Joy28 commited on
Commit
2d5255a
·
1 Parent(s): 0279895

Training in progress, epoch 17

Browse files
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7fe3b3795e10d6eb10436995d24270126b9d7730bda3e8a1b6d43cc2a6f4fea4
3
  size 344946604
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86661308da7bada0d5063cca51b354f9c073640d921a7e7d0d64984a0b7ccef0
3
  size 344946604
tmp-checkpoint-2124/config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "MCG-NJU/videomae-base",
3
+ "architectures": [
4
+ "VideoMAEForVideoClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "decoder_hidden_size": 384,
8
+ "decoder_intermediate_size": 1536,
9
+ "decoder_num_attention_heads": 6,
10
+ "decoder_num_hidden_layers": 4,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.0,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "driving _reverse",
16
+ "1": "driving_reverse(others)",
17
+ "2": "object_falling",
18
+ "3": "pedestrian",
19
+ "4": "stop_vehicle"
20
+ },
21
+ "image_size": 224,
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 3072,
24
+ "label2id": {
25
+ "driving _reverse": 0,
26
+ "driving_reverse(others)": 1,
27
+ "object_falling": 2,
28
+ "pedestrian": 3,
29
+ "stop_vehicle": 4
30
+ },
31
+ "layer_norm_eps": 1e-12,
32
+ "model_type": "videomae",
33
+ "norm_pix_loss": true,
34
+ "num_attention_heads": 12,
35
+ "num_channels": 3,
36
+ "num_frames": 16,
37
+ "num_hidden_layers": 12,
38
+ "patch_size": 16,
39
+ "problem_type": "single_label_classification",
40
+ "qkv_bias": true,
41
+ "torch_dtype": "float32",
42
+ "transformers_version": "4.36.2",
43
+ "tubelet_size": 2,
44
+ "use_mean_pooling": false
45
+ }
tmp-checkpoint-2124/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:86661308da7bada0d5063cca51b354f9c073640d921a7e7d0d64984a0b7ccef0
3
+ size 344946604
tmp-checkpoint-2124/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6afec5b78c4d6c4ca3e2ae1b496f68f16ddfa91fc6b9ded842ad5624a49896e7
3
+ size 690003525
tmp-checkpoint-2124/preprocessor_config.json ADDED
@@ -0,0 +1,27 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": {
3
+ "height": 224,
4
+ "width": 224
5
+ },
6
+ "do_center_crop": true,
7
+ "do_normalize": true,
8
+ "do_rescale": true,
9
+ "do_resize": true,
10
+ "feature_extractor_type": "VideoMAEFeatureExtractor",
11
+ "image_mean": [
12
+ 0.485,
13
+ 0.456,
14
+ 0.406
15
+ ],
16
+ "image_processor_type": "VideoMAEImageProcessor",
17
+ "image_std": [
18
+ 0.229,
19
+ 0.224,
20
+ 0.225
21
+ ],
22
+ "resample": 2,
23
+ "rescale_factor": 0.00392156862745098,
24
+ "size": {
25
+ "shortest_edge": 224
26
+ }
27
+ }
tmp-checkpoint-2124/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2cde81082a354c50d40152410b7298115d178d82f9e447386ab757b99fe0a3f9
3
+ size 14511
tmp-checkpoint-2124/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cb5a4cc9e89e2be6fa962183ff247acf096452d8409a22d7a743ac005d45fc9
3
+ size 627
tmp-checkpoint-2124/trainer_state.json ADDED
@@ -0,0 +1,1455 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7373271889400922,
3
+ "best_model_checkpoint": "videomae-base-finetuned-subset\\checkpoint-1770",
4
+ "epoch": 17.016809116809117,
5
+ "eval_steps": 500,
6
+ "global_step": 2124,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 7.122507122507123e-07,
14
+ "loss": 1.6298,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.0,
19
+ "learning_rate": 1.4245014245014246e-06,
20
+ "loss": 1.6341,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.0,
25
+ "learning_rate": 2.136752136752137e-06,
26
+ "loss": 1.6059,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 2.8490028490028492e-06,
32
+ "loss": 1.6036,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.01,
37
+ "learning_rate": 3.5612535612535615e-06,
38
+ "loss": 1.6528,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.01,
43
+ "learning_rate": 4.273504273504274e-06,
44
+ "loss": 1.62,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.01,
49
+ "learning_rate": 4.985754985754986e-06,
50
+ "loss": 1.6109,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.01,
55
+ "learning_rate": 5.6980056980056985e-06,
56
+ "loss": 1.6268,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.01,
61
+ "learning_rate": 6.41025641025641e-06,
62
+ "loss": 1.6502,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.01,
67
+ "learning_rate": 7.122507122507123e-06,
68
+ "loss": 1.5551,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.02,
73
+ "learning_rate": 7.834757834757835e-06,
74
+ "loss": 1.6277,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.02,
79
+ "eval_accuracy": 0.08755760368663594,
80
+ "eval_loss": 1.579700231552124,
81
+ "eval_runtime": 293.8716,
82
+ "eval_samples_per_second": 0.738,
83
+ "eval_steps_per_second": 0.187,
84
+ "step": 118
85
+ },
86
+ {
87
+ "epoch": 1.0,
88
+ "learning_rate": 8.547008547008548e-06,
89
+ "loss": 1.6441,
90
+ "step": 120
91
+ },
92
+ {
93
+ "epoch": 1.0,
94
+ "learning_rate": 9.259259259259259e-06,
95
+ "loss": 1.6682,
96
+ "step": 130
97
+ },
98
+ {
99
+ "epoch": 1.0,
100
+ "learning_rate": 9.971509971509972e-06,
101
+ "loss": 1.5708,
102
+ "step": 140
103
+ },
104
+ {
105
+ "epoch": 1.0,
106
+ "learning_rate": 1.0683760683760684e-05,
107
+ "loss": 1.5522,
108
+ "step": 150
109
+ },
110
+ {
111
+ "epoch": 1.01,
112
+ "learning_rate": 1.1396011396011397e-05,
113
+ "loss": 1.5703,
114
+ "step": 160
115
+ },
116
+ {
117
+ "epoch": 1.01,
118
+ "learning_rate": 1.2108262108262108e-05,
119
+ "loss": 1.6537,
120
+ "step": 170
121
+ },
122
+ {
123
+ "epoch": 1.01,
124
+ "learning_rate": 1.282051282051282e-05,
125
+ "loss": 1.6126,
126
+ "step": 180
127
+ },
128
+ {
129
+ "epoch": 1.01,
130
+ "learning_rate": 1.3532763532763535e-05,
131
+ "loss": 1.6827,
132
+ "step": 190
133
+ },
134
+ {
135
+ "epoch": 1.01,
136
+ "learning_rate": 1.4245014245014246e-05,
137
+ "loss": 1.5872,
138
+ "step": 200
139
+ },
140
+ {
141
+ "epoch": 1.01,
142
+ "learning_rate": 1.4957264957264958e-05,
143
+ "loss": 1.5702,
144
+ "step": 210
145
+ },
146
+ {
147
+ "epoch": 1.01,
148
+ "learning_rate": 1.566951566951567e-05,
149
+ "loss": 1.5971,
150
+ "step": 220
151
+ },
152
+ {
153
+ "epoch": 1.02,
154
+ "learning_rate": 1.6381766381766382e-05,
155
+ "loss": 1.6542,
156
+ "step": 230
157
+ },
158
+ {
159
+ "epoch": 1.02,
160
+ "eval_accuracy": 0.1336405529953917,
161
+ "eval_loss": 1.654339075088501,
162
+ "eval_runtime": 264.9103,
163
+ "eval_samples_per_second": 0.819,
164
+ "eval_steps_per_second": 0.208,
165
+ "step": 236
166
+ },
167
+ {
168
+ "epoch": 2.0,
169
+ "learning_rate": 1.7094017094017095e-05,
170
+ "loss": 1.5755,
171
+ "step": 240
172
+ },
173
+ {
174
+ "epoch": 2.0,
175
+ "learning_rate": 1.7806267806267805e-05,
176
+ "loss": 1.6276,
177
+ "step": 250
178
+ },
179
+ {
180
+ "epoch": 2.0,
181
+ "learning_rate": 1.8518518518518518e-05,
182
+ "loss": 1.6158,
183
+ "step": 260
184
+ },
185
+ {
186
+ "epoch": 2.0,
187
+ "learning_rate": 1.923076923076923e-05,
188
+ "loss": 1.5842,
189
+ "step": 270
190
+ },
191
+ {
192
+ "epoch": 2.01,
193
+ "learning_rate": 1.9943019943019945e-05,
194
+ "loss": 1.5915,
195
+ "step": 280
196
+ },
197
+ {
198
+ "epoch": 2.01,
199
+ "learning_rate": 2.0655270655270654e-05,
200
+ "loss": 1.4263,
201
+ "step": 290
202
+ },
203
+ {
204
+ "epoch": 2.01,
205
+ "learning_rate": 2.1367521367521368e-05,
206
+ "loss": 1.4421,
207
+ "step": 300
208
+ },
209
+ {
210
+ "epoch": 2.01,
211
+ "learning_rate": 2.207977207977208e-05,
212
+ "loss": 1.2551,
213
+ "step": 310
214
+ },
215
+ {
216
+ "epoch": 2.01,
217
+ "learning_rate": 2.2792022792022794e-05,
218
+ "loss": 1.3329,
219
+ "step": 320
220
+ },
221
+ {
222
+ "epoch": 2.01,
223
+ "learning_rate": 2.3504273504273504e-05,
224
+ "loss": 1.1874,
225
+ "step": 330
226
+ },
227
+ {
228
+ "epoch": 2.01,
229
+ "learning_rate": 2.4216524216524217e-05,
230
+ "loss": 1.3141,
231
+ "step": 340
232
+ },
233
+ {
234
+ "epoch": 2.02,
235
+ "learning_rate": 2.492877492877493e-05,
236
+ "loss": 1.522,
237
+ "step": 350
238
+ },
239
+ {
240
+ "epoch": 2.02,
241
+ "eval_accuracy": 0.3271889400921659,
242
+ "eval_loss": 1.3541377782821655,
243
+ "eval_runtime": 259.8187,
244
+ "eval_samples_per_second": 0.835,
245
+ "eval_steps_per_second": 0.212,
246
+ "step": 354
247
+ },
248
+ {
249
+ "epoch": 3.0,
250
+ "learning_rate": 2.564102564102564e-05,
251
+ "loss": 1.3559,
252
+ "step": 360
253
+ },
254
+ {
255
+ "epoch": 3.0,
256
+ "learning_rate": 2.6353276353276356e-05,
257
+ "loss": 1.4339,
258
+ "step": 370
259
+ },
260
+ {
261
+ "epoch": 3.0,
262
+ "learning_rate": 2.706552706552707e-05,
263
+ "loss": 1.1016,
264
+ "step": 380
265
+ },
266
+ {
267
+ "epoch": 3.01,
268
+ "learning_rate": 2.777777777777778e-05,
269
+ "loss": 1.1426,
270
+ "step": 390
271
+ },
272
+ {
273
+ "epoch": 3.01,
274
+ "learning_rate": 2.8490028490028492e-05,
275
+ "loss": 1.5716,
276
+ "step": 400
277
+ },
278
+ {
279
+ "epoch": 3.01,
280
+ "learning_rate": 2.9202279202279202e-05,
281
+ "loss": 1.201,
282
+ "step": 410
283
+ },
284
+ {
285
+ "epoch": 3.01,
286
+ "learning_rate": 2.9914529914529915e-05,
287
+ "loss": 1.3309,
288
+ "step": 420
289
+ },
290
+ {
291
+ "epoch": 3.01,
292
+ "learning_rate": 3.0626780626780625e-05,
293
+ "loss": 1.3621,
294
+ "step": 430
295
+ },
296
+ {
297
+ "epoch": 3.01,
298
+ "learning_rate": 3.133903133903134e-05,
299
+ "loss": 1.3497,
300
+ "step": 440
301
+ },
302
+ {
303
+ "epoch": 3.01,
304
+ "learning_rate": 3.205128205128206e-05,
305
+ "loss": 1.3392,
306
+ "step": 450
307
+ },
308
+ {
309
+ "epoch": 3.02,
310
+ "learning_rate": 3.2763532763532764e-05,
311
+ "loss": 1.1704,
312
+ "step": 460
313
+ },
314
+ {
315
+ "epoch": 3.02,
316
+ "learning_rate": 3.347578347578348e-05,
317
+ "loss": 1.2632,
318
+ "step": 470
319
+ },
320
+ {
321
+ "epoch": 3.02,
322
+ "eval_accuracy": 0.3640552995391705,
323
+ "eval_loss": 1.2103053331375122,
324
+ "eval_runtime": 269.6766,
325
+ "eval_samples_per_second": 0.805,
326
+ "eval_steps_per_second": 0.204,
327
+ "step": 472
328
+ },
329
+ {
330
+ "epoch": 4.0,
331
+ "learning_rate": 3.418803418803419e-05,
332
+ "loss": 1.2155,
333
+ "step": 480
334
+ },
335
+ {
336
+ "epoch": 4.0,
337
+ "learning_rate": 3.4900284900284904e-05,
338
+ "loss": 1.1909,
339
+ "step": 490
340
+ },
341
+ {
342
+ "epoch": 4.0,
343
+ "learning_rate": 3.561253561253561e-05,
344
+ "loss": 1.355,
345
+ "step": 500
346
+ },
347
+ {
348
+ "epoch": 4.01,
349
+ "learning_rate": 3.6324786324786323e-05,
350
+ "loss": 1.4403,
351
+ "step": 510
352
+ },
353
+ {
354
+ "epoch": 4.01,
355
+ "learning_rate": 3.7037037037037037e-05,
356
+ "loss": 1.2344,
357
+ "step": 520
358
+ },
359
+ {
360
+ "epoch": 4.01,
361
+ "learning_rate": 3.774928774928775e-05,
362
+ "loss": 1.1599,
363
+ "step": 530
364
+ },
365
+ {
366
+ "epoch": 4.01,
367
+ "learning_rate": 3.846153846153846e-05,
368
+ "loss": 1.0906,
369
+ "step": 540
370
+ },
371
+ {
372
+ "epoch": 4.01,
373
+ "learning_rate": 3.9173789173789176e-05,
374
+ "loss": 1.4037,
375
+ "step": 550
376
+ },
377
+ {
378
+ "epoch": 4.01,
379
+ "learning_rate": 3.988603988603989e-05,
380
+ "loss": 1.015,
381
+ "step": 560
382
+ },
383
+ {
384
+ "epoch": 4.01,
385
+ "learning_rate": 4.05982905982906e-05,
386
+ "loss": 1.1146,
387
+ "step": 570
388
+ },
389
+ {
390
+ "epoch": 4.02,
391
+ "learning_rate": 4.131054131054131e-05,
392
+ "loss": 1.3183,
393
+ "step": 580
394
+ },
395
+ {
396
+ "epoch": 4.02,
397
+ "learning_rate": 4.202279202279202e-05,
398
+ "loss": 1.094,
399
+ "step": 590
400
+ },
401
+ {
402
+ "epoch": 4.02,
403
+ "eval_accuracy": 0.6682027649769585,
404
+ "eval_loss": 1.0165023803710938,
405
+ "eval_runtime": 285.7344,
406
+ "eval_samples_per_second": 0.759,
407
+ "eval_steps_per_second": 0.192,
408
+ "step": 590
409
+ },
410
+ {
411
+ "epoch": 5.0,
412
+ "learning_rate": 4.2735042735042735e-05,
413
+ "loss": 1.0792,
414
+ "step": 600
415
+ },
416
+ {
417
+ "epoch": 5.0,
418
+ "learning_rate": 4.344729344729345e-05,
419
+ "loss": 0.9429,
420
+ "step": 610
421
+ },
422
+ {
423
+ "epoch": 5.0,
424
+ "learning_rate": 4.415954415954416e-05,
425
+ "loss": 1.1575,
426
+ "step": 620
427
+ },
428
+ {
429
+ "epoch": 5.01,
430
+ "learning_rate": 4.4871794871794874e-05,
431
+ "loss": 1.03,
432
+ "step": 630
433
+ },
434
+ {
435
+ "epoch": 5.01,
436
+ "learning_rate": 4.558404558404559e-05,
437
+ "loss": 1.2668,
438
+ "step": 640
439
+ },
440
+ {
441
+ "epoch": 5.01,
442
+ "learning_rate": 4.62962962962963e-05,
443
+ "loss": 1.4121,
444
+ "step": 650
445
+ },
446
+ {
447
+ "epoch": 5.01,
448
+ "learning_rate": 4.700854700854701e-05,
449
+ "loss": 1.3624,
450
+ "step": 660
451
+ },
452
+ {
453
+ "epoch": 5.01,
454
+ "learning_rate": 4.772079772079772e-05,
455
+ "loss": 1.0479,
456
+ "step": 670
457
+ },
458
+ {
459
+ "epoch": 5.01,
460
+ "learning_rate": 4.8433048433048433e-05,
461
+ "loss": 1.3261,
462
+ "step": 680
463
+ },
464
+ {
465
+ "epoch": 5.01,
466
+ "learning_rate": 4.9145299145299147e-05,
467
+ "loss": 1.0643,
468
+ "step": 690
469
+ },
470
+ {
471
+ "epoch": 5.02,
472
+ "learning_rate": 4.985754985754986e-05,
473
+ "loss": 1.183,
474
+ "step": 700
475
+ },
476
+ {
477
+ "epoch": 5.02,
478
+ "eval_accuracy": 0.6082949308755761,
479
+ "eval_loss": 1.023329496383667,
480
+ "eval_runtime": 286.9179,
481
+ "eval_samples_per_second": 0.756,
482
+ "eval_steps_per_second": 0.192,
483
+ "step": 708
484
+ },
485
+ {
486
+ "epoch": 6.0,
487
+ "learning_rate": 4.993668882557772e-05,
488
+ "loss": 1.0286,
489
+ "step": 710
490
+ },
491
+ {
492
+ "epoch": 6.0,
493
+ "learning_rate": 4.985754985754986e-05,
494
+ "loss": 1.2985,
495
+ "step": 720
496
+ },
497
+ {
498
+ "epoch": 6.0,
499
+ "learning_rate": 4.9778410889522e-05,
500
+ "loss": 1.3161,
501
+ "step": 730
502
+ },
503
+ {
504
+ "epoch": 6.0,
505
+ "learning_rate": 4.9699271921494144e-05,
506
+ "loss": 1.2498,
507
+ "step": 740
508
+ },
509
+ {
510
+ "epoch": 6.01,
511
+ "learning_rate": 4.962013295346629e-05,
512
+ "loss": 1.0488,
513
+ "step": 750
514
+ },
515
+ {
516
+ "epoch": 6.01,
517
+ "learning_rate": 4.9540993985438435e-05,
518
+ "loss": 1.1916,
519
+ "step": 760
520
+ },
521
+ {
522
+ "epoch": 6.01,
523
+ "learning_rate": 4.946185501741058e-05,
524
+ "loss": 1.3807,
525
+ "step": 770
526
+ },
527
+ {
528
+ "epoch": 6.01,
529
+ "learning_rate": 4.938271604938271e-05,
530
+ "loss": 1.1852,
531
+ "step": 780
532
+ },
533
+ {
534
+ "epoch": 6.01,
535
+ "learning_rate": 4.930357708135486e-05,
536
+ "loss": 1.3454,
537
+ "step": 790
538
+ },
539
+ {
540
+ "epoch": 6.01,
541
+ "learning_rate": 4.9224438113327004e-05,
542
+ "loss": 1.1208,
543
+ "step": 800
544
+ },
545
+ {
546
+ "epoch": 6.01,
547
+ "learning_rate": 4.9145299145299147e-05,
548
+ "loss": 1.0341,
549
+ "step": 810
550
+ },
551
+ {
552
+ "epoch": 6.02,
553
+ "learning_rate": 4.906616017727129e-05,
554
+ "loss": 1.1543,
555
+ "step": 820
556
+ },
557
+ {
558
+ "epoch": 6.02,
559
+ "eval_accuracy": 0.5391705069124424,
560
+ "eval_loss": 1.108931064605713,
561
+ "eval_runtime": 273.7329,
562
+ "eval_samples_per_second": 0.793,
563
+ "eval_steps_per_second": 0.201,
564
+ "step": 826
565
+ },
566
+ {
567
+ "epoch": 7.0,
568
+ "learning_rate": 4.898702120924343e-05,
569
+ "loss": 1.1723,
570
+ "step": 830
571
+ },
572
+ {
573
+ "epoch": 7.0,
574
+ "learning_rate": 4.890788224121557e-05,
575
+ "loss": 1.0579,
576
+ "step": 840
577
+ },
578
+ {
579
+ "epoch": 7.0,
580
+ "learning_rate": 4.882874327318772e-05,
581
+ "loss": 0.9805,
582
+ "step": 850
583
+ },
584
+ {
585
+ "epoch": 7.0,
586
+ "learning_rate": 4.8749604305159865e-05,
587
+ "loss": 1.1434,
588
+ "step": 860
589
+ },
590
+ {
591
+ "epoch": 7.01,
592
+ "learning_rate": 4.867046533713201e-05,
593
+ "loss": 1.2956,
594
+ "step": 870
595
+ },
596
+ {
597
+ "epoch": 7.01,
598
+ "learning_rate": 4.859132636910415e-05,
599
+ "loss": 1.1865,
600
+ "step": 880
601
+ },
602
+ {
603
+ "epoch": 7.01,
604
+ "learning_rate": 4.851218740107629e-05,
605
+ "loss": 1.226,
606
+ "step": 890
607
+ },
608
+ {
609
+ "epoch": 7.01,
610
+ "learning_rate": 4.8433048433048433e-05,
611
+ "loss": 0.9409,
612
+ "step": 900
613
+ },
614
+ {
615
+ "epoch": 7.01,
616
+ "learning_rate": 4.835390946502058e-05,
617
+ "loss": 1.05,
618
+ "step": 910
619
+ },
620
+ {
621
+ "epoch": 7.01,
622
+ "learning_rate": 4.8274770496992725e-05,
623
+ "loss": 1.0455,
624
+ "step": 920
625
+ },
626
+ {
627
+ "epoch": 7.01,
628
+ "learning_rate": 4.819563152896486e-05,
629
+ "loss": 0.8294,
630
+ "step": 930
631
+ },
632
+ {
633
+ "epoch": 7.02,
634
+ "learning_rate": 4.811649256093701e-05,
635
+ "loss": 1.1015,
636
+ "step": 940
637
+ },
638
+ {
639
+ "epoch": 7.02,
640
+ "eval_accuracy": 0.45161290322580644,
641
+ "eval_loss": 1.171296238899231,
642
+ "eval_runtime": 263.2049,
643
+ "eval_samples_per_second": 0.824,
644
+ "eval_steps_per_second": 0.209,
645
+ "step": 944
646
+ },
647
+ {
648
+ "epoch": 8.0,
649
+ "learning_rate": 4.803735359290915e-05,
650
+ "loss": 0.7604,
651
+ "step": 950
652
+ },
653
+ {
654
+ "epoch": 8.0,
655
+ "learning_rate": 4.7958214624881294e-05,
656
+ "loss": 1.3654,
657
+ "step": 960
658
+ },
659
+ {
660
+ "epoch": 8.0,
661
+ "learning_rate": 4.787907565685344e-05,
662
+ "loss": 1.0182,
663
+ "step": 970
664
+ },
665
+ {
666
+ "epoch": 8.01,
667
+ "learning_rate": 4.779993668882558e-05,
668
+ "loss": 1.007,
669
+ "step": 980
670
+ },
671
+ {
672
+ "epoch": 8.01,
673
+ "learning_rate": 4.772079772079772e-05,
674
+ "loss": 0.9339,
675
+ "step": 990
676
+ },
677
+ {
678
+ "epoch": 8.01,
679
+ "learning_rate": 4.764165875276987e-05,
680
+ "loss": 0.8832,
681
+ "step": 1000
682
+ },
683
+ {
684
+ "epoch": 8.01,
685
+ "learning_rate": 4.756251978474201e-05,
686
+ "loss": 0.8878,
687
+ "step": 1010
688
+ },
689
+ {
690
+ "epoch": 8.01,
691
+ "learning_rate": 4.7483380816714154e-05,
692
+ "loss": 1.038,
693
+ "step": 1020
694
+ },
695
+ {
696
+ "epoch": 8.01,
697
+ "learning_rate": 4.7404241848686296e-05,
698
+ "loss": 1.0377,
699
+ "step": 1030
700
+ },
701
+ {
702
+ "epoch": 8.01,
703
+ "learning_rate": 4.732510288065844e-05,
704
+ "loss": 1.1283,
705
+ "step": 1040
706
+ },
707
+ {
708
+ "epoch": 8.02,
709
+ "learning_rate": 4.724596391263058e-05,
710
+ "loss": 0.845,
711
+ "step": 1050
712
+ },
713
+ {
714
+ "epoch": 8.02,
715
+ "learning_rate": 4.716682494460272e-05,
716
+ "loss": 1.012,
717
+ "step": 1060
718
+ },
719
+ {
720
+ "epoch": 8.02,
721
+ "eval_accuracy": 0.6175115207373272,
722
+ "eval_loss": 1.0352449417114258,
723
+ "eval_runtime": 279.9204,
724
+ "eval_samples_per_second": 0.775,
725
+ "eval_steps_per_second": 0.196,
726
+ "step": 1062
727
+ },
728
+ {
729
+ "epoch": 9.0,
730
+ "learning_rate": 4.708768597657487e-05,
731
+ "loss": 1.3229,
732
+ "step": 1070
733
+ },
734
+ {
735
+ "epoch": 9.0,
736
+ "learning_rate": 4.700854700854701e-05,
737
+ "loss": 1.2021,
738
+ "step": 1080
739
+ },
740
+ {
741
+ "epoch": 9.0,
742
+ "learning_rate": 4.692940804051915e-05,
743
+ "loss": 1.2585,
744
+ "step": 1090
745
+ },
746
+ {
747
+ "epoch": 9.01,
748
+ "learning_rate": 4.68502690724913e-05,
749
+ "loss": 1.0603,
750
+ "step": 1100
751
+ },
752
+ {
753
+ "epoch": 9.01,
754
+ "learning_rate": 4.677113010446344e-05,
755
+ "loss": 0.9872,
756
+ "step": 1110
757
+ },
758
+ {
759
+ "epoch": 9.01,
760
+ "learning_rate": 4.669199113643558e-05,
761
+ "loss": 1.0278,
762
+ "step": 1120
763
+ },
764
+ {
765
+ "epoch": 9.01,
766
+ "learning_rate": 4.6612852168407725e-05,
767
+ "loss": 0.8111,
768
+ "step": 1130
769
+ },
770
+ {
771
+ "epoch": 9.01,
772
+ "learning_rate": 4.653371320037987e-05,
773
+ "loss": 1.0985,
774
+ "step": 1140
775
+ },
776
+ {
777
+ "epoch": 9.01,
778
+ "learning_rate": 4.645457423235201e-05,
779
+ "loss": 0.6293,
780
+ "step": 1150
781
+ },
782
+ {
783
+ "epoch": 9.01,
784
+ "learning_rate": 4.637543526432416e-05,
785
+ "loss": 1.0005,
786
+ "step": 1160
787
+ },
788
+ {
789
+ "epoch": 9.02,
790
+ "learning_rate": 4.62962962962963e-05,
791
+ "loss": 1.0029,
792
+ "step": 1170
793
+ },
794
+ {
795
+ "epoch": 9.02,
796
+ "learning_rate": 4.621715732826844e-05,
797
+ "loss": 0.7159,
798
+ "step": 1180
799
+ },
800
+ {
801
+ "epoch": 9.02,
802
+ "eval_accuracy": 0.5622119815668203,
803
+ "eval_loss": 1.004241704940796,
804
+ "eval_runtime": 302.8171,
805
+ "eval_samples_per_second": 0.717,
806
+ "eval_steps_per_second": 0.182,
807
+ "step": 1180
808
+ },
809
+ {
810
+ "epoch": 10.0,
811
+ "learning_rate": 4.6138018360240585e-05,
812
+ "loss": 0.86,
813
+ "step": 1190
814
+ },
815
+ {
816
+ "epoch": 10.0,
817
+ "learning_rate": 4.605887939221273e-05,
818
+ "loss": 0.9204,
819
+ "step": 1200
820
+ },
821
+ {
822
+ "epoch": 10.0,
823
+ "learning_rate": 4.597974042418487e-05,
824
+ "loss": 0.9883,
825
+ "step": 1210
826
+ },
827
+ {
828
+ "epoch": 10.01,
829
+ "learning_rate": 4.590060145615702e-05,
830
+ "loss": 0.9755,
831
+ "step": 1220
832
+ },
833
+ {
834
+ "epoch": 10.01,
835
+ "learning_rate": 4.582146248812916e-05,
836
+ "loss": 0.8854,
837
+ "step": 1230
838
+ },
839
+ {
840
+ "epoch": 10.01,
841
+ "learning_rate": 4.5742323520101296e-05,
842
+ "loss": 1.1736,
843
+ "step": 1240
844
+ },
845
+ {
846
+ "epoch": 10.01,
847
+ "learning_rate": 4.5663184552073445e-05,
848
+ "loss": 0.8579,
849
+ "step": 1250
850
+ },
851
+ {
852
+ "epoch": 10.01,
853
+ "learning_rate": 4.558404558404559e-05,
854
+ "loss": 0.8537,
855
+ "step": 1260
856
+ },
857
+ {
858
+ "epoch": 10.01,
859
+ "learning_rate": 4.550490661601773e-05,
860
+ "loss": 0.8163,
861
+ "step": 1270
862
+ },
863
+ {
864
+ "epoch": 10.01,
865
+ "learning_rate": 4.542576764798987e-05,
866
+ "loss": 0.8938,
867
+ "step": 1280
868
+ },
869
+ {
870
+ "epoch": 10.02,
871
+ "learning_rate": 4.5346628679962014e-05,
872
+ "loss": 0.9417,
873
+ "step": 1290
874
+ },
875
+ {
876
+ "epoch": 10.02,
877
+ "eval_accuracy": 0.5023041474654378,
878
+ "eval_loss": 1.1916040182113647,
879
+ "eval_runtime": 290.3991,
880
+ "eval_samples_per_second": 0.747,
881
+ "eval_steps_per_second": 0.189,
882
+ "step": 1298
883
+ },
884
+ {
885
+ "epoch": 11.0,
886
+ "learning_rate": 4.5267489711934157e-05,
887
+ "loss": 0.7397,
888
+ "step": 1300
889
+ },
890
+ {
891
+ "epoch": 11.0,
892
+ "learning_rate": 4.51883507439063e-05,
893
+ "loss": 0.9699,
894
+ "step": 1310
895
+ },
896
+ {
897
+ "epoch": 11.0,
898
+ "learning_rate": 4.510921177587845e-05,
899
+ "loss": 1.0963,
900
+ "step": 1320
901
+ },
902
+ {
903
+ "epoch": 11.0,
904
+ "learning_rate": 4.503007280785059e-05,
905
+ "loss": 0.6996,
906
+ "step": 1330
907
+ },
908
+ {
909
+ "epoch": 11.01,
910
+ "learning_rate": 4.4950933839822725e-05,
911
+ "loss": 0.878,
912
+ "step": 1340
913
+ },
914
+ {
915
+ "epoch": 11.01,
916
+ "learning_rate": 4.4871794871794874e-05,
917
+ "loss": 1.1484,
918
+ "step": 1350
919
+ },
920
+ {
921
+ "epoch": 11.01,
922
+ "learning_rate": 4.479265590376702e-05,
923
+ "loss": 1.0544,
924
+ "step": 1360
925
+ },
926
+ {
927
+ "epoch": 11.01,
928
+ "learning_rate": 4.471351693573916e-05,
929
+ "loss": 0.9162,
930
+ "step": 1370
931
+ },
932
+ {
933
+ "epoch": 11.01,
934
+ "learning_rate": 4.463437796771131e-05,
935
+ "loss": 0.8283,
936
+ "step": 1380
937
+ },
938
+ {
939
+ "epoch": 11.01,
940
+ "learning_rate": 4.455523899968344e-05,
941
+ "loss": 1.0451,
942
+ "step": 1390
943
+ },
944
+ {
945
+ "epoch": 11.01,
946
+ "learning_rate": 4.4476100031655586e-05,
947
+ "loss": 1.1554,
948
+ "step": 1400
949
+ },
950
+ {
951
+ "epoch": 11.02,
952
+ "learning_rate": 4.4396961063627735e-05,
953
+ "loss": 1.0331,
954
+ "step": 1410
955
+ },
956
+ {
957
+ "epoch": 11.02,
958
+ "eval_accuracy": 0.6359447004608295,
959
+ "eval_loss": 0.9247654676437378,
960
+ "eval_runtime": 279.9164,
961
+ "eval_samples_per_second": 0.775,
962
+ "eval_steps_per_second": 0.196,
963
+ "step": 1416
964
+ },
965
+ {
966
+ "epoch": 12.0,
967
+ "learning_rate": 4.431782209559988e-05,
968
+ "loss": 0.7141,
969
+ "step": 1420
970
+ },
971
+ {
972
+ "epoch": 12.0,
973
+ "learning_rate": 4.423868312757202e-05,
974
+ "loss": 0.6275,
975
+ "step": 1430
976
+ },
977
+ {
978
+ "epoch": 12.0,
979
+ "learning_rate": 4.415954415954416e-05,
980
+ "loss": 0.8072,
981
+ "step": 1440
982
+ },
983
+ {
984
+ "epoch": 12.0,
985
+ "learning_rate": 4.4080405191516304e-05,
986
+ "loss": 0.9574,
987
+ "step": 1450
988
+ },
989
+ {
990
+ "epoch": 12.01,
991
+ "learning_rate": 4.4001266223488446e-05,
992
+ "loss": 1.0716,
993
+ "step": 1460
994
+ },
995
+ {
996
+ "epoch": 12.01,
997
+ "learning_rate": 4.3922127255460595e-05,
998
+ "loss": 1.2938,
999
+ "step": 1470
1000
+ },
1001
+ {
1002
+ "epoch": 12.01,
1003
+ "learning_rate": 4.384298828743274e-05,
1004
+ "loss": 1.0527,
1005
+ "step": 1480
1006
+ },
1007
+ {
1008
+ "epoch": 12.01,
1009
+ "learning_rate": 4.376384931940488e-05,
1010
+ "loss": 0.9223,
1011
+ "step": 1490
1012
+ },
1013
+ {
1014
+ "epoch": 12.01,
1015
+ "learning_rate": 4.368471035137702e-05,
1016
+ "loss": 0.9105,
1017
+ "step": 1500
1018
+ },
1019
+ {
1020
+ "epoch": 12.01,
1021
+ "learning_rate": 4.3605571383349164e-05,
1022
+ "loss": 0.8926,
1023
+ "step": 1510
1024
+ },
1025
+ {
1026
+ "epoch": 12.01,
1027
+ "learning_rate": 4.3526432415321306e-05,
1028
+ "loss": 0.7864,
1029
+ "step": 1520
1030
+ },
1031
+ {
1032
+ "epoch": 12.02,
1033
+ "learning_rate": 4.344729344729345e-05,
1034
+ "loss": 1.208,
1035
+ "step": 1530
1036
+ },
1037
+ {
1038
+ "epoch": 12.02,
1039
+ "eval_accuracy": 0.7096774193548387,
1040
+ "eval_loss": 0.7162204384803772,
1041
+ "eval_runtime": 253.0914,
1042
+ "eval_samples_per_second": 0.857,
1043
+ "eval_steps_per_second": 0.217,
1044
+ "step": 1534
1045
+ },
1046
+ {
1047
+ "epoch": 13.0,
1048
+ "learning_rate": 4.336815447926559e-05,
1049
+ "loss": 0.5751,
1050
+ "step": 1540
1051
+ },
1052
+ {
1053
+ "epoch": 13.0,
1054
+ "learning_rate": 4.328901551123773e-05,
1055
+ "loss": 0.892,
1056
+ "step": 1550
1057
+ },
1058
+ {
1059
+ "epoch": 13.0,
1060
+ "learning_rate": 4.3209876543209875e-05,
1061
+ "loss": 0.8127,
1062
+ "step": 1560
1063
+ },
1064
+ {
1065
+ "epoch": 13.01,
1066
+ "learning_rate": 4.3130737575182024e-05,
1067
+ "loss": 0.7734,
1068
+ "step": 1570
1069
+ },
1070
+ {
1071
+ "epoch": 13.01,
1072
+ "learning_rate": 4.3051598607154166e-05,
1073
+ "loss": 0.6491,
1074
+ "step": 1580
1075
+ },
1076
+ {
1077
+ "epoch": 13.01,
1078
+ "learning_rate": 4.297245963912631e-05,
1079
+ "loss": 1.2019,
1080
+ "step": 1590
1081
+ },
1082
+ {
1083
+ "epoch": 13.01,
1084
+ "learning_rate": 4.289332067109845e-05,
1085
+ "loss": 0.8835,
1086
+ "step": 1600
1087
+ },
1088
+ {
1089
+ "epoch": 13.01,
1090
+ "learning_rate": 4.281418170307059e-05,
1091
+ "loss": 1.1494,
1092
+ "step": 1610
1093
+ },
1094
+ {
1095
+ "epoch": 13.01,
1096
+ "learning_rate": 4.2735042735042735e-05,
1097
+ "loss": 0.8985,
1098
+ "step": 1620
1099
+ },
1100
+ {
1101
+ "epoch": 13.01,
1102
+ "learning_rate": 4.2655903767014884e-05,
1103
+ "loss": 0.6637,
1104
+ "step": 1630
1105
+ },
1106
+ {
1107
+ "epoch": 13.02,
1108
+ "learning_rate": 4.2576764798987026e-05,
1109
+ "loss": 0.9194,
1110
+ "step": 1640
1111
+ },
1112
+ {
1113
+ "epoch": 13.02,
1114
+ "learning_rate": 4.249762583095916e-05,
1115
+ "loss": 1.0481,
1116
+ "step": 1650
1117
+ },
1118
+ {
1119
+ "epoch": 13.02,
1120
+ "eval_accuracy": 0.6129032258064516,
1121
+ "eval_loss": 1.2836726903915405,
1122
+ "eval_runtime": 266.9839,
1123
+ "eval_samples_per_second": 0.813,
1124
+ "eval_steps_per_second": 0.206,
1125
+ "step": 1652
1126
+ },
1127
+ {
1128
+ "epoch": 14.0,
1129
+ "learning_rate": 4.241848686293131e-05,
1130
+ "loss": 0.8832,
1131
+ "step": 1660
1132
+ },
1133
+ {
1134
+ "epoch": 14.0,
1135
+ "learning_rate": 4.233934789490345e-05,
1136
+ "loss": 0.5931,
1137
+ "step": 1670
1138
+ },
1139
+ {
1140
+ "epoch": 14.0,
1141
+ "learning_rate": 4.2260208926875595e-05,
1142
+ "loss": 0.7721,
1143
+ "step": 1680
1144
+ },
1145
+ {
1146
+ "epoch": 14.01,
1147
+ "learning_rate": 4.2181069958847744e-05,
1148
+ "loss": 0.8753,
1149
+ "step": 1690
1150
+ },
1151
+ {
1152
+ "epoch": 14.01,
1153
+ "learning_rate": 4.210193099081988e-05,
1154
+ "loss": 0.9211,
1155
+ "step": 1700
1156
+ },
1157
+ {
1158
+ "epoch": 14.01,
1159
+ "learning_rate": 4.202279202279202e-05,
1160
+ "loss": 0.8866,
1161
+ "step": 1710
1162
+ },
1163
+ {
1164
+ "epoch": 14.01,
1165
+ "learning_rate": 4.194365305476417e-05,
1166
+ "loss": 0.9061,
1167
+ "step": 1720
1168
+ },
1169
+ {
1170
+ "epoch": 14.01,
1171
+ "learning_rate": 4.186451408673631e-05,
1172
+ "loss": 0.8832,
1173
+ "step": 1730
1174
+ },
1175
+ {
1176
+ "epoch": 14.01,
1177
+ "learning_rate": 4.1785375118708455e-05,
1178
+ "loss": 0.6224,
1179
+ "step": 1740
1180
+ },
1181
+ {
1182
+ "epoch": 14.01,
1183
+ "learning_rate": 4.17062361506806e-05,
1184
+ "loss": 0.752,
1185
+ "step": 1750
1186
+ },
1187
+ {
1188
+ "epoch": 14.02,
1189
+ "learning_rate": 4.162709718265274e-05,
1190
+ "loss": 1.1106,
1191
+ "step": 1760
1192
+ },
1193
+ {
1194
+ "epoch": 14.02,
1195
+ "learning_rate": 4.154795821462488e-05,
1196
+ "loss": 0.7895,
1197
+ "step": 1770
1198
+ },
1199
+ {
1200
+ "epoch": 14.02,
1201
+ "eval_accuracy": 0.7373271889400922,
1202
+ "eval_loss": 0.7213783264160156,
1203
+ "eval_runtime": 302.9001,
1204
+ "eval_samples_per_second": 0.716,
1205
+ "eval_steps_per_second": 0.182,
1206
+ "step": 1770
1207
+ },
1208
+ {
1209
+ "epoch": 15.0,
1210
+ "learning_rate": 4.1468819246597024e-05,
1211
+ "loss": 0.5983,
1212
+ "step": 1780
1213
+ },
1214
+ {
1215
+ "epoch": 15.0,
1216
+ "learning_rate": 4.138968027856917e-05,
1217
+ "loss": 0.7442,
1218
+ "step": 1790
1219
+ },
1220
+ {
1221
+ "epoch": 15.0,
1222
+ "learning_rate": 4.131054131054131e-05,
1223
+ "loss": 0.8338,
1224
+ "step": 1800
1225
+ },
1226
+ {
1227
+ "epoch": 15.01,
1228
+ "learning_rate": 4.123140234251345e-05,
1229
+ "loss": 1.0226,
1230
+ "step": 1810
1231
+ },
1232
+ {
1233
+ "epoch": 15.01,
1234
+ "learning_rate": 4.11522633744856e-05,
1235
+ "loss": 0.8336,
1236
+ "step": 1820
1237
+ },
1238
+ {
1239
+ "epoch": 15.01,
1240
+ "learning_rate": 4.107312440645774e-05,
1241
+ "loss": 0.6392,
1242
+ "step": 1830
1243
+ },
1244
+ {
1245
+ "epoch": 15.01,
1246
+ "learning_rate": 4.0993985438429884e-05,
1247
+ "loss": 0.8357,
1248
+ "step": 1840
1249
+ },
1250
+ {
1251
+ "epoch": 15.01,
1252
+ "learning_rate": 4.091484647040203e-05,
1253
+ "loss": 0.9604,
1254
+ "step": 1850
1255
+ },
1256
+ {
1257
+ "epoch": 15.01,
1258
+ "learning_rate": 4.083570750237417e-05,
1259
+ "loss": 0.7828,
1260
+ "step": 1860
1261
+ },
1262
+ {
1263
+ "epoch": 15.01,
1264
+ "learning_rate": 4.075656853434631e-05,
1265
+ "loss": 0.8063,
1266
+ "step": 1870
1267
+ },
1268
+ {
1269
+ "epoch": 15.02,
1270
+ "learning_rate": 4.067742956631846e-05,
1271
+ "loss": 0.9859,
1272
+ "step": 1880
1273
+ },
1274
+ {
1275
+ "epoch": 15.02,
1276
+ "eval_accuracy": 0.6866359447004609,
1277
+ "eval_loss": 0.8335953950881958,
1278
+ "eval_runtime": 299.0538,
1279
+ "eval_samples_per_second": 0.726,
1280
+ "eval_steps_per_second": 0.184,
1281
+ "step": 1888
1282
+ },
1283
+ {
1284
+ "epoch": 16.0,
1285
+ "learning_rate": 4.05982905982906e-05,
1286
+ "loss": 0.672,
1287
+ "step": 1890
1288
+ },
1289
+ {
1290
+ "epoch": 16.0,
1291
+ "learning_rate": 4.0519151630262745e-05,
1292
+ "loss": 0.8109,
1293
+ "step": 1900
1294
+ },
1295
+ {
1296
+ "epoch": 16.0,
1297
+ "learning_rate": 4.044001266223489e-05,
1298
+ "loss": 0.6525,
1299
+ "step": 1910
1300
+ },
1301
+ {
1302
+ "epoch": 16.0,
1303
+ "learning_rate": 4.036087369420703e-05,
1304
+ "loss": 0.7089,
1305
+ "step": 1920
1306
+ },
1307
+ {
1308
+ "epoch": 16.01,
1309
+ "learning_rate": 4.028173472617917e-05,
1310
+ "loss": 1.0346,
1311
+ "step": 1930
1312
+ },
1313
+ {
1314
+ "epoch": 16.01,
1315
+ "learning_rate": 4.020259575815132e-05,
1316
+ "loss": 1.1721,
1317
+ "step": 1940
1318
+ },
1319
+ {
1320
+ "epoch": 16.01,
1321
+ "learning_rate": 4.012345679012346e-05,
1322
+ "loss": 0.6268,
1323
+ "step": 1950
1324
+ },
1325
+ {
1326
+ "epoch": 16.01,
1327
+ "learning_rate": 4.00443178220956e-05,
1328
+ "loss": 0.8538,
1329
+ "step": 1960
1330
+ },
1331
+ {
1332
+ "epoch": 16.01,
1333
+ "learning_rate": 3.996517885406775e-05,
1334
+ "loss": 1.0732,
1335
+ "step": 1970
1336
+ },
1337
+ {
1338
+ "epoch": 16.01,
1339
+ "learning_rate": 3.988603988603989e-05,
1340
+ "loss": 0.8913,
1341
+ "step": 1980
1342
+ },
1343
+ {
1344
+ "epoch": 16.01,
1345
+ "learning_rate": 3.980690091801203e-05,
1346
+ "loss": 0.6379,
1347
+ "step": 1990
1348
+ },
1349
+ {
1350
+ "epoch": 16.02,
1351
+ "learning_rate": 3.9727761949984174e-05,
1352
+ "loss": 0.6517,
1353
+ "step": 2000
1354
+ },
1355
+ {
1356
+ "epoch": 16.02,
1357
+ "eval_accuracy": 0.6405529953917051,
1358
+ "eval_loss": 0.8792047500610352,
1359
+ "eval_runtime": 289.7798,
1360
+ "eval_samples_per_second": 0.749,
1361
+ "eval_steps_per_second": 0.19,
1362
+ "step": 2006
1363
+ },
1364
+ {
1365
+ "epoch": 17.0,
1366
+ "learning_rate": 3.9648622981956316e-05,
1367
+ "loss": 0.5995,
1368
+ "step": 2010
1369
+ },
1370
+ {
1371
+ "epoch": 17.0,
1372
+ "learning_rate": 3.956948401392846e-05,
1373
+ "loss": 1.0005,
1374
+ "step": 2020
1375
+ },
1376
+ {
1377
+ "epoch": 17.0,
1378
+ "learning_rate": 3.94903450459006e-05,
1379
+ "loss": 0.7818,
1380
+ "step": 2030
1381
+ },
1382
+ {
1383
+ "epoch": 17.0,
1384
+ "learning_rate": 3.941120607787275e-05,
1385
+ "loss": 0.7927,
1386
+ "step": 2040
1387
+ },
1388
+ {
1389
+ "epoch": 17.01,
1390
+ "learning_rate": 3.933206710984489e-05,
1391
+ "loss": 0.751,
1392
+ "step": 2050
1393
+ },
1394
+ {
1395
+ "epoch": 17.01,
1396
+ "learning_rate": 3.925292814181703e-05,
1397
+ "loss": 1.001,
1398
+ "step": 2060
1399
+ },
1400
+ {
1401
+ "epoch": 17.01,
1402
+ "learning_rate": 3.9173789173789176e-05,
1403
+ "loss": 0.8409,
1404
+ "step": 2070
1405
+ },
1406
+ {
1407
+ "epoch": 17.01,
1408
+ "learning_rate": 3.909465020576132e-05,
1409
+ "loss": 0.6746,
1410
+ "step": 2080
1411
+ },
1412
+ {
1413
+ "epoch": 17.01,
1414
+ "learning_rate": 3.901551123773346e-05,
1415
+ "loss": 0.6879,
1416
+ "step": 2090
1417
+ },
1418
+ {
1419
+ "epoch": 17.01,
1420
+ "learning_rate": 3.893637226970561e-05,
1421
+ "loss": 0.6906,
1422
+ "step": 2100
1423
+ },
1424
+ {
1425
+ "epoch": 17.01,
1426
+ "learning_rate": 3.8857233301677745e-05,
1427
+ "loss": 0.9602,
1428
+ "step": 2110
1429
+ },
1430
+ {
1431
+ "epoch": 17.02,
1432
+ "learning_rate": 3.877809433364989e-05,
1433
+ "loss": 0.7542,
1434
+ "step": 2120
1435
+ },
1436
+ {
1437
+ "epoch": 17.02,
1438
+ "eval_accuracy": 0.631336405529954,
1439
+ "eval_loss": 1.075119137763977,
1440
+ "eval_runtime": 259.3934,
1441
+ "eval_samples_per_second": 0.837,
1442
+ "eval_steps_per_second": 0.212,
1443
+ "step": 2124
1444
+ }
1445
+ ],
1446
+ "logging_steps": 10,
1447
+ "max_steps": 7020,
1448
+ "num_input_tokens_seen": 0,
1449
+ "num_train_epochs": 9223372036854775807,
1450
+ "save_steps": 500,
1451
+ "total_flos": 1.051955870149804e+19,
1452
+ "train_batch_size": 4,
1453
+ "trial_name": null,
1454
+ "trial_params": null
1455
+ }
tmp-checkpoint-2124/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eeef339b6eb8462b14393409df1be12ab41b1a33c83fbb4135a13f4ce965bd02
3
+ size 4283