693
+ "epoch": 0.0809090909090909,
694
+ "grad_norm": 6461672914944.0,
695
+ "learning_rate": 9.839703655824197e-06,
696
+ "loss": 30.379254150390626,
697
+ "step": 890
698
+ },
699
+ {
700
+ "epoch": 0.08181818181818182,
701
+ "grad_norm": 8673656569856.0,
702
+ "learning_rate": 9.836097099219058e-06,
703
+ "loss": 30.13455810546875,
704
+ "step": 900
705
+ },
706
+ {
707
+ "epoch": 0.08181818181818182,
708
+ "eval_loss": 30.246681213378906,
709
+ "eval_runtime": 2.2478,
710
+ "eval_samples_per_second": 1.779,
711
+ "eval_steps_per_second": 0.445,
712
+ "step": 900
713
+ },
714
+ {
715
+ "epoch": 0.08272727272727273,
716
+ "grad_norm": 6762599022592.0,
717
+ "learning_rate": 9.832451096057912e-06,
718
+ "loss": 30.483001708984375,
719
+ "step": 910
720
+ },
721
+ {
722
+ "epoch": 0.08363636363636363,
723
+ "grad_norm": 6819155542016.0,
724
+ "learning_rate": 9.828765676080081e-06,
725
+ "loss": 30.585333251953124,
726
+ "step": 920
727
+ },
728
+ {
729
+ "epoch": 0.08454545454545455,
730
+ "grad_norm": 3836970532864.0,
731
+ "learning_rate": 9.825040869346407e-06,
732
+ "loss": 30.220254516601564,
733
+ "step": 930
734
+ },
735
+ {
736
+ "epoch": 0.08545454545454545,
737
+ "grad_norm": 16401004756992.0,
738
+ "learning_rate": 9.821276706238986e-06,
739
+ "loss": 30.71534423828125,
740
+ "step": 940
741
+ },
742
+ {
743
+ "epoch": 0.08636363636363636,
744
+ "grad_norm": 3000969199616.0,
745
+ "learning_rate": 9.817473217460942e-06,
746
+ "loss": 30.396319580078124,
747
+ "step": 950
748
+ },
749
+ {
750
+ "epoch": 0.08727272727272728,
751
+ "grad_norm": 13839672279040.0,
752
+ "learning_rate": 9.813630434036161e-06,
753
+ "loss": 30.110427856445312,
754
+ "step": 960
755
+ },
756
+ {
757
+ "epoch": 0.08818181818181818,
758
+ "grad_norm": 6212048388096.0,
759
+ "learning_rate": 9.809748387309049e-06,
760
+ "loss": 30.376641845703126,
761
+ "step": 970
762
+ },
763
+ {
764
+ "epoch": 0.0890909090909091,
765
+ "grad_norm": 8919135027200.0,
766
+ "learning_rate": 9.80582710894426e-06,
767
+ "loss": 30.46407470703125,
768
+ "step": 980
769
+ },
770
+ {
771
+ "epoch": 0.09,
772
+ "grad_norm": 3707678490624.0,
773
+ "learning_rate": 9.80186663092646e-06,
774
+ "loss": 30.339068603515624,
775
+ "step": 990
776
+ },
777
+ {
778
+ "epoch": 0.09090909090909091,
779
+ "grad_norm": 7284239892480.0,
780
+ "learning_rate": 9.79786698556005e-06,
781
+ "loss": 30.35057373046875,
782
+ "step": 1000
783
+ },
784
+ {
785
+ "epoch": 0.09090909090909091,
786
+ "eval_loss": 30.29996109008789,
787
+ "eval_runtime": 2.2418,
788
+ "eval_samples_per_second": 1.784,
789
+ "eval_steps_per_second": 0.446,
790
+ "step": 1000
791
+ }
792
+ ],
793
+ "logging_steps": 10,
794
+ "max_steps": 11000,
795
+ "num_input_tokens_seen": 0,
796
+ "num_train_epochs": 9223372036854775807,
797
+ "save_steps": 1000,
798
+ "stateful_callbacks": {
799
+ "TrainerControl": {
800
+ "args": {
801
+ "should_epoch_stop": false,
802
+ "should_evaluate": false,
803
+ "should_log": false,
804
+ "should_save": true,
805
+ "should_training_stop": false
806
+ },
807
+ "attributes": {}
808
+ }
809
+ },
810
+ "total_flos": 4529923817472000.0,
811
+ "train_batch_size": 1,
812
+ "trial_name": null,
813
+ "trial_params": null
814
+ }
last-checkpoint/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b7b7630962008facfc553dae23e1b281e5855fa53113fd26848cdb4e9d6e896
3
+ size 5265