MostLime commited on
Commit
39cd858
·
verified ·
1 Parent(s): 3ab01ae

init upload

Browse files
config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Lfm2ForCausalLM"
4
+ ],
5
+ "block_auto_adjust_ff_dim": true,
6
+ "block_ffn_dim_multiplier": 1.0,
7
+ "block_multiple_of": 256,
8
+ "bos_token_id": 1,
9
+ "conv_L_cache": 3,
10
+ "conv_bias": false,
11
+ "dtype": "float32",
12
+ "eos_token_id": 2,
13
+ "hidden_size": 1024,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 2720,
16
+ "layer_types": [
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention"
29
+ ],
30
+ "max_batch_size": 512,
31
+ "max_position_embeddings": 2048,
32
+ "model_type": "lfm2",
33
+ "norm_eps": 1e-05,
34
+ "num_attention_heads": 16,
35
+ "num_hidden_layers": 12,
36
+ "num_key_value_heads": 4,
37
+ "pad_token_id": 0,
38
+ "rope_theta": 1000000.0,
39
+ "transformers_version": "4.56.0",
40
+ "use_cache": false,
41
+ "vocab_size": 50257
42
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.56.0",
7
+ "use_cache": false
8
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3bb92b5865aa709b54c9d7bb64d60f2c8a690fbf68b0c92744e34df75eeec855
3
+ size 633795192
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62ac04c1e118e5c1bc21ecafff703a9ba1d10c75001a54cb5b8b10a73889ddb7
3
+ size 633943947
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62fae89a87dda02c920f53346ec8f529110400776b4bc34106565d26314c1d04
3
+ size 14645
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13a1c6e38638738c0bc5a722660016155dc0152f6443cf33650b074db30a7091
3
+ size 1465
trainer_state.json ADDED
@@ -0,0 +1,1091 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.3932878867330886,
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": 0.00026219192448872575,
14
+ "grad_norm": 22.20619010925293,
15
+ "learning_rate": 0.0,
16
+ "loss": 10.5131,
17
+ "step": 1
18
+ },
19
+ {
20
+ "epoch": 0.0026219192448872575,
21
+ "grad_norm": 22.429588317871094,
22
+ "learning_rate": 4.4999999999999996e-05,
23
+ "loss": 10.4662,
24
+ "step": 10
25
+ },
26
+ {
27
+ "epoch": 0.005243838489774515,
28
+ "grad_norm": 22.83245086669922,
29
+ "learning_rate": 9.5e-05,
30
+ "loss": 10.1612,
31
+ "step": 20
32
+ },
33
+ {
34
+ "epoch": 0.007865757734661772,
35
+ "grad_norm": 23.247602462768555,
36
+ "learning_rate": 0.000145,
37
+ "loss": 9.5256,
38
+ "step": 30
39
+ },
40
+ {
41
+ "epoch": 0.01048767697954903,
42
+ "grad_norm": 23.51291275024414,
43
+ "learning_rate": 0.00019500000000000002,
44
+ "loss": 8.5708,
45
+ "step": 40
46
+ },
47
+ {
48
+ "epoch": 0.013109596224436287,
49
+ "grad_norm": 22.496492385864258,
50
+ "learning_rate": 0.000245,
51
+ "loss": 7.3388,
52
+ "step": 50
53
+ },
54
+ {
55
+ "epoch": 0.015731515469323543,
56
+ "grad_norm": 16.345460891723633,
57
+ "learning_rate": 0.000295,
58
+ "loss": 5.9703,
59
+ "step": 60
60
+ },
61
+ {
62
+ "epoch": 0.018353434714210803,
63
+ "grad_norm": 3.921259880065918,
64
+ "learning_rate": 0.000345,
65
+ "loss": 4.9478,
66
+ "step": 70
67
+ },
68
+ {
69
+ "epoch": 0.02097535395909806,
70
+ "grad_norm": 7.0385589599609375,
71
+ "learning_rate": 0.000395,
72
+ "loss": 4.6803,
73
+ "step": 80
74
+ },
75
+ {
76
+ "epoch": 0.023597273203985317,
77
+ "grad_norm": 2.6207873821258545,
78
+ "learning_rate": 0.00044500000000000003,
79
+ "loss": 4.4974,
80
+ "step": 90
81
+ },
82
+ {
83
+ "epoch": 0.026219192448872573,
84
+ "grad_norm": 1.9961260557174683,
85
+ "learning_rate": 0.000495,
86
+ "loss": 4.3314,
87
+ "step": 100
88
+ },
89
+ {
90
+ "epoch": 0.028841111693759833,
91
+ "grad_norm": 1.6183704137802124,
92
+ "learning_rate": 0.000545,
93
+ "loss": 4.1959,
94
+ "step": 110
95
+ },
96
+ {
97
+ "epoch": 0.03146303093864709,
98
+ "grad_norm": 1.331021785736084,
99
+ "learning_rate": 0.0005949999999999999,
100
+ "loss": 4.0158,
101
+ "step": 120
102
+ },
103
+ {
104
+ "epoch": 0.03408495018353435,
105
+ "grad_norm": 1.14554762840271,
106
+ "learning_rate": 0.0006450000000000001,
107
+ "loss": 3.9321,
108
+ "step": 130
109
+ },
110
+ {
111
+ "epoch": 0.03670686942842161,
112
+ "grad_norm": 0.9175837635993958,
113
+ "learning_rate": 0.000695,
114
+ "loss": 3.802,
115
+ "step": 140
116
+ },
117
+ {
118
+ "epoch": 0.03932878867330886,
119
+ "grad_norm": 0.7335033416748047,
120
+ "learning_rate": 0.000745,
121
+ "loss": 3.6618,
122
+ "step": 150
123
+ },
124
+ {
125
+ "epoch": 0.04195070791819612,
126
+ "grad_norm": 0.5916274785995483,
127
+ "learning_rate": 0.000795,
128
+ "loss": 3.5341,
129
+ "step": 160
130
+ },
131
+ {
132
+ "epoch": 0.04457262716308338,
133
+ "grad_norm": 0.4947799742221832,
134
+ "learning_rate": 0.0008449999999999999,
135
+ "loss": 3.5311,
136
+ "step": 170
137
+ },
138
+ {
139
+ "epoch": 0.04719454640797063,
140
+ "grad_norm": 0.40263015031814575,
141
+ "learning_rate": 0.0008950000000000001,
142
+ "loss": 3.4709,
143
+ "step": 180
144
+ },
145
+ {
146
+ "epoch": 0.04981646565285789,
147
+ "grad_norm": 0.32677406072616577,
148
+ "learning_rate": 0.000945,
149
+ "loss": 3.2973,
150
+ "step": 190
151
+ },
152
+ {
153
+ "epoch": 0.05243838489774515,
154
+ "grad_norm": 0.3071628212928772,
155
+ "learning_rate": 0.000995,
156
+ "loss": 3.28,
157
+ "step": 200
158
+ },
159
+ {
160
+ "epoch": 0.05506030414263241,
161
+ "grad_norm": 0.3233015835285187,
162
+ "learning_rate": 0.001045,
163
+ "loss": 3.2038,
164
+ "step": 210
165
+ },
166
+ {
167
+ "epoch": 0.05768222338751967,
168
+ "grad_norm": 0.39402100443840027,
169
+ "learning_rate": 0.001095,
170
+ "loss": 3.1627,
171
+ "step": 220
172
+ },
173
+ {
174
+ "epoch": 0.060304142632406924,
175
+ "grad_norm": 0.5528343915939331,
176
+ "learning_rate": 0.001145,
177
+ "loss": 3.1341,
178
+ "step": 230
179
+ },
180
+ {
181
+ "epoch": 0.06292606187729417,
182
+ "grad_norm": 0.4888489842414856,
183
+ "learning_rate": 0.001195,
184
+ "loss": 3.0192,
185
+ "step": 240
186
+ },
187
+ {
188
+ "epoch": 0.06554798112218144,
189
+ "grad_norm": 0.5662292838096619,
190
+ "learning_rate": 0.0012450000000000002,
191
+ "loss": 2.991,
192
+ "step": 250
193
+ },
194
+ {
195
+ "epoch": 0.0681699003670687,
196
+ "grad_norm": 0.5800466537475586,
197
+ "learning_rate": 0.001295,
198
+ "loss": 2.992,
199
+ "step": 260
200
+ },
201
+ {
202
+ "epoch": 0.07079181961195595,
203
+ "grad_norm": 0.5511091947555542,
204
+ "learning_rate": 0.001345,
205
+ "loss": 2.9246,
206
+ "step": 270
207
+ },
208
+ {
209
+ "epoch": 0.07341373885684321,
210
+ "grad_norm": 0.7486537098884583,
211
+ "learning_rate": 0.001395,
212
+ "loss": 2.8996,
213
+ "step": 280
214
+ },
215
+ {
216
+ "epoch": 0.07603565810173046,
217
+ "grad_norm": 0.6995801329612732,
218
+ "learning_rate": 0.001445,
219
+ "loss": 2.7945,
220
+ "step": 290
221
+ },
222
+ {
223
+ "epoch": 0.07865757734661773,
224
+ "grad_norm": 0.7938666939735413,
225
+ "learning_rate": 0.0014950000000000002,
226
+ "loss": 2.7632,
227
+ "step": 300
228
+ },
229
+ {
230
+ "epoch": 0.08127949659150498,
231
+ "grad_norm": 0.7555065155029297,
232
+ "learning_rate": 0.001545,
233
+ "loss": 2.7513,
234
+ "step": 310
235
+ },
236
+ {
237
+ "epoch": 0.08390141583639224,
238
+ "grad_norm": 0.7714865803718567,
239
+ "learning_rate": 0.001595,
240
+ "loss": 2.6165,
241
+ "step": 320
242
+ },
243
+ {
244
+ "epoch": 0.08652333508127949,
245
+ "grad_norm": 0.7604843974113464,
246
+ "learning_rate": 0.001645,
247
+ "loss": 2.6391,
248
+ "step": 330
249
+ },
250
+ {
251
+ "epoch": 0.08914525432616675,
252
+ "grad_norm": 0.7840315699577332,
253
+ "learning_rate": 0.0016950000000000001,
254
+ "loss": 2.5818,
255
+ "step": 340
256
+ },
257
+ {
258
+ "epoch": 0.09176717357105402,
259
+ "grad_norm": 1.0126832723617554,
260
+ "learning_rate": 0.0017450000000000002,
261
+ "loss": 2.5417,
262
+ "step": 350
263
+ },
264
+ {
265
+ "epoch": 0.09438909281594127,
266
+ "grad_norm": 1.0092129707336426,
267
+ "learning_rate": 0.001795,
268
+ "loss": 2.4844,
269
+ "step": 360
270
+ },
271
+ {
272
+ "epoch": 0.09701101206082853,
273
+ "grad_norm": 1.1585489511489868,
274
+ "learning_rate": 0.001845,
275
+ "loss": 2.4645,
276
+ "step": 370
277
+ },
278
+ {
279
+ "epoch": 0.09963293130571578,
280
+ "grad_norm": 1.0778034925460815,
281
+ "learning_rate": 0.001895,
282
+ "loss": 2.4003,
283
+ "step": 380
284
+ },
285
+ {
286
+ "epoch": 0.10225485055060304,
287
+ "grad_norm": 1.146636962890625,
288
+ "learning_rate": 0.0019450000000000001,
289
+ "loss": 2.3466,
290
+ "step": 390
291
+ },
292
+ {
293
+ "epoch": 0.1048767697954903,
294
+ "grad_norm": 0.9742526412010193,
295
+ "learning_rate": 0.0019950000000000002,
296
+ "loss": 2.3088,
297
+ "step": 400
298
+ },
299
+ {
300
+ "epoch": 0.10749868904037756,
301
+ "grad_norm": 1.3035728931427002,
302
+ "learning_rate": 0.0019999657054386192,
303
+ "loss": 2.2834,
304
+ "step": 410
305
+ },
306
+ {
307
+ "epoch": 0.11012060828526482,
308
+ "grad_norm": 1.0689384937286377,
309
+ "learning_rate": 0.0019998471593574603,
310
+ "loss": 2.2473,
311
+ "step": 420
312
+ },
313
+ {
314
+ "epoch": 0.11274252753015207,
315
+ "grad_norm": 1.1519441604614258,
316
+ "learning_rate": 0.001999643948402709,
317
+ "loss": 2.1925,
318
+ "step": 430
319
+ },
320
+ {
321
+ "epoch": 0.11536444677503933,
322
+ "grad_norm": 0.9427940249443054,
323
+ "learning_rate": 0.0019993560897818255,
324
+ "loss": 2.1774,
325
+ "step": 440
326
+ },
327
+ {
328
+ "epoch": 0.11798636601992658,
329
+ "grad_norm": 0.9017934203147888,
330
+ "learning_rate": 0.0019989836078700496,
331
+ "loss": 2.152,
332
+ "step": 450
333
+ },
334
+ {
335
+ "epoch": 0.12060828526481385,
336
+ "grad_norm": 1.018966555595398,
337
+ "learning_rate": 0.001998526534208335,
338
+ "loss": 2.0825,
339
+ "step": 460
340
+ },
341
+ {
342
+ "epoch": 0.1232302045097011,
343
+ "grad_norm": 1.0533466339111328,
344
+ "learning_rate": 0.0019979849075006813,
345
+ "loss": 2.1358,
346
+ "step": 470
347
+ },
348
+ {
349
+ "epoch": 0.12585212375458835,
350
+ "grad_norm": 0.941605806350708,
351
+ "learning_rate": 0.001997358773610856,
352
+ "loss": 2.0524,
353
+ "step": 480
354
+ },
355
+ {
356
+ "epoch": 0.12847404299947562,
357
+ "grad_norm": 0.8877449035644531,
358
+ "learning_rate": 0.0019966481855585075,
359
+ "loss": 2.0308,
360
+ "step": 490
361
+ },
362
+ {
363
+ "epoch": 0.13109596224436287,
364
+ "grad_norm": 0.8652307391166687,
365
+ "learning_rate": 0.001995853203514682,
366
+ "loss": 2.012,
367
+ "step": 500
368
+ },
369
+ {
370
+ "epoch": 0.13371788148925012,
371
+ "grad_norm": 0.8943641781806946,
372
+ "learning_rate": 0.0019949738947967217,
373
+ "loss": 1.9729,
374
+ "step": 510
375
+ },
376
+ {
377
+ "epoch": 0.1363398007341374,
378
+ "grad_norm": 0.9359736442565918,
379
+ "learning_rate": 0.001994010333862568,
380
+ "loss": 1.9997,
381
+ "step": 520
382
+ },
383
+ {
384
+ "epoch": 0.13896171997902465,
385
+ "grad_norm": 1.0085017681121826,
386
+ "learning_rate": 0.001992962602304456,
387
+ "loss": 1.937,
388
+ "step": 530
389
+ },
390
+ {
391
+ "epoch": 0.1415836392239119,
392
+ "grad_norm": 0.7549618482589722,
393
+ "learning_rate": 0.0019918307888420065,
394
+ "loss": 1.9268,
395
+ "step": 540
396
+ },
397
+ {
398
+ "epoch": 0.14420555846879915,
399
+ "grad_norm": 0.8932085037231445,
400
+ "learning_rate": 0.0019906149893147104,
401
+ "loss": 1.9014,
402
+ "step": 550
403
+ },
404
+ {
405
+ "epoch": 0.14682747771368643,
406
+ "grad_norm": 0.8130724430084229,
407
+ "learning_rate": 0.001989315306673817,
408
+ "loss": 1.8577,
409
+ "step": 560
410
+ },
411
+ {
412
+ "epoch": 0.14944939695857368,
413
+ "grad_norm": 0.8497139811515808,
414
+ "learning_rate": 0.0019879318509736137,
415
+ "loss": 1.8185,
416
+ "step": 570
417
+ },
418
+ {
419
+ "epoch": 0.15207131620346093,
420
+ "grad_norm": 0.6299962997436523,
421
+ "learning_rate": 0.001986464739362106,
422
+ "loss": 1.811,
423
+ "step": 580
424
+ },
425
+ {
426
+ "epoch": 0.1546932354483482,
427
+ "grad_norm": 0.7180768251419067,
428
+ "learning_rate": 0.0019849140960711024,
429
+ "loss": 1.7944,
430
+ "step": 590
431
+ },
432
+ {
433
+ "epoch": 0.15731515469323545,
434
+ "grad_norm": 0.8082334399223328,
435
+ "learning_rate": 0.0019832800524056888,
436
+ "loss": 1.8333,
437
+ "step": 600
438
+ },
439
+ {
440
+ "epoch": 0.1599370739381227,
441
+ "grad_norm": 0.8284159302711487,
442
+ "learning_rate": 0.0019815627467331142,
443
+ "loss": 1.811,
444
+ "step": 610
445
+ },
446
+ {
447
+ "epoch": 0.16255899318300995,
448
+ "grad_norm": 0.7332941293716431,
449
+ "learning_rate": 0.0019797623244710715,
450
+ "loss": 1.7704,
451
+ "step": 620
452
+ },
453
+ {
454
+ "epoch": 0.16518091242789723,
455
+ "grad_norm": 0.7234723567962646,
456
+ "learning_rate": 0.0019778789380753862,
457
+ "loss": 1.7558,
458
+ "step": 630
459
+ },
460
+ {
461
+ "epoch": 0.16780283167278448,
462
+ "grad_norm": 0.693242073059082,
463
+ "learning_rate": 0.001975912747027104,
464
+ "loss": 1.742,
465
+ "step": 640
466
+ },
467
+ {
468
+ "epoch": 0.17042475091767173,
469
+ "grad_norm": 0.8523733019828796,
470
+ "learning_rate": 0.0019738639178189885,
471
+ "loss": 1.7438,
472
+ "step": 650
473
+ },
474
+ {
475
+ "epoch": 0.17304667016255898,
476
+ "grad_norm": 0.7505561709403992,
477
+ "learning_rate": 0.001971732623941422,
478
+ "loss": 1.7251,
479
+ "step": 660
480
+ },
481
+ {
482
+ "epoch": 0.17566858940744626,
483
+ "grad_norm": 0.7338821887969971,
484
+ "learning_rate": 0.0019695190458677144,
485
+ "loss": 1.7281,
486
+ "step": 670
487
+ },
488
+ {
489
+ "epoch": 0.1782905086523335,
490
+ "grad_norm": 0.8278585076332092,
491
+ "learning_rate": 0.001967223371038823,
492
+ "loss": 1.6983,
493
+ "step": 680
494
+ },
495
+ {
496
+ "epoch": 0.18091242789722076,
497
+ "grad_norm": 0.6785498261451721,
498
+ "learning_rate": 0.0019648457938474776,
499
+ "loss": 1.7018,
500
+ "step": 690
501
+ },
502
+ {
503
+ "epoch": 0.18353434714210803,
504
+ "grad_norm": 0.7954968810081482,
505
+ "learning_rate": 0.0019623865156217215,
506
+ "loss": 1.6978,
507
+ "step": 700
508
+ },
509
+ {
510
+ "epoch": 0.18615626638699528,
511
+ "grad_norm": 0.6877925992012024,
512
+ "learning_rate": 0.001959845744607864,
513
+ "loss": 1.6693,
514
+ "step": 710
515
+ },
516
+ {
517
+ "epoch": 0.18877818563188253,
518
+ "grad_norm": 0.6183112859725952,
519
+ "learning_rate": 0.001957223695952844,
520
+ "loss": 1.656,
521
+ "step": 720
522
+ },
523
+ {
524
+ "epoch": 0.19140010487676978,
525
+ "grad_norm": 0.6864896416664124,
526
+ "learning_rate": 0.0019545205916860152,
527
+ "loss": 1.6188,
528
+ "step": 730
529
+ },
530
+ {
531
+ "epoch": 0.19402202412165706,
532
+ "grad_norm": 0.6678555011749268,
533
+ "learning_rate": 0.0019517366607003429,
534
+ "loss": 1.6195,
535
+ "step": 740
536
+ },
537
+ {
538
+ "epoch": 0.1966439433665443,
539
+ "grad_norm": 0.724320113658905,
540
+ "learning_rate": 0.0019488721387330222,
541
+ "loss": 1.6067,
542
+ "step": 750
543
+ },
544
+ {
545
+ "epoch": 0.19926586261143156,
546
+ "grad_norm": 0.6665757298469543,
547
+ "learning_rate": 0.0019459272683455162,
548
+ "loss": 1.5781,
549
+ "step": 760
550
+ },
551
+ {
552
+ "epoch": 0.20188778185631884,
553
+ "grad_norm": 0.7139772772789001,
554
+ "learning_rate": 0.0019429022989030176,
555
+ "loss": 1.5647,
556
+ "step": 770
557
+ },
558
+ {
559
+ "epoch": 0.2045097011012061,
560
+ "grad_norm": 0.6505457758903503,
561
+ "learning_rate": 0.0019397974865533315,
562
+ "loss": 1.5869,
563
+ "step": 780
564
+ },
565
+ {
566
+ "epoch": 0.20713162034609334,
567
+ "grad_norm": 0.6815754175186157,
568
+ "learning_rate": 0.001936613094205186,
569
+ "loss": 1.5848,
570
+ "step": 790
571
+ },
572
+ {
573
+ "epoch": 0.2097535395909806,
574
+ "grad_norm": 0.6977171897888184,
575
+ "learning_rate": 0.00193334939150597,
576
+ "loss": 1.5284,
577
+ "step": 800
578
+ },
579
+ {
580
+ "epoch": 0.21237545883586786,
581
+ "grad_norm": 0.5965753197669983,
582
+ "learning_rate": 0.0019300066548188998,
583
+ "loss": 1.5468,
584
+ "step": 810
585
+ },
586
+ {
587
+ "epoch": 0.2149973780807551,
588
+ "grad_norm": 0.596052885055542,
589
+ "learning_rate": 0.001926585167199616,
590
+ "loss": 1.5579,
591
+ "step": 820
592
+ },
593
+ {
594
+ "epoch": 0.21761929732564236,
595
+ "grad_norm": 0.6821017861366272,
596
+ "learning_rate": 0.001923085218372218,
597
+ "loss": 1.4984,
598
+ "step": 830
599
+ },
600
+ {
601
+ "epoch": 0.22024121657052964,
602
+ "grad_norm": 0.6523297429084778,
603
+ "learning_rate": 0.0019195071047047277,
604
+ "loss": 1.537,
605
+ "step": 840
606
+ },
607
+ {
608
+ "epoch": 0.2228631358154169,
609
+ "grad_norm": 0.648935079574585,
610
+ "learning_rate": 0.0019158511291839945,
611
+ "loss": 1.5192,
612
+ "step": 850
613
+ },
614
+ {
615
+ "epoch": 0.22548505506030414,
616
+ "grad_norm": 0.6102792620658875,
617
+ "learning_rate": 0.0019121176013900407,
618
+ "loss": 1.5209,
619
+ "step": 860
620
+ },
621
+ {
622
+ "epoch": 0.2281069743051914,
623
+ "grad_norm": 0.6573307514190674,
624
+ "learning_rate": 0.0019083068374698448,
625
+ "loss": 1.49,
626
+ "step": 870
627
+ },
628
+ {
629
+ "epoch": 0.23072889355007867,
630
+ "grad_norm": 0.6355723738670349,
631
+ "learning_rate": 0.0019044191601105727,
632
+ "loss": 1.4929,
633
+ "step": 880
634
+ },
635
+ {
636
+ "epoch": 0.23335081279496592,
637
+ "grad_norm": 0.5931225419044495,
638
+ "learning_rate": 0.0019004548985122511,
639
+ "loss": 1.4813,
640
+ "step": 890
641
+ },
642
+ {
643
+ "epoch": 0.23597273203985317,
644
+ "grad_norm": 0.6640650629997253,
645
+ "learning_rate": 0.0018964143883598936,
646
+ "loss": 1.4808,
647
+ "step": 900
648
+ },
649
+ {
650
+ "epoch": 0.23859465128474042,
651
+ "grad_norm": 0.6377866268157959,
652
+ "learning_rate": 0.0018922979717950748,
653
+ "loss": 1.4901,
654
+ "step": 910
655
+ },
656
+ {
657
+ "epoch": 0.2412165705296277,
658
+ "grad_norm": 0.6502982378005981,
659
+ "learning_rate": 0.0018881059973869581,
660
+ "loss": 1.4501,
661
+ "step": 920
662
+ },
663
+ {
664
+ "epoch": 0.24383848977451494,
665
+ "grad_norm": 0.602969765663147,
666
+ "learning_rate": 0.0018838388201027805,
667
+ "loss": 1.4661,
668
+ "step": 930
669
+ },
670
+ {
671
+ "epoch": 0.2464604090194022,
672
+ "grad_norm": 0.6061879396438599,
673
+ "learning_rate": 0.001879496801277794,
674
+ "loss": 1.4408,
675
+ "step": 940
676
+ },
677
+ {
678
+ "epoch": 0.24908232826428947,
679
+ "grad_norm": 0.8049127459526062,
680
+ "learning_rate": 0.001875080308584669,
681
+ "loss": 1.4466,
682
+ "step": 950
683
+ },
684
+ {
685
+ "epoch": 0.2517042475091767,
686
+ "grad_norm": 0.46771517395973206,
687
+ "learning_rate": 0.00187058971600236,
688
+ "loss": 1.4382,
689
+ "step": 960
690
+ },
691
+ {
692
+ "epoch": 0.254326166754064,
693
+ "grad_norm": 0.6081333756446838,
694
+ "learning_rate": 0.001866025403784439,
695
+ "loss": 1.4518,
696
+ "step": 970
697
+ },
698
+ {
699
+ "epoch": 0.25694808599895125,
700
+ "grad_norm": 0.6247040033340454,
701
+ "learning_rate": 0.0018613877584268944,
702
+ "loss": 1.4639,
703
+ "step": 980
704
+ },
705
+ {
706
+ "epoch": 0.2595700052438385,
707
+ "grad_norm": 0.5699506998062134,
708
+ "learning_rate": 0.0018566771726354063,
709
+ "loss": 1.4218,
710
+ "step": 990
711
+ },
712
+ {
713
+ "epoch": 0.26219192448872575,
714
+ "grad_norm": 0.5360729694366455,
715
+ "learning_rate": 0.0018518940452920906,
716
+ "loss": 1.4189,
717
+ "step": 1000
718
+ },
719
+ {
720
+ "epoch": 0.264813843733613,
721
+ "grad_norm": 0.5921474695205688,
722
+ "learning_rate": 0.0018470387814217232,
723
+ "loss": 1.424,
724
+ "step": 1010
725
+ },
726
+ {
727
+ "epoch": 0.26743576297850025,
728
+ "grad_norm": 0.6162559986114502,
729
+ "learning_rate": 0.0018421117921574438,
730
+ "loss": 1.4307,
731
+ "step": 1020
732
+ },
733
+ {
734
+ "epoch": 0.2700576822233875,
735
+ "grad_norm": 0.5530286431312561,
736
+ "learning_rate": 0.001837113494705942,
737
+ "loss": 1.4158,
738
+ "step": 1030
739
+ },
740
+ {
741
+ "epoch": 0.2726796014682748,
742
+ "grad_norm": 0.5585499405860901,
743
+ "learning_rate": 0.0018320443123121283,
744
+ "loss": 1.3861,
745
+ "step": 1040
746
+ },
747
+ {
748
+ "epoch": 0.27530152071316205,
749
+ "grad_norm": 0.6225973963737488,
750
+ "learning_rate": 0.0018269046742232966,
751
+ "loss": 1.3942,
752
+ "step": 1050
753
+ },
754
+ {
755
+ "epoch": 0.2779234399580493,
756
+ "grad_norm": 0.49642321467399597,
757
+ "learning_rate": 0.0018216950156527737,
758
+ "loss": 1.3912,
759
+ "step": 1060
760
+ },
761
+ {
762
+ "epoch": 0.28054535920293655,
763
+ "grad_norm": 0.6089576482772827,
764
+ "learning_rate": 0.0018164157777430681,
765
+ "loss": 1.3732,
766
+ "step": 1070
767
+ },
768
+ {
769
+ "epoch": 0.2831672784478238,
770
+ "grad_norm": 0.5753847360610962,
771
+ "learning_rate": 0.0018110674075285157,
772
+ "loss": 1.398,
773
+ "step": 1080
774
+ },
775
+ {
776
+ "epoch": 0.28578919769271105,
777
+ "grad_norm": 0.5357734560966492,
778
+ "learning_rate": 0.0018056503578974242,
779
+ "loss": 1.3851,
780
+ "step": 1090
781
+ },
782
+ {
783
+ "epoch": 0.2884111169375983,
784
+ "grad_norm": 0.5319791436195374,
785
+ "learning_rate": 0.001800165087553724,
786
+ "loss": 1.3804,
787
+ "step": 1100
788
+ },
789
+ {
790
+ "epoch": 0.2910330361824856,
791
+ "grad_norm": 0.5765709280967712,
792
+ "learning_rate": 0.0017946120609781276,
793
+ "loss": 1.3534,
794
+ "step": 1110
795
+ },
796
+ {
797
+ "epoch": 0.29365495542737285,
798
+ "grad_norm": 0.48765453696250916,
799
+ "learning_rate": 0.001788991748388796,
800
+ "loss": 1.3693,
801
+ "step": 1120
802
+ },
803
+ {
804
+ "epoch": 0.2962768746722601,
805
+ "grad_norm": 0.5916075110435486,
806
+ "learning_rate": 0.001783304625701524,
807
+ "loss": 1.3697,
808
+ "step": 1130
809
+ },
810
+ {
811
+ "epoch": 0.29889879391714735,
812
+ "grad_norm": 0.411699503660202,
813
+ "learning_rate": 0.0017775511744894384,
814
+ "loss": 1.3588,
815
+ "step": 1140
816
+ },
817
+ {
818
+ "epoch": 0.3015207131620346,
819
+ "grad_norm": 0.5155631899833679,
820
+ "learning_rate": 0.0017717318819422214,
821
+ "loss": 1.3697,
822
+ "step": 1150
823
+ },
824
+ {
825
+ "epoch": 0.30414263240692185,
826
+ "grad_norm": 0.5687488913536072,
827
+ "learning_rate": 0.0017658472408248551,
828
+ "loss": 1.3558,
829
+ "step": 1160
830
+ },
831
+ {
832
+ "epoch": 0.3067645516518091,
833
+ "grad_norm": 0.5609891414642334,
834
+ "learning_rate": 0.0017598977494358967,
835
+ "loss": 1.3376,
836
+ "step": 1170
837
+ },
838
+ {
839
+ "epoch": 0.3093864708966964,
840
+ "grad_norm": 0.5137512683868408,
841
+ "learning_rate": 0.0017538839115652817,
842
+ "loss": 1.3534,
843
+ "step": 1180
844
+ },
845
+ {
846
+ "epoch": 0.31200839014158366,
847
+ "grad_norm": 0.5840641260147095,
848
+ "learning_rate": 0.001747806236451666,
849
+ "loss": 1.3394,
850
+ "step": 1190
851
+ },
852
+ {
853
+ "epoch": 0.3146303093864709,
854
+ "grad_norm": 0.5758949518203735,
855
+ "learning_rate": 0.0017416652387393027,
856
+ "loss": 1.3417,
857
+ "step": 1200
858
+ },
859
+ {
860
+ "epoch": 0.31725222863135816,
861
+ "grad_norm": 0.5121742486953735,
862
+ "learning_rate": 0.0017354614384344658,
863
+ "loss": 1.341,
864
+ "step": 1210
865
+ },
866
+ {
867
+ "epoch": 0.3198741478762454,
868
+ "grad_norm": 0.5056650638580322,
869
+ "learning_rate": 0.001729195360861414,
870
+ "loss": 1.316,
871
+ "step": 1220
872
+ },
873
+ {
874
+ "epoch": 0.32249606712113266,
875
+ "grad_norm": 0.4782615602016449,
876
+ "learning_rate": 0.0017228675366179106,
877
+ "loss": 1.3226,
878
+ "step": 1230
879
+ },
880
+ {
881
+ "epoch": 0.3251179863660199,
882
+ "grad_norm": 0.49403342604637146,
883
+ "learning_rate": 0.0017164785015302906,
884
+ "loss": 1.37,
885
+ "step": 1240
886
+ },
887
+ {
888
+ "epoch": 0.3277399056109072,
889
+ "grad_norm": 0.4836321175098419,
890
+ "learning_rate": 0.0017100287966080906,
891
+ "loss": 1.3272,
892
+ "step": 1250
893
+ },
894
+ {
895
+ "epoch": 0.33036182485579446,
896
+ "grad_norm": 0.48174890875816345,
897
+ "learning_rate": 0.001703518967998236,
898
+ "loss": 1.3148,
899
+ "step": 1260
900
+ },
901
+ {
902
+ "epoch": 0.3329837441006817,
903
+ "grad_norm": 0.4627121090888977,
904
+ "learning_rate": 0.001696949566938795,
905
+ "loss": 1.3161,
906
+ "step": 1270
907
+ },
908
+ {
909
+ "epoch": 0.33560566334556896,
910
+ "grad_norm": 0.470414936542511,
911
+ "learning_rate": 0.0016903211497123003,
912
+ "loss": 1.3313,
913
+ "step": 1280
914
+ },
915
+ {
916
+ "epoch": 0.3382275825904562,
917
+ "grad_norm": 0.4437310993671417,
918
+ "learning_rate": 0.0016836342775986446,
919
+ "loss": 1.3073,
920
+ "step": 1290
921
+ },
922
+ {
923
+ "epoch": 0.34084950183534346,
924
+ "grad_norm": 0.47688329219818115,
925
+ "learning_rate": 0.0016768895168275534,
926
+ "loss": 1.3128,
927
+ "step": 1300
928
+ },
929
+ {
930
+ "epoch": 0.3434714210802307,
931
+ "grad_norm": 0.5143507122993469,
932
+ "learning_rate": 0.0016700874385306363,
933
+ "loss": 1.3357,
934
+ "step": 1310
935
+ },
936
+ {
937
+ "epoch": 0.34609334032511796,
938
+ "grad_norm": 0.4100657105445862,
939
+ "learning_rate": 0.0016632286186930275,
940
+ "loss": 1.3061,
941
+ "step": 1320
942
+ },
943
+ {
944
+ "epoch": 0.34871525957000526,
945
+ "grad_norm": 0.4421868920326233,
946
+ "learning_rate": 0.0016563136381046088,
947
+ "loss": 1.3158,
948
+ "step": 1330
949
+ },
950
+ {
951
+ "epoch": 0.3513371788148925,
952
+ "grad_norm": 0.4668099582195282,
953
+ "learning_rate": 0.0016493430823108332,
954
+ "loss": 1.3088,
955
+ "step": 1340
956
+ },
957
+ {
958
+ "epoch": 0.35395909805977976,
959
+ "grad_norm": 0.5451709032058716,
960
+ "learning_rate": 0.0016423175415631404,
961
+ "loss": 1.3344,
962
+ "step": 1350
963
+ },
964
+ {
965
+ "epoch": 0.356581017304667,
966
+ "grad_norm": 0.45294106006622314,
967
+ "learning_rate": 0.0016352376107689754,
968
+ "loss": 1.2778,
969
+ "step": 1360
970
+ },
971
+ {
972
+ "epoch": 0.35920293654955426,
973
+ "grad_norm": 0.4404051601886749,
974
+ "learning_rate": 0.0016281038894414143,
975
+ "loss": 1.2871,
976
+ "step": 1370
977
+ },
978
+ {
979
+ "epoch": 0.3618248557944415,
980
+ "grad_norm": 0.45863279700279236,
981
+ "learning_rate": 0.0016209169816483971,
982
+ "loss": 1.3286,
983
+ "step": 1380
984
+ },
985
+ {
986
+ "epoch": 0.36444677503932876,
987
+ "grad_norm": 0.45011425018310547,
988
+ "learning_rate": 0.0016136774959615784,
989
+ "loss": 1.2979,
990
+ "step": 1390
991
+ },
992
+ {
993
+ "epoch": 0.36706869428421607,
994
+ "grad_norm": 0.5113876461982727,
995
+ "learning_rate": 0.0016063860454047943,
996
+ "loss": 1.3088,
997
+ "step": 1400
998
+ },
999
+ {
1000
+ "epoch": 0.3696906135291033,
1001
+ "grad_norm": 0.40740302205085754,
1002
+ "learning_rate": 0.001599043247402151,
1003
+ "loss": 1.2703,
1004
+ "step": 1410
1005
+ },
1006
+ {
1007
+ "epoch": 0.37231253277399057,
1008
+ "grad_norm": 0.4261358976364136,
1009
+ "learning_rate": 0.0015916497237257455,
1010
+ "loss": 1.2681,
1011
+ "step": 1420
1012
+ },
1013
+ {
1014
+ "epoch": 0.3749344520188778,
1015
+ "grad_norm": 0.4349290132522583,
1016
+ "learning_rate": 0.0015842061004430145,
1017
+ "loss": 1.317,
1018
+ "step": 1430
1019
+ },
1020
+ {
1021
+ "epoch": 0.37755637126376507,
1022
+ "grad_norm": 0.4363626539707184,
1023
+ "learning_rate": 0.0015767130078637183,
1024
+ "loss": 1.2707,
1025
+ "step": 1440
1026
+ },
1027
+ {
1028
+ "epoch": 0.3801782905086523,
1029
+ "grad_norm": 0.41238006949424744,
1030
+ "learning_rate": 0.0015691710804865706,
1031
+ "loss": 1.2763,
1032
+ "step": 1450
1033
+ },
1034
+ {
1035
+ "epoch": 0.38280020975353957,
1036
+ "grad_norm": 0.476226270198822,
1037
+ "learning_rate": 0.0015615809569455089,
1038
+ "loss": 1.3037,
1039
+ "step": 1460
1040
+ },
1041
+ {
1042
+ "epoch": 0.38542212899842687,
1043
+ "grad_norm": 0.45900896191596985,
1044
+ "learning_rate": 0.0015539432799556159,
1045
+ "loss": 1.287,
1046
+ "step": 1470
1047
+ },
1048
+ {
1049
+ "epoch": 0.3880440482433141,
1050
+ "grad_norm": 0.3873949348926544,
1051
+ "learning_rate": 0.0015462586962586972,
1052
+ "loss": 1.2793,
1053
+ "step": 1480
1054
+ },
1055
+ {
1056
+ "epoch": 0.39066596748820137,
1057
+ "grad_norm": 0.4380306601524353,
1058
+ "learning_rate": 0.001538527856568515,
1059
+ "loss": 1.2916,
1060
+ "step": 1490
1061
+ },
1062
+ {
1063
+ "epoch": 0.3932878867330886,
1064
+ "grad_norm": 0.39479300379753113,
1065
+ "learning_rate": 0.0015307514155156895,
1066
+ "loss": 1.272,
1067
+ "step": 1500
1068
+ }
1069
+ ],
1070
+ "logging_steps": 10,
1071
+ "max_steps": 3814,
1072
+ "num_input_tokens_seen": 0,
1073
+ "num_train_epochs": 9223372036854775807,
1074
+ "save_steps": 500,
1075
+ "stateful_callbacks": {
1076
+ "TrainerControl": {
1077
+ "args": {
1078
+ "should_epoch_stop": false,
1079
+ "should_evaluate": false,
1080
+ "should_log": false,
1081
+ "should_save": true,
1082
+ "should_training_stop": false
1083
+ },
1084
+ "attributes": {}
1085
+ }
1086
+ },
1087
+ "total_flos": 1.0095396499845284e+18,
1088
+ "train_batch_size": 64,
1089
+ "trial_name": null,
1090
+ "trial_params": null
1091
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81c91fc77496ffd4fb9e10a9ec46455f5246c9c164ddf0a8d2f8b08013987959
3
+ size 5777