Synth-Ecosystem commited on
Commit
c84cd51
·
verified ·
1 Parent(s): 1e940a4

Training in progress, step 50, checkpoint

Browse files
last-checkpoint/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:64b76747407c4e35f82034686c204e47c073dcbd4896d28d888988234fd4be93
3
  size 38572808
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f043eac6489bbc6287f3aaaaadc5830f160d3bfd157fcdda7262517b534c4c4f
3
  size 38572808
last-checkpoint/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:df890ec8e4a14bf7d664d8fd147f1541aec59d062ef9c7e66534fdd2527e739f
3
- size 39471557
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69216f02fdc7869623e7f66ffea6cc5d5aa2413984919305f0b4ef843c872d43
3
+ size 39471301
last-checkpoint/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:18efa3d16ff8a661446c6070fb2493803837ba9e7ee897995792e655b332ffc8
3
  size 14645
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0786d799bd9ca9eae2880fb41a0c195765ff948c40e6a01cd4c84edb1955805b
3
  size 14645
last-checkpoint/scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:90d8c8575f4efeeaec2cef3c0422fa1c701b9e4c6a42193dced73fa5ae08ddb0
3
  size 1465
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:103f48442c46172aa72965a6fa450502ab3136566ae7bb88ba108ea05eab43ec
3
  size 1465
last-checkpoint/trainer_state.json CHANGED
@@ -2,312 +2,62 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.48,
6
  "eval_steps": 500,
7
- "global_step": 300,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
- "entropy": 5.946600067615509,
14
- "epoch": 0.016,
15
- "grad_norm": 3.484375,
16
- "learning_rate": 0.00017460000000000002,
17
- "loss": 6.29835319519043,
18
- "mean_token_accuracy": 0.15247181179001928,
19
- "num_tokens": 22300.0,
20
  "step": 10
21
  },
22
  {
23
- "entropy": 3.9428867518901827,
24
- "epoch": 0.032,
25
- "grad_norm": 3.0,
26
- "learning_rate": 0.0001686,
27
- "loss": 3.982344055175781,
28
- "mean_token_accuracy": 0.385134282708168,
29
- "num_tokens": 46965.0,
30
  "step": 20
31
  },
32
  {
33
- "entropy": 3.1769737243652343,
34
- "epoch": 0.048,
35
- "grad_norm": 3.015625,
36
- "learning_rate": 0.0001626,
37
- "loss": 3.274928665161133,
38
- "mean_token_accuracy": 0.5095567837357521,
39
- "num_tokens": 71076.0,
40
  "step": 30
41
  },
42
  {
43
- "entropy": 2.869546139240265,
44
- "epoch": 0.064,
45
- "grad_norm": 2.640625,
46
- "learning_rate": 0.0001566,
47
- "loss": 2.9933839797973634,
48
- "mean_token_accuracy": 0.5253676064312458,
49
- "num_tokens": 97013.0,
50
  "step": 40
51
  },
52
  {
53
- "entropy": 2.465953287482262,
54
- "epoch": 0.08,
55
- "grad_norm": 2.21875,
56
- "learning_rate": 0.0001506,
57
- "loss": 2.62363224029541,
58
- "mean_token_accuracy": 0.5868253096938133,
59
- "num_tokens": 120513.0,
60
- "step": 50
61
- },
62
- {
63
- "entropy": 2.52132785320282,
64
- "epoch": 0.096,
65
- "grad_norm": 2.265625,
66
- "learning_rate": 0.00014460000000000002,
67
- "loss": 2.5625015258789063,
68
- "mean_token_accuracy": 0.5602172784507274,
69
- "num_tokens": 145386.0,
70
- "step": 60
71
- },
72
- {
73
- "entropy": 2.4292581140995027,
74
- "epoch": 0.112,
75
- "grad_norm": 2.671875,
76
- "learning_rate": 0.0001386,
77
- "loss": 2.5273454666137694,
78
- "mean_token_accuracy": 0.5630464181303978,
79
- "num_tokens": 171111.0,
80
- "step": 70
81
- },
82
- {
83
- "entropy": 2.053872545063496,
84
- "epoch": 0.128,
85
- "grad_norm": 2.28125,
86
- "learning_rate": 0.00013260000000000002,
87
- "loss": 2.1423622131347657,
88
- "mean_token_accuracy": 0.6182360425591469,
89
- "num_tokens": 195172.0,
90
- "step": 80
91
- },
92
- {
93
- "entropy": 1.8883190169930457,
94
- "epoch": 0.144,
95
- "grad_norm": 2.046875,
96
- "learning_rate": 0.00012660000000000001,
97
- "loss": 1.9964679718017577,
98
- "mean_token_accuracy": 0.635661780834198,
99
- "num_tokens": 218330.0,
100
- "step": 90
101
- },
102
- {
103
- "entropy": 1.9393853038549422,
104
- "epoch": 0.16,
105
  "grad_norm": 2.40625,
106
- "learning_rate": 0.00012060000000000002,
107
- "loss": 2.022273063659668,
108
- "mean_token_accuracy": 0.6229535043239594,
109
- "num_tokens": 244458.0,
110
- "step": 100
111
- },
112
- {
113
- "entropy": 1.7916773557662964,
114
- "epoch": 0.176,
115
- "grad_norm": 2.4375,
116
- "learning_rate": 0.00011460000000000002,
117
- "loss": 1.837610626220703,
118
- "mean_token_accuracy": 0.6510058090090751,
119
- "num_tokens": 268597.0,
120
- "step": 110
121
- },
122
- {
123
- "entropy": 1.6508349940180778,
124
- "epoch": 0.192,
125
- "grad_norm": 2.1875,
126
- "learning_rate": 0.00010860000000000001,
127
- "loss": 1.7676301956176759,
128
- "mean_token_accuracy": 0.6538835391402245,
129
- "num_tokens": 293619.0,
130
- "step": 120
131
- },
132
- {
133
- "entropy": 1.550324660539627,
134
- "epoch": 0.208,
135
- "grad_norm": 2.453125,
136
- "learning_rate": 0.0001026,
137
- "loss": 1.6161380767822267,
138
- "mean_token_accuracy": 0.6810771860182285,
139
- "num_tokens": 316757.0,
140
- "step": 130
141
- },
142
- {
143
- "entropy": 1.5475640408694744,
144
- "epoch": 0.224,
145
- "grad_norm": 2.859375,
146
- "learning_rate": 9.66e-05,
147
- "loss": 1.6725698471069337,
148
- "mean_token_accuracy": 0.676793497800827,
149
- "num_tokens": 341371.0,
150
- "step": 140
151
- },
152
- {
153
- "entropy": 1.5467152088880538,
154
- "epoch": 0.24,
155
- "grad_norm": 1.890625,
156
- "learning_rate": 9.059999999999999e-05,
157
- "loss": 1.5721678733825684,
158
- "mean_token_accuracy": 0.6773946091532708,
159
- "num_tokens": 366204.0,
160
- "step": 150
161
- },
162
- {
163
- "entropy": 1.4623933732509613,
164
- "epoch": 0.256,
165
- "grad_norm": 2.296875,
166
- "learning_rate": 8.46e-05,
167
- "loss": 1.5432461738586425,
168
- "mean_token_accuracy": 0.6877943009138108,
169
- "num_tokens": 390253.0,
170
- "step": 160
171
- },
172
- {
173
- "entropy": 1.4584963083267213,
174
- "epoch": 0.272,
175
- "grad_norm": 1.7578125,
176
- "learning_rate": 7.86e-05,
177
- "loss": 1.5031940460205078,
178
- "mean_token_accuracy": 0.6966861024498939,
179
- "num_tokens": 414662.0,
180
- "step": 170
181
- },
182
- {
183
- "entropy": 1.2456398598849774,
184
- "epoch": 0.288,
185
- "grad_norm": 1.9921875,
186
- "learning_rate": 7.26e-05,
187
- "loss": 1.378309154510498,
188
- "mean_token_accuracy": 0.726311057806015,
189
- "num_tokens": 437835.0,
190
- "step": 180
191
- },
192
- {
193
- "entropy": 1.2711421206593514,
194
- "epoch": 0.304,
195
- "grad_norm": 2.4375,
196
- "learning_rate": 6.66e-05,
197
- "loss": 1.3943452835083008,
198
- "mean_token_accuracy": 0.7222944989800453,
199
- "num_tokens": 462225.0,
200
- "step": 190
201
- },
202
- {
203
- "entropy": 1.3176883950829505,
204
- "epoch": 0.32,
205
- "grad_norm": 1.9765625,
206
- "learning_rate": 6.06e-05,
207
- "loss": 1.3720195770263672,
208
- "mean_token_accuracy": 0.7194708675146103,
209
- "num_tokens": 485468.0,
210
- "step": 200
211
- },
212
- {
213
- "entropy": 1.4132127992808818,
214
- "epoch": 0.336,
215
- "grad_norm": 1.71875,
216
- "learning_rate": 5.4600000000000006e-05,
217
- "loss": 1.532619094848633,
218
- "mean_token_accuracy": 0.6904794305562973,
219
- "num_tokens": 511653.0,
220
- "step": 210
221
- },
222
- {
223
- "entropy": 1.3034262195229531,
224
- "epoch": 0.352,
225
- "grad_norm": 1.875,
226
- "learning_rate": 4.860000000000001e-05,
227
- "loss": 1.3831446647644043,
228
- "mean_token_accuracy": 0.7147367805242538,
229
- "num_tokens": 535939.0,
230
- "step": 220
231
- },
232
- {
233
- "entropy": 1.2737671211361885,
234
- "epoch": 0.368,
235
- "grad_norm": 1.828125,
236
- "learning_rate": 4.2600000000000005e-05,
237
- "loss": 1.321816349029541,
238
- "mean_token_accuracy": 0.721889540553093,
239
- "num_tokens": 558895.0,
240
- "step": 230
241
- },
242
- {
243
- "entropy": 1.2787693426012994,
244
- "epoch": 0.384,
245
- "grad_norm": 1.9453125,
246
- "learning_rate": 3.66e-05,
247
- "loss": 1.3669401168823243,
248
- "mean_token_accuracy": 0.721854430437088,
249
- "num_tokens": 581828.0,
250
- "step": 240
251
- },
252
- {
253
- "entropy": 1.3367563620209695,
254
- "epoch": 0.4,
255
- "grad_norm": 1.6328125,
256
- "learning_rate": 3.0600000000000005e-05,
257
- "loss": 1.382511520385742,
258
- "mean_token_accuracy": 0.7014128029346466,
259
- "num_tokens": 607581.0,
260
- "step": 250
261
- },
262
- {
263
- "entropy": 1.2327990546822547,
264
- "epoch": 0.416,
265
- "grad_norm": 1.640625,
266
- "learning_rate": 2.46e-05,
267
- "loss": 1.3629666328430177,
268
- "mean_token_accuracy": 0.7295151136815547,
269
- "num_tokens": 630821.0,
270
- "step": 260
271
- },
272
- {
273
- "entropy": 1.2706819713115691,
274
- "epoch": 0.432,
275
- "grad_norm": 1.90625,
276
- "learning_rate": 1.86e-05,
277
- "loss": 1.3413145065307617,
278
- "mean_token_accuracy": 0.7252940103411675,
279
- "num_tokens": 654722.0,
280
- "step": 270
281
- },
282
- {
283
- "entropy": 1.4314473286271094,
284
- "epoch": 0.448,
285
- "grad_norm": 1.7421875,
286
- "learning_rate": 1.2600000000000001e-05,
287
- "loss": 1.48402681350708,
288
- "mean_token_accuracy": 0.6887813597917557,
289
- "num_tokens": 680750.0,
290
- "step": 280
291
- },
292
- {
293
- "entropy": 1.2526817061007023,
294
- "epoch": 0.464,
295
- "grad_norm": 1.7421875,
296
- "learning_rate": 6.6e-06,
297
- "loss": 1.2900440216064453,
298
- "mean_token_accuracy": 0.7276664525270462,
299
- "num_tokens": 704141.0,
300
- "step": 290
301
- },
302
- {
303
- "entropy": 1.272147350013256,
304
- "epoch": 0.48,
305
- "grad_norm": 1.609375,
306
- "learning_rate": 6.000000000000001e-07,
307
- "loss": 1.3638032913208007,
308
- "mean_token_accuracy": 0.7200124055147171,
309
- "num_tokens": 727493.0,
310
- "step": 300
311
  }
312
  ],
313
  "logging_steps": 10,
@@ -322,12 +72,12 @@
322
  "should_evaluate": false,
323
  "should_log": false,
324
  "should_save": true,
325
- "should_training_stop": true
326
  },
327
  "attributes": {}
328
  }
329
  },
330
- "total_flos": 8051407652167680.0,
331
  "train_batch_size": 2,
332
  "trial_name": null,
333
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 0.06666666666666667,
6
  "eval_steps": 500,
7
+ "global_step": 50,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
+ "entropy": 6.141439843177795,
14
+ "epoch": 0.013333333333333334,
15
+ "grad_norm": 5.21875,
16
+ "learning_rate": 0.00019955619646030802,
17
+ "loss": 6.9075462341308596,
18
+ "mean_token_accuracy": 0.10883389501832426,
19
+ "num_tokens": 25203.0,
20
  "step": 10
21
  },
22
  {
23
+ "entropy": 4.384208405017853,
24
+ "epoch": 0.02666666666666667,
25
+ "grad_norm": 4.46875,
26
+ "learning_rate": 0.00019802711746217218,
27
+ "loss": 4.253509140014648,
28
+ "mean_token_accuracy": 0.35271222703158855,
29
+ "num_tokens": 51133.0,
30
  "step": 20
31
  },
32
  {
33
+ "entropy": 3.155897554755211,
34
+ "epoch": 0.04,
35
+ "grad_norm": 2.375,
36
+ "learning_rate": 0.0001954240328516277,
37
+ "loss": 3.1908872604370115,
38
+ "mean_token_accuracy": 0.5070156037807465,
39
+ "num_tokens": 76227.0,
40
  "step": 30
41
  },
42
  {
43
+ "entropy": 2.6240377008914946,
44
+ "epoch": 0.05333333333333334,
45
+ "grad_norm": 2.609375,
46
+ "learning_rate": 0.00019177546256839812,
47
+ "loss": 2.829987716674805,
48
+ "mean_token_accuracy": 0.5530258767306805,
49
+ "num_tokens": 101311.0,
50
  "step": 40
51
  },
52
  {
53
+ "entropy": 2.5415185183286666,
54
+ "epoch": 0.06666666666666667,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
55
  "grad_norm": 2.40625,
56
+ "learning_rate": 0.00018712138111201895,
57
+ "loss": 2.619968223571777,
58
+ "mean_token_accuracy": 0.5668626762926579,
59
+ "num_tokens": 125668.0,
60
+ "step": 50
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
61
  }
62
  ],
63
  "logging_steps": 10,
 
72
  "should_evaluate": false,
73
  "should_log": false,
74
  "should_save": true,
75
+ "should_training_stop": false
76
  },
77
  "attributes": {}
78
  }
79
  },
80
+ "total_flos": 1371325075783680.0,
81
  "train_batch_size": 2,
82
  "trial_name": null,
83
  "trial_params": null
last-checkpoint/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:91a9d9e638a3272d95834e80ada6476d79b2f340c5de36a184447e88aa44de9c
3
  size 5649
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:53a49b0c5ca7d1384ef8f74aac0703b3aa931913c3690963fa6eedfdfd6edbcb
3
  size 5649