desarrolloasesoreslocales commited on
Commit
6c9aff5
·
verified ·
1 Parent(s): 9030bb5

Model save

Browse files
README.md CHANGED
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
33
 
34
  This model is a fine-tuned version of [therealcyberlord/stanford-car-vit-patch16](https://huggingface.co/therealcyberlord/stanford-car-vit-patch16) on the imagefolder dataset.
35
  It achieves the following results on the evaluation set:
36
- - Loss: 1.4404
37
  - Accuracy: 0.5412
 
38
 
39
  ## Model description
40
 
@@ -54,11 +54,11 @@ More information needed
54
 
55
  The following hyperparameters were used during training:
56
  - learning_rate: 0.001
57
- - train_batch_size: 256
58
- - eval_batch_size: 256
59
  - seed: 42
60
  - gradient_accumulation_steps: 4
61
- - total_train_batch_size: 1024
62
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
63
  - lr_scheduler_type: linear
64
  - lr_scheduler_warmup_ratio: 0.1
@@ -66,18 +66,18 @@ The following hyperparameters were used during training:
66
 
67
  ### Training results
68
 
69
- | Training Loss | Epoch | Step | Validation Loss | Accuracy |
70
- |:-------------:|:------:|:----:|:---------------:|:--------:|
71
- | 16.0903 | 0.9967 | 226 | 3.4455 | 0.1969 |
72
- | 11.5438 | 1.9967 | 452 | 2.4033 | 0.2992 |
73
- | 9.5963 | 2.9967 | 678 | 2.0761 | 0.3492 |
74
- | 8.3873 | 3.9967 | 904 | 1.8921 | 0.3934 |
75
- | 7.3127 | 4.9967 | 1130 | 1.7535 | 0.4288 |
76
- | 6.2178 | 5.9967 | 1356 | 1.6533 | 0.4609 |
77
- | 5.4619 | 6.9967 | 1582 | 1.5859 | 0.4856 |
78
- | 4.618 | 7.9967 | 1808 | 1.5253 | 0.5129 |
79
- | 3.9349 | 8.9967 | 2034 | 1.4693 | 0.5315 |
80
- | 3.4667 | 9.9967 | 2260 | 1.4404 | 0.5412 |
81
 
82
 
83
  ### Framework versions
 
33
 
34
  This model is a fine-tuned version of [therealcyberlord/stanford-car-vit-patch16](https://huggingface.co/therealcyberlord/stanford-car-vit-patch16) on the imagefolder dataset.
35
  It achieves the following results on the evaluation set:
 
36
  - Accuracy: 0.5412
37
+ - Loss: 1.4404
38
 
39
  ## Model description
40
 
 
54
 
55
  The following hyperparameters were used during training:
56
  - learning_rate: 0.001
57
+ - train_batch_size: 128
58
+ - eval_batch_size: 128
59
  - seed: 42
60
  - gradient_accumulation_steps: 4
61
+ - total_train_batch_size: 512
62
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
63
  - lr_scheduler_type: linear
64
  - lr_scheduler_warmup_ratio: 0.1
 
66
 
67
  ### Training results
68
 
69
+ | Training Loss | Epoch | Step | Accuracy | Validation Loss |
70
+ |:-------------:|:------:|:----:|:--------:|:---------------:|
71
+ | 16.0903 | 0.9967 | 226 | 0.1969 | 3.4455 |
72
+ | 11.5438 | 1.9967 | 452 | 0.2992 | 2.4033 |
73
+ | 9.5963 | 2.9967 | 678 | 0.3492 | 2.0761 |
74
+ | 8.3873 | 3.9967 | 904 | 0.3934 | 1.8921 |
75
+ | 7.3127 | 4.9967 | 1130 | 0.4288 | 1.7535 |
76
+ | 6.2178 | 5.9967 | 1356 | 0.4609 | 1.6533 |
77
+ | 5.4619 | 6.9967 | 1582 | 0.4856 | 1.5859 |
78
+ | 4.618 | 7.9967 | 1808 | 0.5129 | 1.5253 |
79
+ | 3.9349 | 8.9967 | 2034 | 0.5315 | 1.4693 |
80
+ | 3.4667 | 9.9967 | 2260 | 0.5412 | 1.4404 |
81
 
82
 
83
  ### Framework versions
all_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.996692392502757,
3
+ "total_flos": 1.8565430446260014e+20,
4
+ "train_loss": 8.705859319298668,
5
+ "train_runtime": 25889.9118,
6
+ "train_samples_per_second": 89.633,
7
+ "train_steps_per_second": 0.087
8
+ }
runs/Jan29_08-40-35_d0e85ba8394a/events.out.tfevents.1738140041.d0e85ba8394a.21250.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ebc231b261860717c5deddb997d56642121b05af893772bed16490b126ee856f
3
+ size 251249
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 9.996692392502757,
3
+ "total_flos": 1.8565430446260014e+20,
4
+ "train_loss": 8.705859319298668,
5
+ "train_runtime": 25889.9118,
6
+ "train_samples_per_second": 89.633,
7
+ "train_steps_per_second": 0.087
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1714 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.541167345355827,
3
+ "best_model_checkpoint": "/content/drive/MyDrive/Inventariado_Mobiliario_Urbano/cars_brand_model/cmmy/checkpoint-2260",
4
+ "epoch": 9.996692392502757,
5
+ "eval_steps": 500,
6
+ "global_step": 2260,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.044101433296582136,
13
+ "grad_norm": 4.8150153160095215,
14
+ "learning_rate": 4.424778761061947e-05,
15
+ "loss": 33.4771,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.08820286659316427,
20
+ "grad_norm": 4.733597278594971,
21
+ "learning_rate": 8.849557522123894e-05,
22
+ "loss": 33.1339,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.13230429988974643,
27
+ "grad_norm": 4.442318439483643,
28
+ "learning_rate": 0.00013274336283185842,
29
+ "loss": 32.4884,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.17640573318632854,
34
+ "grad_norm": 4.178449630737305,
35
+ "learning_rate": 0.00017699115044247788,
36
+ "loss": 31.6444,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.2205071664829107,
41
+ "grad_norm": 4.139791965484619,
42
+ "learning_rate": 0.00022123893805309737,
43
+ "loss": 30.7515,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.26460859977949286,
48
+ "grad_norm": 4.265659332275391,
49
+ "learning_rate": 0.00026548672566371683,
50
+ "loss": 29.8171,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 0.308710033076075,
55
+ "grad_norm": 4.5766921043396,
56
+ "learning_rate": 0.00030973451327433627,
57
+ "loss": 28.6353,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 0.3528114663726571,
62
+ "grad_norm": 5.000542640686035,
63
+ "learning_rate": 0.00035398230088495576,
64
+ "loss": 27.6647,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 0.39691289966923926,
69
+ "grad_norm": 4.588747501373291,
70
+ "learning_rate": 0.00039823008849557525,
71
+ "loss": 26.439,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 0.4410143329658214,
76
+ "grad_norm": 5.457170486450195,
77
+ "learning_rate": 0.00044247787610619474,
78
+ "loss": 25.3025,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 0.48511576626240355,
83
+ "grad_norm": 5.326786518096924,
84
+ "learning_rate": 0.0004867256637168142,
85
+ "loss": 24.4399,
86
+ "step": 110
87
+ },
88
+ {
89
+ "epoch": 0.5292171995589857,
90
+ "grad_norm": 5.408795356750488,
91
+ "learning_rate": 0.0005309734513274337,
92
+ "loss": 23.267,
93
+ "step": 120
94
+ },
95
+ {
96
+ "epoch": 0.5733186328555678,
97
+ "grad_norm": 5.888697624206543,
98
+ "learning_rate": 0.0005752212389380532,
99
+ "loss": 22.239,
100
+ "step": 130
101
+ },
102
+ {
103
+ "epoch": 0.61742006615215,
104
+ "grad_norm": 6.699647903442383,
105
+ "learning_rate": 0.0006194690265486725,
106
+ "loss": 21.443,
107
+ "step": 140
108
+ },
109
+ {
110
+ "epoch": 0.6615214994487321,
111
+ "grad_norm": 5.940579891204834,
112
+ "learning_rate": 0.0006637168141592921,
113
+ "loss": 20.2496,
114
+ "step": 150
115
+ },
116
+ {
117
+ "epoch": 0.7056229327453142,
118
+ "grad_norm": 6.729783058166504,
119
+ "learning_rate": 0.0007079646017699115,
120
+ "loss": 19.7215,
121
+ "step": 160
122
+ },
123
+ {
124
+ "epoch": 0.7497243660418964,
125
+ "grad_norm": 7.232001781463623,
126
+ "learning_rate": 0.0007522123893805309,
127
+ "loss": 18.8327,
128
+ "step": 170
129
+ },
130
+ {
131
+ "epoch": 0.7938257993384785,
132
+ "grad_norm": 6.139113426208496,
133
+ "learning_rate": 0.0007964601769911505,
134
+ "loss": 18.3153,
135
+ "step": 180
136
+ },
137
+ {
138
+ "epoch": 0.8379272326350606,
139
+ "grad_norm": 6.233616828918457,
140
+ "learning_rate": 0.0008407079646017699,
141
+ "loss": 17.7032,
142
+ "step": 190
143
+ },
144
+ {
145
+ "epoch": 0.8820286659316428,
146
+ "grad_norm": 6.897170543670654,
147
+ "learning_rate": 0.0008849557522123895,
148
+ "loss": 16.9044,
149
+ "step": 200
150
+ },
151
+ {
152
+ "epoch": 0.9261300992282249,
153
+ "grad_norm": 6.186814785003662,
154
+ "learning_rate": 0.0009292035398230089,
155
+ "loss": 16.6313,
156
+ "step": 210
157
+ },
158
+ {
159
+ "epoch": 0.9702315325248071,
160
+ "grad_norm": 6.237786293029785,
161
+ "learning_rate": 0.0009734513274336283,
162
+ "loss": 16.0903,
163
+ "step": 220
164
+ },
165
+ {
166
+ "epoch": 0.9966923925027563,
167
+ "eval_accuracy": 0.19693620321892574,
168
+ "eval_loss": 3.4455316066741943,
169
+ "eval_runtime": 157.5981,
170
+ "eval_samples_per_second": 163.612,
171
+ "eval_steps_per_second": 0.641,
172
+ "step": 226
173
+ },
174
+ {
175
+ "epoch": 1.017640573318633,
176
+ "grad_norm": 5.7300639152526855,
177
+ "learning_rate": 0.0009980334316617503,
178
+ "loss": 16.2564,
179
+ "step": 230
180
+ },
181
+ {
182
+ "epoch": 1.061742006615215,
183
+ "grad_norm": 6.365210056304932,
184
+ "learning_rate": 0.000993117010816126,
185
+ "loss": 14.7841,
186
+ "step": 240
187
+ },
188
+ {
189
+ "epoch": 1.1058434399117971,
190
+ "grad_norm": 6.162064552307129,
191
+ "learning_rate": 0.0009882005899705015,
192
+ "loss": 14.7691,
193
+ "step": 250
194
+ },
195
+ {
196
+ "epoch": 1.1499448732083792,
197
+ "grad_norm": 6.440285682678223,
198
+ "learning_rate": 0.0009832841691248771,
199
+ "loss": 14.1738,
200
+ "step": 260
201
+ },
202
+ {
203
+ "epoch": 1.1940463065049614,
204
+ "grad_norm": 5.520109176635742,
205
+ "learning_rate": 0.0009783677482792527,
206
+ "loss": 14.0754,
207
+ "step": 270
208
+ },
209
+ {
210
+ "epoch": 1.2381477398015435,
211
+ "grad_norm": 6.146816730499268,
212
+ "learning_rate": 0.0009734513274336283,
213
+ "loss": 13.9686,
214
+ "step": 280
215
+ },
216
+ {
217
+ "epoch": 1.2822491730981258,
218
+ "grad_norm": 5.696677207946777,
219
+ "learning_rate": 0.000968534906588004,
220
+ "loss": 13.6213,
221
+ "step": 290
222
+ },
223
+ {
224
+ "epoch": 1.326350606394708,
225
+ "grad_norm": 5.706403732299805,
226
+ "learning_rate": 0.0009636184857423795,
227
+ "loss": 13.2106,
228
+ "step": 300
229
+ },
230
+ {
231
+ "epoch": 1.37045203969129,
232
+ "grad_norm": 6.3477067947387695,
233
+ "learning_rate": 0.0009587020648967551,
234
+ "loss": 13.1036,
235
+ "step": 310
236
+ },
237
+ {
238
+ "epoch": 1.4145534729878722,
239
+ "grad_norm": 6.014346599578857,
240
+ "learning_rate": 0.0009537856440511307,
241
+ "loss": 13.0983,
242
+ "step": 320
243
+ },
244
+ {
245
+ "epoch": 1.4586549062844543,
246
+ "grad_norm": 5.742127895355225,
247
+ "learning_rate": 0.0009488692232055063,
248
+ "loss": 12.6127,
249
+ "step": 330
250
+ },
251
+ {
252
+ "epoch": 1.5027563395810364,
253
+ "grad_norm": 5.869048595428467,
254
+ "learning_rate": 0.0009439528023598821,
255
+ "loss": 12.7718,
256
+ "step": 340
257
+ },
258
+ {
259
+ "epoch": 1.5468577728776185,
260
+ "grad_norm": 5.569489479064941,
261
+ "learning_rate": 0.0009390363815142577,
262
+ "loss": 12.4881,
263
+ "step": 350
264
+ },
265
+ {
266
+ "epoch": 1.5909592061742006,
267
+ "grad_norm": 6.744396686553955,
268
+ "learning_rate": 0.0009341199606686333,
269
+ "loss": 12.3667,
270
+ "step": 360
271
+ },
272
+ {
273
+ "epoch": 1.6350606394707827,
274
+ "grad_norm": 5.721745014190674,
275
+ "learning_rate": 0.0009292035398230089,
276
+ "loss": 12.169,
277
+ "step": 370
278
+ },
279
+ {
280
+ "epoch": 1.6791620727673648,
281
+ "grad_norm": 6.00828742980957,
282
+ "learning_rate": 0.0009242871189773845,
283
+ "loss": 12.163,
284
+ "step": 380
285
+ },
286
+ {
287
+ "epoch": 1.723263506063947,
288
+ "grad_norm": 6.328344345092773,
289
+ "learning_rate": 0.0009193706981317601,
290
+ "loss": 11.9877,
291
+ "step": 390
292
+ },
293
+ {
294
+ "epoch": 1.767364939360529,
295
+ "grad_norm": 5.9762492179870605,
296
+ "learning_rate": 0.0009144542772861358,
297
+ "loss": 11.8959,
298
+ "step": 400
299
+ },
300
+ {
301
+ "epoch": 1.8114663726571112,
302
+ "grad_norm": 5.462543487548828,
303
+ "learning_rate": 0.0009095378564405114,
304
+ "loss": 11.8443,
305
+ "step": 410
306
+ },
307
+ {
308
+ "epoch": 1.8555678059536935,
309
+ "grad_norm": 5.628457069396973,
310
+ "learning_rate": 0.000904621435594887,
311
+ "loss": 11.7006,
312
+ "step": 420
313
+ },
314
+ {
315
+ "epoch": 1.8996692392502756,
316
+ "grad_norm": 5.508877754211426,
317
+ "learning_rate": 0.0008997050147492626,
318
+ "loss": 11.6014,
319
+ "step": 430
320
+ },
321
+ {
322
+ "epoch": 1.9437706725468578,
323
+ "grad_norm": 5.510633945465088,
324
+ "learning_rate": 0.0008947885939036382,
325
+ "loss": 11.4442,
326
+ "step": 440
327
+ },
328
+ {
329
+ "epoch": 1.9878721058434399,
330
+ "grad_norm": 5.567264080047607,
331
+ "learning_rate": 0.0008898721730580139,
332
+ "loss": 11.5438,
333
+ "step": 450
334
+ },
335
+ {
336
+ "epoch": 1.9966923925027564,
337
+ "eval_accuracy": 0.299166181888695,
338
+ "eval_loss": 2.4033162593841553,
339
+ "eval_runtime": 157.919,
340
+ "eval_samples_per_second": 163.28,
341
+ "eval_steps_per_second": 0.64,
342
+ "step": 452
343
+ },
344
+ {
345
+ "epoch": 2.035281146637266,
346
+ "grad_norm": 5.653784275054932,
347
+ "learning_rate": 0.0008849557522123895,
348
+ "loss": 11.2957,
349
+ "step": 460
350
+ },
351
+ {
352
+ "epoch": 2.079382579933848,
353
+ "grad_norm": 5.515305519104004,
354
+ "learning_rate": 0.0008800393313667651,
355
+ "loss": 10.4382,
356
+ "step": 470
357
+ },
358
+ {
359
+ "epoch": 2.12348401323043,
360
+ "grad_norm": 5.540599346160889,
361
+ "learning_rate": 0.0008751229105211407,
362
+ "loss": 10.2655,
363
+ "step": 480
364
+ },
365
+ {
366
+ "epoch": 2.167585446527012,
367
+ "grad_norm": 5.588015079498291,
368
+ "learning_rate": 0.0008702064896755162,
369
+ "loss": 10.4359,
370
+ "step": 490
371
+ },
372
+ {
373
+ "epoch": 2.2116868798235942,
374
+ "grad_norm": 5.268083095550537,
375
+ "learning_rate": 0.0008652900688298918,
376
+ "loss": 10.4373,
377
+ "step": 500
378
+ },
379
+ {
380
+ "epoch": 2.2557883131201764,
381
+ "grad_norm": 5.2158427238464355,
382
+ "learning_rate": 0.0008603736479842675,
383
+ "loss": 10.3617,
384
+ "step": 510
385
+ },
386
+ {
387
+ "epoch": 2.2998897464167585,
388
+ "grad_norm": 5.967540264129639,
389
+ "learning_rate": 0.0008554572271386431,
390
+ "loss": 10.343,
391
+ "step": 520
392
+ },
393
+ {
394
+ "epoch": 2.3439911797133406,
395
+ "grad_norm": 6.330170154571533,
396
+ "learning_rate": 0.0008505408062930187,
397
+ "loss": 10.3236,
398
+ "step": 530
399
+ },
400
+ {
401
+ "epoch": 2.3880926130099227,
402
+ "grad_norm": 5.394957065582275,
403
+ "learning_rate": 0.0008456243854473943,
404
+ "loss": 10.1741,
405
+ "step": 540
406
+ },
407
+ {
408
+ "epoch": 2.432194046306505,
409
+ "grad_norm": 5.637348651885986,
410
+ "learning_rate": 0.0008407079646017699,
411
+ "loss": 10.1634,
412
+ "step": 550
413
+ },
414
+ {
415
+ "epoch": 2.476295479603087,
416
+ "grad_norm": 5.205583572387695,
417
+ "learning_rate": 0.0008357915437561455,
418
+ "loss": 10.0739,
419
+ "step": 560
420
+ },
421
+ {
422
+ "epoch": 2.5203969128996695,
423
+ "grad_norm": 5.155258655548096,
424
+ "learning_rate": 0.0008308751229105212,
425
+ "loss": 10.1469,
426
+ "step": 570
427
+ },
428
+ {
429
+ "epoch": 2.5644983461962516,
430
+ "grad_norm": 5.169893741607666,
431
+ "learning_rate": 0.0008259587020648968,
432
+ "loss": 10.0088,
433
+ "step": 580
434
+ },
435
+ {
436
+ "epoch": 2.6085997794928337,
437
+ "grad_norm": 5.327938556671143,
438
+ "learning_rate": 0.0008210422812192724,
439
+ "loss": 9.9356,
440
+ "step": 590
441
+ },
442
+ {
443
+ "epoch": 2.652701212789416,
444
+ "grad_norm": 5.036328315734863,
445
+ "learning_rate": 0.000816125860373648,
446
+ "loss": 9.8759,
447
+ "step": 600
448
+ },
449
+ {
450
+ "epoch": 2.696802646085998,
451
+ "grad_norm": 5.075291633605957,
452
+ "learning_rate": 0.0008112094395280236,
453
+ "loss": 9.8351,
454
+ "step": 610
455
+ },
456
+ {
457
+ "epoch": 2.74090407938258,
458
+ "grad_norm": 5.24298620223999,
459
+ "learning_rate": 0.0008062930186823992,
460
+ "loss": 9.6878,
461
+ "step": 620
462
+ },
463
+ {
464
+ "epoch": 2.785005512679162,
465
+ "grad_norm": 4.871809005737305,
466
+ "learning_rate": 0.0008013765978367749,
467
+ "loss": 9.9713,
468
+ "step": 630
469
+ },
470
+ {
471
+ "epoch": 2.8291069459757443,
472
+ "grad_norm": 4.931336879730225,
473
+ "learning_rate": 0.0007964601769911505,
474
+ "loss": 9.8542,
475
+ "step": 640
476
+ },
477
+ {
478
+ "epoch": 2.8732083792723264,
479
+ "grad_norm": 5.142650127410889,
480
+ "learning_rate": 0.0007915437561455261,
481
+ "loss": 9.8176,
482
+ "step": 650
483
+ },
484
+ {
485
+ "epoch": 2.9173098125689085,
486
+ "grad_norm": 5.100261211395264,
487
+ "learning_rate": 0.0007866273352999017,
488
+ "loss": 9.5995,
489
+ "step": 660
490
+ },
491
+ {
492
+ "epoch": 2.9614112458654906,
493
+ "grad_norm": 4.906562805175781,
494
+ "learning_rate": 0.0007817109144542773,
495
+ "loss": 9.5963,
496
+ "step": 670
497
+ },
498
+ {
499
+ "epoch": 2.9966923925027564,
500
+ "eval_accuracy": 0.3491564863292612,
501
+ "eval_loss": 2.0761232376098633,
502
+ "eval_runtime": 157.9952,
503
+ "eval_samples_per_second": 163.201,
504
+ "eval_steps_per_second": 0.639,
505
+ "step": 678
506
+ },
507
+ {
508
+ "epoch": 3.0088202866593163,
509
+ "grad_norm": 4.665123462677002,
510
+ "learning_rate": 0.000776794493608653,
511
+ "loss": 9.8294,
512
+ "step": 680
513
+ },
514
+ {
515
+ "epoch": 3.0529217199558984,
516
+ "grad_norm": 5.142998695373535,
517
+ "learning_rate": 0.0007718780727630286,
518
+ "loss": 8.6602,
519
+ "step": 690
520
+ },
521
+ {
522
+ "epoch": 3.0970231532524806,
523
+ "grad_norm": 4.909497261047363,
524
+ "learning_rate": 0.0007669616519174042,
525
+ "loss": 8.5178,
526
+ "step": 700
527
+ },
528
+ {
529
+ "epoch": 3.1411245865490627,
530
+ "grad_norm": 4.914758682250977,
531
+ "learning_rate": 0.0007620452310717798,
532
+ "loss": 8.5225,
533
+ "step": 710
534
+ },
535
+ {
536
+ "epoch": 3.185226019845645,
537
+ "grad_norm": 4.778907299041748,
538
+ "learning_rate": 0.0007571288102261554,
539
+ "loss": 8.6442,
540
+ "step": 720
541
+ },
542
+ {
543
+ "epoch": 3.2293274531422274,
544
+ "grad_norm": 4.7863240242004395,
545
+ "learning_rate": 0.0007522123893805309,
546
+ "loss": 8.6121,
547
+ "step": 730
548
+ },
549
+ {
550
+ "epoch": 3.2734288864388095,
551
+ "grad_norm": 4.866987228393555,
552
+ "learning_rate": 0.0007472959685349066,
553
+ "loss": 8.6773,
554
+ "step": 740
555
+ },
556
+ {
557
+ "epoch": 3.3175303197353916,
558
+ "grad_norm": 4.968977451324463,
559
+ "learning_rate": 0.0007423795476892822,
560
+ "loss": 8.6663,
561
+ "step": 750
562
+ },
563
+ {
564
+ "epoch": 3.3616317530319737,
565
+ "grad_norm": 4.805993556976318,
566
+ "learning_rate": 0.0007374631268436578,
567
+ "loss": 8.6653,
568
+ "step": 760
569
+ },
570
+ {
571
+ "epoch": 3.405733186328556,
572
+ "grad_norm": 4.871044635772705,
573
+ "learning_rate": 0.0007325467059980334,
574
+ "loss": 8.5913,
575
+ "step": 770
576
+ },
577
+ {
578
+ "epoch": 3.449834619625138,
579
+ "grad_norm": 5.305060863494873,
580
+ "learning_rate": 0.000727630285152409,
581
+ "loss": 8.4683,
582
+ "step": 780
583
+ },
584
+ {
585
+ "epoch": 3.49393605292172,
586
+ "grad_norm": 4.906181812286377,
587
+ "learning_rate": 0.0007227138643067846,
588
+ "loss": 8.5715,
589
+ "step": 790
590
+ },
591
+ {
592
+ "epoch": 3.538037486218302,
593
+ "grad_norm": 4.955236434936523,
594
+ "learning_rate": 0.0007177974434611603,
595
+ "loss": 8.5461,
596
+ "step": 800
597
+ },
598
+ {
599
+ "epoch": 3.5821389195148843,
600
+ "grad_norm": 4.7065863609313965,
601
+ "learning_rate": 0.0007128810226155359,
602
+ "loss": 8.5517,
603
+ "step": 810
604
+ },
605
+ {
606
+ "epoch": 3.6262403528114664,
607
+ "grad_norm": 4.8048930168151855,
608
+ "learning_rate": 0.0007079646017699115,
609
+ "loss": 8.6689,
610
+ "step": 820
611
+ },
612
+ {
613
+ "epoch": 3.6703417861080485,
614
+ "grad_norm": 5.108726978302002,
615
+ "learning_rate": 0.0007030481809242871,
616
+ "loss": 8.4989,
617
+ "step": 830
618
+ },
619
+ {
620
+ "epoch": 3.7144432194046306,
621
+ "grad_norm": 4.784049034118652,
622
+ "learning_rate": 0.0006981317600786627,
623
+ "loss": 8.5436,
624
+ "step": 840
625
+ },
626
+ {
627
+ "epoch": 3.7585446527012127,
628
+ "grad_norm": 4.953752517700195,
629
+ "learning_rate": 0.0006932153392330383,
630
+ "loss": 8.3707,
631
+ "step": 850
632
+ },
633
+ {
634
+ "epoch": 3.802646085997795,
635
+ "grad_norm": 4.866833686828613,
636
+ "learning_rate": 0.000688298918387414,
637
+ "loss": 8.5086,
638
+ "step": 860
639
+ },
640
+ {
641
+ "epoch": 3.846747519294377,
642
+ "grad_norm": 4.966568946838379,
643
+ "learning_rate": 0.0006833824975417896,
644
+ "loss": 8.3376,
645
+ "step": 870
646
+ },
647
+ {
648
+ "epoch": 3.890848952590959,
649
+ "grad_norm": 4.906131744384766,
650
+ "learning_rate": 0.0006784660766961652,
651
+ "loss": 8.3634,
652
+ "step": 880
653
+ },
654
+ {
655
+ "epoch": 3.934950385887541,
656
+ "grad_norm": 4.776595115661621,
657
+ "learning_rate": 0.0006735496558505408,
658
+ "loss": 8.3092,
659
+ "step": 890
660
+ },
661
+ {
662
+ "epoch": 3.9790518191841233,
663
+ "grad_norm": 4.59867525100708,
664
+ "learning_rate": 0.0006686332350049164,
665
+ "loss": 8.3873,
666
+ "step": 900
667
+ },
668
+ {
669
+ "epoch": 3.9966923925027564,
670
+ "eval_accuracy": 0.3934458018227652,
671
+ "eval_loss": 1.892139196395874,
672
+ "eval_runtime": 157.998,
673
+ "eval_samples_per_second": 163.198,
674
+ "eval_steps_per_second": 0.639,
675
+ "step": 904
676
+ },
677
+ {
678
+ "epoch": 4.0264608599779494,
679
+ "grad_norm": 4.579000473022461,
680
+ "learning_rate": 0.0006637168141592921,
681
+ "loss": 8.0129,
682
+ "step": 910
683
+ },
684
+ {
685
+ "epoch": 4.070562293274532,
686
+ "grad_norm": 4.628970623016357,
687
+ "learning_rate": 0.0006588003933136677,
688
+ "loss": 7.3258,
689
+ "step": 920
690
+ },
691
+ {
692
+ "epoch": 4.114663726571114,
693
+ "grad_norm": 4.4933953285217285,
694
+ "learning_rate": 0.0006538839724680433,
695
+ "loss": 7.3052,
696
+ "step": 930
697
+ },
698
+ {
699
+ "epoch": 4.158765159867696,
700
+ "grad_norm": 4.572768211364746,
701
+ "learning_rate": 0.0006489675516224189,
702
+ "loss": 7.3455,
703
+ "step": 940
704
+ },
705
+ {
706
+ "epoch": 4.202866593164278,
707
+ "grad_norm": 4.468959331512451,
708
+ "learning_rate": 0.0006440511307767945,
709
+ "loss": 7.3741,
710
+ "step": 950
711
+ },
712
+ {
713
+ "epoch": 4.24696802646086,
714
+ "grad_norm": 4.594937324523926,
715
+ "learning_rate": 0.00063913470993117,
716
+ "loss": 7.3599,
717
+ "step": 960
718
+ },
719
+ {
720
+ "epoch": 4.291069459757442,
721
+ "grad_norm": 4.6465935707092285,
722
+ "learning_rate": 0.0006342182890855457,
723
+ "loss": 7.3735,
724
+ "step": 970
725
+ },
726
+ {
727
+ "epoch": 4.335170893054024,
728
+ "grad_norm": 4.683464527130127,
729
+ "learning_rate": 0.0006293018682399213,
730
+ "loss": 7.447,
731
+ "step": 980
732
+ },
733
+ {
734
+ "epoch": 4.379272326350606,
735
+ "grad_norm": 4.466670989990234,
736
+ "learning_rate": 0.0006243854473942969,
737
+ "loss": 7.4649,
738
+ "step": 990
739
+ },
740
+ {
741
+ "epoch": 4.4233737596471885,
742
+ "grad_norm": 4.653200149536133,
743
+ "learning_rate": 0.0006194690265486725,
744
+ "loss": 7.3462,
745
+ "step": 1000
746
+ },
747
+ {
748
+ "epoch": 4.467475192943771,
749
+ "grad_norm": 4.518767356872559,
750
+ "learning_rate": 0.0006145526057030481,
751
+ "loss": 7.2768,
752
+ "step": 1010
753
+ },
754
+ {
755
+ "epoch": 4.511576626240353,
756
+ "grad_norm": 4.512480735778809,
757
+ "learning_rate": 0.0006096361848574237,
758
+ "loss": 7.306,
759
+ "step": 1020
760
+ },
761
+ {
762
+ "epoch": 4.555678059536935,
763
+ "grad_norm": 4.512141704559326,
764
+ "learning_rate": 0.0006047197640117994,
765
+ "loss": 7.3665,
766
+ "step": 1030
767
+ },
768
+ {
769
+ "epoch": 4.599779492833517,
770
+ "grad_norm": 4.769530773162842,
771
+ "learning_rate": 0.000599803343166175,
772
+ "loss": 7.4248,
773
+ "step": 1040
774
+ },
775
+ {
776
+ "epoch": 4.643880926130099,
777
+ "grad_norm": 4.538891792297363,
778
+ "learning_rate": 0.0005948869223205506,
779
+ "loss": 7.3548,
780
+ "step": 1050
781
+ },
782
+ {
783
+ "epoch": 4.687982359426681,
784
+ "grad_norm": 4.579529762268066,
785
+ "learning_rate": 0.0005899705014749262,
786
+ "loss": 7.1825,
787
+ "step": 1060
788
+ },
789
+ {
790
+ "epoch": 4.732083792723263,
791
+ "grad_norm": 4.407522201538086,
792
+ "learning_rate": 0.0005850540806293018,
793
+ "loss": 7.3086,
794
+ "step": 1070
795
+ },
796
+ {
797
+ "epoch": 4.776185226019845,
798
+ "grad_norm": 4.507376194000244,
799
+ "learning_rate": 0.0005801376597836774,
800
+ "loss": 7.4489,
801
+ "step": 1080
802
+ },
803
+ {
804
+ "epoch": 4.8202866593164275,
805
+ "grad_norm": 4.580904960632324,
806
+ "learning_rate": 0.0005752212389380532,
807
+ "loss": 7.2829,
808
+ "step": 1090
809
+ },
810
+ {
811
+ "epoch": 4.86438809261301,
812
+ "grad_norm": 4.334172248840332,
813
+ "learning_rate": 0.0005703048180924288,
814
+ "loss": 7.3591,
815
+ "step": 1100
816
+ },
817
+ {
818
+ "epoch": 4.908489525909592,
819
+ "grad_norm": 4.6015095710754395,
820
+ "learning_rate": 0.0005653883972468044,
821
+ "loss": 7.3586,
822
+ "step": 1110
823
+ },
824
+ {
825
+ "epoch": 4.952590959206174,
826
+ "grad_norm": 4.470212936401367,
827
+ "learning_rate": 0.00056047197640118,
828
+ "loss": 7.3276,
829
+ "step": 1120
830
+ },
831
+ {
832
+ "epoch": 4.996692392502756,
833
+ "grad_norm": 4.509373664855957,
834
+ "learning_rate": 0.0005555555555555556,
835
+ "loss": 7.3127,
836
+ "step": 1130
837
+ },
838
+ {
839
+ "epoch": 4.996692392502756,
840
+ "eval_accuracy": 0.42877642039945707,
841
+ "eval_loss": 1.7534652948379517,
842
+ "eval_runtime": 157.6275,
843
+ "eval_samples_per_second": 163.582,
844
+ "eval_steps_per_second": 0.641,
845
+ "step": 1130
846
+ },
847
+ {
848
+ "epoch": 5.044101433296582,
849
+ "grad_norm": 4.370789527893066,
850
+ "learning_rate": 0.0005506391347099313,
851
+ "loss": 6.6219,
852
+ "step": 1140
853
+ },
854
+ {
855
+ "epoch": 5.088202866593164,
856
+ "grad_norm": 4.587760925292969,
857
+ "learning_rate": 0.0005457227138643069,
858
+ "loss": 6.382,
859
+ "step": 1150
860
+ },
861
+ {
862
+ "epoch": 5.132304299889746,
863
+ "grad_norm": 4.437044143676758,
864
+ "learning_rate": 0.0005408062930186825,
865
+ "loss": 6.3493,
866
+ "step": 1160
867
+ },
868
+ {
869
+ "epoch": 5.1764057331863285,
870
+ "grad_norm": 4.376111030578613,
871
+ "learning_rate": 0.0005358898721730581,
872
+ "loss": 6.3537,
873
+ "step": 1170
874
+ },
875
+ {
876
+ "epoch": 5.220507166482911,
877
+ "grad_norm": 4.685988903045654,
878
+ "learning_rate": 0.0005309734513274337,
879
+ "loss": 6.3465,
880
+ "step": 1180
881
+ },
882
+ {
883
+ "epoch": 5.264608599779493,
884
+ "grad_norm": 4.62017297744751,
885
+ "learning_rate": 0.0005260570304818093,
886
+ "loss": 6.3306,
887
+ "step": 1190
888
+ },
889
+ {
890
+ "epoch": 5.308710033076075,
891
+ "grad_norm": 4.6863112449646,
892
+ "learning_rate": 0.0005211406096361849,
893
+ "loss": 6.4408,
894
+ "step": 1200
895
+ },
896
+ {
897
+ "epoch": 5.352811466372657,
898
+ "grad_norm": 4.470117092132568,
899
+ "learning_rate": 0.0005162241887905605,
900
+ "loss": 6.4383,
901
+ "step": 1210
902
+ },
903
+ {
904
+ "epoch": 5.396912899669239,
905
+ "grad_norm": 4.46852445602417,
906
+ "learning_rate": 0.0005113077679449361,
907
+ "loss": 6.2687,
908
+ "step": 1220
909
+ },
910
+ {
911
+ "epoch": 5.441014332965821,
912
+ "grad_norm": 4.40092134475708,
913
+ "learning_rate": 0.0005063913470993117,
914
+ "loss": 6.3344,
915
+ "step": 1230
916
+ },
917
+ {
918
+ "epoch": 5.485115766262403,
919
+ "grad_norm": 4.5052595138549805,
920
+ "learning_rate": 0.0005014749262536873,
921
+ "loss": 6.3289,
922
+ "step": 1240
923
+ },
924
+ {
925
+ "epoch": 5.529217199558985,
926
+ "grad_norm": 4.2100419998168945,
927
+ "learning_rate": 0.000496558505408063,
928
+ "loss": 6.4074,
929
+ "step": 1250
930
+ },
931
+ {
932
+ "epoch": 5.5733186328555675,
933
+ "grad_norm": 4.383902549743652,
934
+ "learning_rate": 0.0004916420845624386,
935
+ "loss": 6.3748,
936
+ "step": 1260
937
+ },
938
+ {
939
+ "epoch": 5.61742006615215,
940
+ "grad_norm": 4.348648548126221,
941
+ "learning_rate": 0.0004867256637168142,
942
+ "loss": 6.2771,
943
+ "step": 1270
944
+ },
945
+ {
946
+ "epoch": 5.661521499448732,
947
+ "grad_norm": 4.597408294677734,
948
+ "learning_rate": 0.0004818092428711898,
949
+ "loss": 6.4728,
950
+ "step": 1280
951
+ },
952
+ {
953
+ "epoch": 5.705622932745314,
954
+ "grad_norm": 4.483145236968994,
955
+ "learning_rate": 0.0004768928220255654,
956
+ "loss": 6.3158,
957
+ "step": 1290
958
+ },
959
+ {
960
+ "epoch": 5.749724366041896,
961
+ "grad_norm": 4.486307144165039,
962
+ "learning_rate": 0.00047197640117994103,
963
+ "loss": 6.3443,
964
+ "step": 1300
965
+ },
966
+ {
967
+ "epoch": 5.793825799338478,
968
+ "grad_norm": 4.527980804443359,
969
+ "learning_rate": 0.00046705998033431663,
970
+ "loss": 6.306,
971
+ "step": 1310
972
+ },
973
+ {
974
+ "epoch": 5.83792723263506,
975
+ "grad_norm": 4.423805236816406,
976
+ "learning_rate": 0.00046214355948869223,
977
+ "loss": 6.2912,
978
+ "step": 1320
979
+ },
980
+ {
981
+ "epoch": 5.882028665931642,
982
+ "grad_norm": 4.3704752922058105,
983
+ "learning_rate": 0.0004572271386430679,
984
+ "loss": 6.2006,
985
+ "step": 1330
986
+ },
987
+ {
988
+ "epoch": 5.926130099228224,
989
+ "grad_norm": 4.610422611236572,
990
+ "learning_rate": 0.0004523107177974435,
991
+ "loss": 6.2838,
992
+ "step": 1340
993
+ },
994
+ {
995
+ "epoch": 5.970231532524807,
996
+ "grad_norm": 4.506854057312012,
997
+ "learning_rate": 0.0004473942969518191,
998
+ "loss": 6.2178,
999
+ "step": 1350
1000
+ },
1001
+ {
1002
+ "epoch": 5.996692392502756,
1003
+ "eval_accuracy": 0.4608881132441342,
1004
+ "eval_loss": 1.6532503366470337,
1005
+ "eval_runtime": 157.0061,
1006
+ "eval_samples_per_second": 164.229,
1007
+ "eval_steps_per_second": 0.643,
1008
+ "step": 1356
1009
+ },
1010
+ {
1011
+ "epoch": 6.017640573318633,
1012
+ "grad_norm": 4.212904930114746,
1013
+ "learning_rate": 0.00044247787610619474,
1014
+ "loss": 6.1728,
1015
+ "step": 1360
1016
+ },
1017
+ {
1018
+ "epoch": 6.061742006615215,
1019
+ "grad_norm": 4.3509931564331055,
1020
+ "learning_rate": 0.00043756145526057034,
1021
+ "loss": 5.4675,
1022
+ "step": 1370
1023
+ },
1024
+ {
1025
+ "epoch": 6.105843439911797,
1026
+ "grad_norm": 4.346857070922852,
1027
+ "learning_rate": 0.0004326450344149459,
1028
+ "loss": 5.4391,
1029
+ "step": 1380
1030
+ },
1031
+ {
1032
+ "epoch": 6.149944873208379,
1033
+ "grad_norm": 4.269370079040527,
1034
+ "learning_rate": 0.00042772861356932154,
1035
+ "loss": 5.6166,
1036
+ "step": 1390
1037
+ },
1038
+ {
1039
+ "epoch": 6.194046306504961,
1040
+ "grad_norm": 4.528676986694336,
1041
+ "learning_rate": 0.00042281219272369714,
1042
+ "loss": 5.4683,
1043
+ "step": 1400
1044
+ },
1045
+ {
1046
+ "epoch": 6.238147739801543,
1047
+ "grad_norm": 4.301924705505371,
1048
+ "learning_rate": 0.00041789577187807274,
1049
+ "loss": 5.6475,
1050
+ "step": 1410
1051
+ },
1052
+ {
1053
+ "epoch": 6.282249173098125,
1054
+ "grad_norm": 4.408478736877441,
1055
+ "learning_rate": 0.0004129793510324484,
1056
+ "loss": 5.5012,
1057
+ "step": 1420
1058
+ },
1059
+ {
1060
+ "epoch": 6.3263506063947075,
1061
+ "grad_norm": 4.205230712890625,
1062
+ "learning_rate": 0.000408062930186824,
1063
+ "loss": 5.4751,
1064
+ "step": 1430
1065
+ },
1066
+ {
1067
+ "epoch": 6.37045203969129,
1068
+ "grad_norm": 4.417777061462402,
1069
+ "learning_rate": 0.0004031465093411996,
1070
+ "loss": 5.4574,
1071
+ "step": 1440
1072
+ },
1073
+ {
1074
+ "epoch": 6.414553472987872,
1075
+ "grad_norm": 4.562683582305908,
1076
+ "learning_rate": 0.00039823008849557525,
1077
+ "loss": 5.5681,
1078
+ "step": 1450
1079
+ },
1080
+ {
1081
+ "epoch": 6.458654906284455,
1082
+ "grad_norm": 4.4911112785339355,
1083
+ "learning_rate": 0.00039331366764995085,
1084
+ "loss": 5.5747,
1085
+ "step": 1460
1086
+ },
1087
+ {
1088
+ "epoch": 6.502756339581037,
1089
+ "grad_norm": 4.3542890548706055,
1090
+ "learning_rate": 0.0003883972468043265,
1091
+ "loss": 5.4824,
1092
+ "step": 1470
1093
+ },
1094
+ {
1095
+ "epoch": 6.546857772877619,
1096
+ "grad_norm": 4.291258811950684,
1097
+ "learning_rate": 0.0003834808259587021,
1098
+ "loss": 5.3834,
1099
+ "step": 1480
1100
+ },
1101
+ {
1102
+ "epoch": 6.590959206174201,
1103
+ "grad_norm": 4.43005895614624,
1104
+ "learning_rate": 0.0003785644051130777,
1105
+ "loss": 5.4797,
1106
+ "step": 1490
1107
+ },
1108
+ {
1109
+ "epoch": 6.635060639470783,
1110
+ "grad_norm": 4.461911201477051,
1111
+ "learning_rate": 0.0003736479842674533,
1112
+ "loss": 5.5312,
1113
+ "step": 1500
1114
+ },
1115
+ {
1116
+ "epoch": 6.679162072767365,
1117
+ "grad_norm": 4.246000289916992,
1118
+ "learning_rate": 0.0003687315634218289,
1119
+ "loss": 5.5821,
1120
+ "step": 1510
1121
+ },
1122
+ {
1123
+ "epoch": 6.723263506063947,
1124
+ "grad_norm": 4.318939685821533,
1125
+ "learning_rate": 0.0003638151425762045,
1126
+ "loss": 5.5103,
1127
+ "step": 1520
1128
+ },
1129
+ {
1130
+ "epoch": 6.7673649393605295,
1131
+ "grad_norm": 4.241775989532471,
1132
+ "learning_rate": 0.00035889872173058016,
1133
+ "loss": 5.4244,
1134
+ "step": 1530
1135
+ },
1136
+ {
1137
+ "epoch": 6.811466372657112,
1138
+ "grad_norm": 4.246124267578125,
1139
+ "learning_rate": 0.00035398230088495576,
1140
+ "loss": 5.4415,
1141
+ "step": 1540
1142
+ },
1143
+ {
1144
+ "epoch": 6.855567805953694,
1145
+ "grad_norm": 4.474425792694092,
1146
+ "learning_rate": 0.00034906588003933136,
1147
+ "loss": 5.4259,
1148
+ "step": 1550
1149
+ },
1150
+ {
1151
+ "epoch": 6.899669239250276,
1152
+ "grad_norm": 4.427676677703857,
1153
+ "learning_rate": 0.000344149459193707,
1154
+ "loss": 5.4803,
1155
+ "step": 1560
1156
+ },
1157
+ {
1158
+ "epoch": 6.943770672546858,
1159
+ "grad_norm": 4.402575969696045,
1160
+ "learning_rate": 0.0003392330383480826,
1161
+ "loss": 5.3709,
1162
+ "step": 1570
1163
+ },
1164
+ {
1165
+ "epoch": 6.98787210584344,
1166
+ "grad_norm": 4.089934349060059,
1167
+ "learning_rate": 0.0003343166175024582,
1168
+ "loss": 5.4619,
1169
+ "step": 1580
1170
+ },
1171
+ {
1172
+ "epoch": 6.996692392502756,
1173
+ "eval_accuracy": 0.4855923986814039,
1174
+ "eval_loss": 1.5859359502792358,
1175
+ "eval_runtime": 157.4699,
1176
+ "eval_samples_per_second": 163.746,
1177
+ "eval_steps_per_second": 0.641,
1178
+ "step": 1582
1179
+ },
1180
+ {
1181
+ "epoch": 7.035281146637265,
1182
+ "grad_norm": 4.048954963684082,
1183
+ "learning_rate": 0.00032940019665683387,
1184
+ "loss": 4.9956,
1185
+ "step": 1590
1186
+ },
1187
+ {
1188
+ "epoch": 7.0793825799338475,
1189
+ "grad_norm": 4.142811298370361,
1190
+ "learning_rate": 0.00032448377581120947,
1191
+ "loss": 4.637,
1192
+ "step": 1600
1193
+ },
1194
+ {
1195
+ "epoch": 7.12348401323043,
1196
+ "grad_norm": 4.157500267028809,
1197
+ "learning_rate": 0.000319567354965585,
1198
+ "loss": 4.7621,
1199
+ "step": 1610
1200
+ },
1201
+ {
1202
+ "epoch": 7.167585446527012,
1203
+ "grad_norm": 4.244820594787598,
1204
+ "learning_rate": 0.00031465093411996067,
1205
+ "loss": 4.671,
1206
+ "step": 1620
1207
+ },
1208
+ {
1209
+ "epoch": 7.211686879823595,
1210
+ "grad_norm": 4.291553497314453,
1211
+ "learning_rate": 0.00030973451327433627,
1212
+ "loss": 4.7641,
1213
+ "step": 1630
1214
+ },
1215
+ {
1216
+ "epoch": 7.255788313120177,
1217
+ "grad_norm": 4.3009233474731445,
1218
+ "learning_rate": 0.00030481809242871187,
1219
+ "loss": 4.7801,
1220
+ "step": 1640
1221
+ },
1222
+ {
1223
+ "epoch": 7.299889746416759,
1224
+ "grad_norm": 4.151345729827881,
1225
+ "learning_rate": 0.0002999016715830875,
1226
+ "loss": 4.7008,
1227
+ "step": 1650
1228
+ },
1229
+ {
1230
+ "epoch": 7.343991179713341,
1231
+ "grad_norm": 4.11318826675415,
1232
+ "learning_rate": 0.0002949852507374631,
1233
+ "loss": 4.5749,
1234
+ "step": 1660
1235
+ },
1236
+ {
1237
+ "epoch": 7.388092613009923,
1238
+ "grad_norm": 4.272958755493164,
1239
+ "learning_rate": 0.0002900688298918387,
1240
+ "loss": 4.7279,
1241
+ "step": 1670
1242
+ },
1243
+ {
1244
+ "epoch": 7.432194046306505,
1245
+ "grad_norm": 4.370232105255127,
1246
+ "learning_rate": 0.0002851524090462144,
1247
+ "loss": 4.6739,
1248
+ "step": 1680
1249
+ },
1250
+ {
1251
+ "epoch": 7.476295479603087,
1252
+ "grad_norm": 4.224992752075195,
1253
+ "learning_rate": 0.00028023598820059,
1254
+ "loss": 4.6791,
1255
+ "step": 1690
1256
+ },
1257
+ {
1258
+ "epoch": 7.5203969128996695,
1259
+ "grad_norm": 4.131043434143066,
1260
+ "learning_rate": 0.00027531956735496563,
1261
+ "loss": 4.742,
1262
+ "step": 1700
1263
+ },
1264
+ {
1265
+ "epoch": 7.564498346196252,
1266
+ "grad_norm": 4.325319290161133,
1267
+ "learning_rate": 0.00027040314650934123,
1268
+ "loss": 4.6848,
1269
+ "step": 1710
1270
+ },
1271
+ {
1272
+ "epoch": 7.608599779492834,
1273
+ "grad_norm": 4.375499248504639,
1274
+ "learning_rate": 0.00026548672566371683,
1275
+ "loss": 4.7289,
1276
+ "step": 1720
1277
+ },
1278
+ {
1279
+ "epoch": 7.652701212789416,
1280
+ "grad_norm": 4.138423442840576,
1281
+ "learning_rate": 0.00026057030481809243,
1282
+ "loss": 4.7507,
1283
+ "step": 1730
1284
+ },
1285
+ {
1286
+ "epoch": 7.696802646085998,
1287
+ "grad_norm": 4.30584716796875,
1288
+ "learning_rate": 0.00025565388397246803,
1289
+ "loss": 4.7498,
1290
+ "step": 1740
1291
+ },
1292
+ {
1293
+ "epoch": 7.74090407938258,
1294
+ "grad_norm": 4.196134567260742,
1295
+ "learning_rate": 0.00025073746312684363,
1296
+ "loss": 4.6912,
1297
+ "step": 1750
1298
+ },
1299
+ {
1300
+ "epoch": 7.785005512679162,
1301
+ "grad_norm": 4.179983615875244,
1302
+ "learning_rate": 0.0002458210422812193,
1303
+ "loss": 4.6327,
1304
+ "step": 1760
1305
+ },
1306
+ {
1307
+ "epoch": 7.829106945975744,
1308
+ "grad_norm": 4.482637405395508,
1309
+ "learning_rate": 0.0002409046214355949,
1310
+ "loss": 4.7178,
1311
+ "step": 1770
1312
+ },
1313
+ {
1314
+ "epoch": 7.873208379272326,
1315
+ "grad_norm": 4.053654670715332,
1316
+ "learning_rate": 0.00023598820058997051,
1317
+ "loss": 4.6353,
1318
+ "step": 1780
1319
+ },
1320
+ {
1321
+ "epoch": 7.9173098125689085,
1322
+ "grad_norm": 4.109724998474121,
1323
+ "learning_rate": 0.00023107177974434611,
1324
+ "loss": 4.6531,
1325
+ "step": 1790
1326
+ },
1327
+ {
1328
+ "epoch": 7.961411245865491,
1329
+ "grad_norm": 4.190576553344727,
1330
+ "learning_rate": 0.00022615535889872174,
1331
+ "loss": 4.618,
1332
+ "step": 1800
1333
+ },
1334
+ {
1335
+ "epoch": 7.996692392502756,
1336
+ "eval_accuracy": 0.5128950940469265,
1337
+ "eval_loss": 1.52533757686615,
1338
+ "eval_runtime": 159.0581,
1339
+ "eval_samples_per_second": 162.111,
1340
+ "eval_steps_per_second": 0.635,
1341
+ "step": 1808
1342
+ },
1343
+ {
1344
+ "epoch": 8.008820286659317,
1345
+ "grad_norm": 3.9392812252044678,
1346
+ "learning_rate": 0.00022123893805309737,
1347
+ "loss": 4.6935,
1348
+ "step": 1810
1349
+ },
1350
+ {
1351
+ "epoch": 8.052921719955899,
1352
+ "grad_norm": 3.8019185066223145,
1353
+ "learning_rate": 0.00021632251720747294,
1354
+ "loss": 3.9004,
1355
+ "step": 1820
1356
+ },
1357
+ {
1358
+ "epoch": 8.097023153252481,
1359
+ "grad_norm": 3.8372962474823,
1360
+ "learning_rate": 0.00021140609636184857,
1361
+ "loss": 4.0764,
1362
+ "step": 1830
1363
+ },
1364
+ {
1365
+ "epoch": 8.141124586549063,
1366
+ "grad_norm": 4.013173580169678,
1367
+ "learning_rate": 0.0002064896755162242,
1368
+ "loss": 3.921,
1369
+ "step": 1840
1370
+ },
1371
+ {
1372
+ "epoch": 8.185226019845645,
1373
+ "grad_norm": 4.06724739074707,
1374
+ "learning_rate": 0.0002015732546705998,
1375
+ "loss": 4.1819,
1376
+ "step": 1850
1377
+ },
1378
+ {
1379
+ "epoch": 8.229327453142227,
1380
+ "grad_norm": 3.967484474182129,
1381
+ "learning_rate": 0.00019665683382497542,
1382
+ "loss": 4.1189,
1383
+ "step": 1860
1384
+ },
1385
+ {
1386
+ "epoch": 8.27342888643881,
1387
+ "grad_norm": 3.8714957237243652,
1388
+ "learning_rate": 0.00019174041297935105,
1389
+ "loss": 4.0143,
1390
+ "step": 1870
1391
+ },
1392
+ {
1393
+ "epoch": 8.317530319735392,
1394
+ "grad_norm": 3.9495551586151123,
1395
+ "learning_rate": 0.00018682399213372665,
1396
+ "loss": 4.1982,
1397
+ "step": 1880
1398
+ },
1399
+ {
1400
+ "epoch": 8.361631753031974,
1401
+ "grad_norm": 4.08094596862793,
1402
+ "learning_rate": 0.00018190757128810225,
1403
+ "loss": 4.0732,
1404
+ "step": 1890
1405
+ },
1406
+ {
1407
+ "epoch": 8.405733186328556,
1408
+ "grad_norm": 3.997749090194702,
1409
+ "learning_rate": 0.00017699115044247788,
1410
+ "loss": 4.0657,
1411
+ "step": 1900
1412
+ },
1413
+ {
1414
+ "epoch": 8.449834619625138,
1415
+ "grad_norm": 3.9958107471466064,
1416
+ "learning_rate": 0.0001720747295968535,
1417
+ "loss": 4.0651,
1418
+ "step": 1910
1419
+ },
1420
+ {
1421
+ "epoch": 8.49393605292172,
1422
+ "grad_norm": 4.039346218109131,
1423
+ "learning_rate": 0.0001671583087512291,
1424
+ "loss": 4.0022,
1425
+ "step": 1920
1426
+ },
1427
+ {
1428
+ "epoch": 8.538037486218302,
1429
+ "grad_norm": 4.07537317276001,
1430
+ "learning_rate": 0.00016224188790560473,
1431
+ "loss": 4.0364,
1432
+ "step": 1930
1433
+ },
1434
+ {
1435
+ "epoch": 8.582138919514884,
1436
+ "grad_norm": 3.949800491333008,
1437
+ "learning_rate": 0.00015732546705998033,
1438
+ "loss": 4.0273,
1439
+ "step": 1940
1440
+ },
1441
+ {
1442
+ "epoch": 8.626240352811466,
1443
+ "grad_norm": 3.9785361289978027,
1444
+ "learning_rate": 0.00015240904621435593,
1445
+ "loss": 4.1008,
1446
+ "step": 1950
1447
+ },
1448
+ {
1449
+ "epoch": 8.670341786108049,
1450
+ "grad_norm": 4.0533318519592285,
1451
+ "learning_rate": 0.00014749262536873156,
1452
+ "loss": 3.9856,
1453
+ "step": 1960
1454
+ },
1455
+ {
1456
+ "epoch": 8.71444321940463,
1457
+ "grad_norm": 3.9734625816345215,
1458
+ "learning_rate": 0.0001425762045231072,
1459
+ "loss": 3.9327,
1460
+ "step": 1970
1461
+ },
1462
+ {
1463
+ "epoch": 8.758544652701213,
1464
+ "grad_norm": 4.144639015197754,
1465
+ "learning_rate": 0.00013765978367748282,
1466
+ "loss": 4.1105,
1467
+ "step": 1980
1468
+ },
1469
+ {
1470
+ "epoch": 8.802646085997795,
1471
+ "grad_norm": 4.224411487579346,
1472
+ "learning_rate": 0.00013274336283185842,
1473
+ "loss": 4.0302,
1474
+ "step": 1990
1475
+ },
1476
+ {
1477
+ "epoch": 8.846747519294377,
1478
+ "grad_norm": 4.210290908813477,
1479
+ "learning_rate": 0.00012782694198623402,
1480
+ "loss": 3.9702,
1481
+ "step": 2000
1482
+ },
1483
+ {
1484
+ "epoch": 8.890848952590959,
1485
+ "grad_norm": 3.948779821395874,
1486
+ "learning_rate": 0.00012291052114060964,
1487
+ "loss": 4.0179,
1488
+ "step": 2010
1489
+ },
1490
+ {
1491
+ "epoch": 8.934950385887541,
1492
+ "grad_norm": 4.020263671875,
1493
+ "learning_rate": 0.00011799410029498526,
1494
+ "loss": 3.9426,
1495
+ "step": 2020
1496
+ },
1497
+ {
1498
+ "epoch": 8.979051819184123,
1499
+ "grad_norm": 3.9727530479431152,
1500
+ "learning_rate": 0.00011307767944936087,
1501
+ "loss": 3.9349,
1502
+ "step": 2030
1503
+ },
1504
+ {
1505
+ "epoch": 8.996692392502757,
1506
+ "eval_accuracy": 0.5314717859220477,
1507
+ "eval_loss": 1.4693009853363037,
1508
+ "eval_runtime": 158.202,
1509
+ "eval_samples_per_second": 162.988,
1510
+ "eval_steps_per_second": 0.638,
1511
+ "step": 2034
1512
+ },
1513
+ {
1514
+ "epoch": 9.026460859977949,
1515
+ "grad_norm": 4.023420333862305,
1516
+ "learning_rate": 0.00010816125860373647,
1517
+ "loss": 4.0029,
1518
+ "step": 2040
1519
+ },
1520
+ {
1521
+ "epoch": 9.07056229327453,
1522
+ "grad_norm": 3.8481717109680176,
1523
+ "learning_rate": 0.0001032448377581121,
1524
+ "loss": 3.5067,
1525
+ "step": 2050
1526
+ },
1527
+ {
1528
+ "epoch": 9.114663726571113,
1529
+ "grad_norm": 3.807828187942505,
1530
+ "learning_rate": 9.832841691248771e-05,
1531
+ "loss": 3.5354,
1532
+ "step": 2060
1533
+ },
1534
+ {
1535
+ "epoch": 9.158765159867695,
1536
+ "grad_norm": 3.808962106704712,
1537
+ "learning_rate": 9.341199606686333e-05,
1538
+ "loss": 3.6032,
1539
+ "step": 2070
1540
+ },
1541
+ {
1542
+ "epoch": 9.202866593164277,
1543
+ "grad_norm": 3.6250662803649902,
1544
+ "learning_rate": 8.849557522123894e-05,
1545
+ "loss": 3.5225,
1546
+ "step": 2080
1547
+ },
1548
+ {
1549
+ "epoch": 9.24696802646086,
1550
+ "grad_norm": 3.9860610961914062,
1551
+ "learning_rate": 8.357915437561455e-05,
1552
+ "loss": 3.6428,
1553
+ "step": 2090
1554
+ },
1555
+ {
1556
+ "epoch": 9.291069459757441,
1557
+ "grad_norm": 3.9992544651031494,
1558
+ "learning_rate": 7.866273352999017e-05,
1559
+ "loss": 3.4771,
1560
+ "step": 2100
1561
+ },
1562
+ {
1563
+ "epoch": 9.335170893054023,
1564
+ "grad_norm": 3.782989740371704,
1565
+ "learning_rate": 7.374631268436578e-05,
1566
+ "loss": 3.506,
1567
+ "step": 2110
1568
+ },
1569
+ {
1570
+ "epoch": 9.379272326350605,
1571
+ "grad_norm": 3.9225122928619385,
1572
+ "learning_rate": 6.882989183874141e-05,
1573
+ "loss": 3.4989,
1574
+ "step": 2120
1575
+ },
1576
+ {
1577
+ "epoch": 9.42337375964719,
1578
+ "grad_norm": 3.853450298309326,
1579
+ "learning_rate": 6.391347099311701e-05,
1580
+ "loss": 3.49,
1581
+ "step": 2130
1582
+ },
1583
+ {
1584
+ "epoch": 9.467475192943771,
1585
+ "grad_norm": 3.887131452560425,
1586
+ "learning_rate": 5.899705014749263e-05,
1587
+ "loss": 3.5517,
1588
+ "step": 2140
1589
+ },
1590
+ {
1591
+ "epoch": 9.511576626240354,
1592
+ "grad_norm": 3.7988171577453613,
1593
+ "learning_rate": 5.4080629301868236e-05,
1594
+ "loss": 3.4626,
1595
+ "step": 2150
1596
+ },
1597
+ {
1598
+ "epoch": 9.555678059536936,
1599
+ "grad_norm": 3.80769419670105,
1600
+ "learning_rate": 4.9164208456243856e-05,
1601
+ "loss": 3.5627,
1602
+ "step": 2160
1603
+ },
1604
+ {
1605
+ "epoch": 9.599779492833518,
1606
+ "grad_norm": 3.883881092071533,
1607
+ "learning_rate": 4.424778761061947e-05,
1608
+ "loss": 3.5904,
1609
+ "step": 2170
1610
+ },
1611
+ {
1612
+ "epoch": 9.6438809261301,
1613
+ "grad_norm": 3.729762554168701,
1614
+ "learning_rate": 3.9331366764995083e-05,
1615
+ "loss": 3.5366,
1616
+ "step": 2180
1617
+ },
1618
+ {
1619
+ "epoch": 9.687982359426682,
1620
+ "grad_norm": 3.7784054279327393,
1621
+ "learning_rate": 3.4414945919370704e-05,
1622
+ "loss": 3.5186,
1623
+ "step": 2190
1624
+ },
1625
+ {
1626
+ "epoch": 9.732083792723264,
1627
+ "grad_norm": 3.8702545166015625,
1628
+ "learning_rate": 2.9498525073746314e-05,
1629
+ "loss": 3.385,
1630
+ "step": 2200
1631
+ },
1632
+ {
1633
+ "epoch": 9.776185226019846,
1634
+ "grad_norm": 3.82987380027771,
1635
+ "learning_rate": 2.4582104228121928e-05,
1636
+ "loss": 3.4502,
1637
+ "step": 2210
1638
+ },
1639
+ {
1640
+ "epoch": 9.820286659316428,
1641
+ "grad_norm": 3.7813146114349365,
1642
+ "learning_rate": 1.9665683382497542e-05,
1643
+ "loss": 3.4203,
1644
+ "step": 2220
1645
+ },
1646
+ {
1647
+ "epoch": 9.86438809261301,
1648
+ "grad_norm": 3.8914685249328613,
1649
+ "learning_rate": 1.4749262536873157e-05,
1650
+ "loss": 3.5059,
1651
+ "step": 2230
1652
+ },
1653
+ {
1654
+ "epoch": 9.908489525909593,
1655
+ "grad_norm": 3.8533997535705566,
1656
+ "learning_rate": 9.832841691248771e-06,
1657
+ "loss": 3.483,
1658
+ "step": 2240
1659
+ },
1660
+ {
1661
+ "epoch": 9.952590959206175,
1662
+ "grad_norm": 3.7153797149658203,
1663
+ "learning_rate": 4.9164208456243854e-06,
1664
+ "loss": 3.5258,
1665
+ "step": 2250
1666
+ },
1667
+ {
1668
+ "epoch": 9.996692392502757,
1669
+ "grad_norm": 3.670651435852051,
1670
+ "learning_rate": 0.0,
1671
+ "loss": 3.4667,
1672
+ "step": 2260
1673
+ },
1674
+ {
1675
+ "epoch": 9.996692392502757,
1676
+ "eval_accuracy": 0.541167345355827,
1677
+ "eval_loss": 1.4404356479644775,
1678
+ "eval_runtime": 159.5598,
1679
+ "eval_samples_per_second": 161.601,
1680
+ "eval_steps_per_second": 0.633,
1681
+ "step": 2260
1682
+ },
1683
+ {
1684
+ "epoch": 9.996692392502757,
1685
+ "step": 2260,
1686
+ "total_flos": 1.8565430446260014e+20,
1687
+ "train_loss": 8.705859319298668,
1688
+ "train_runtime": 25889.9118,
1689
+ "train_samples_per_second": 89.633,
1690
+ "train_steps_per_second": 0.087
1691
+ }
1692
+ ],
1693
+ "logging_steps": 10,
1694
+ "max_steps": 2260,
1695
+ "num_input_tokens_seen": 0,
1696
+ "num_train_epochs": 10,
1697
+ "save_steps": 500,
1698
+ "stateful_callbacks": {
1699
+ "TrainerControl": {
1700
+ "args": {
1701
+ "should_epoch_stop": false,
1702
+ "should_evaluate": false,
1703
+ "should_log": false,
1704
+ "should_save": true,
1705
+ "should_training_stop": true
1706
+ },
1707
+ "attributes": {}
1708
+ }
1709
+ },
1710
+ "total_flos": 1.8565430446260014e+20,
1711
+ "train_batch_size": 256,
1712
+ "trial_name": null,
1713
+ "trial_params": null
1714
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:3533ba38e2be96525323398ed11e8dbbfd1d3b254b562fbcf4b42a505fdeeb5c
3
  size 5496
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e86ee97e08f747eea8db0b784efda5bfb2453e173b613fed5408bef54109a2e7
3
  size 5496