vishalkatheriya18
commited on
Commit
•
41c8669
1
Parent(s):
a9b9bca
End of training
Browse files- README.md +196 -0
- all_results.json +13 -0
- config.json +67 -0
- eval_results.json +8 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- train_results.json +8 -0
- trainer_state.json +3465 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,196 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: facebook/convnextv2-tiny-1k-224
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- imagefolder
|
8 |
+
metrics:
|
9 |
+
- accuracy
|
10 |
+
model-index:
|
11 |
+
- name: convnextv2-tiny-1k-224-finetuned-pattern-rgb
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Image Classification
|
15 |
+
type: image-classification
|
16 |
+
dataset:
|
17 |
+
name: imagefolder
|
18 |
+
type: imagefolder
|
19 |
+
config: default
|
20 |
+
split: train
|
21 |
+
args: default
|
22 |
+
metrics:
|
23 |
+
- name: Accuracy
|
24 |
+
type: accuracy
|
25 |
+
value: 0.875
|
26 |
+
---
|
27 |
+
|
28 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
+
should probably proofread and complete it, then remove this comment. -->
|
30 |
+
|
31 |
+
# convnextv2-tiny-1k-224-finetuned-pattern-rgb
|
32 |
+
|
33 |
+
This model is a fine-tuned version of [facebook/convnextv2-tiny-1k-224](https://huggingface.co/facebook/convnextv2-tiny-1k-224) on the imagefolder dataset.
|
34 |
+
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.5797
|
36 |
+
- Accuracy: 0.875
|
37 |
+
|
38 |
+
## Model description
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Intended uses & limitations
|
43 |
+
|
44 |
+
More information needed
|
45 |
+
|
46 |
+
## Training and evaluation data
|
47 |
+
|
48 |
+
More information needed
|
49 |
+
|
50 |
+
## Training procedure
|
51 |
+
|
52 |
+
### Training hyperparameters
|
53 |
+
|
54 |
+
The following hyperparameters were used during training:
|
55 |
+
- learning_rate: 5e-05
|
56 |
+
- train_batch_size: 32
|
57 |
+
- eval_batch_size: 32
|
58 |
+
- seed: 42
|
59 |
+
- gradient_accumulation_steps: 4
|
60 |
+
- total_train_batch_size: 128
|
61 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
62 |
+
- lr_scheduler_type: linear
|
63 |
+
- lr_scheduler_warmup_ratio: 0.1
|
64 |
+
- num_epochs: 120
|
65 |
+
|
66 |
+
### Training results
|
67 |
+
|
68 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
69 |
+
|:-------------:|:--------:|:----:|:---------------:|:--------:|
|
70 |
+
| 1.6954 | 0.9912 | 28 | 1.4482 | 0.58 |
|
71 |
+
| 1.015 | 1.9823 | 56 | 0.9088 | 0.7175 |
|
72 |
+
| 0.7953 | 2.9735 | 84 | 0.7266 | 0.7625 |
|
73 |
+
| 0.627 | 4.0 | 113 | 0.5872 | 0.8 |
|
74 |
+
| 0.4684 | 4.9912 | 141 | 0.5534 | 0.8175 |
|
75 |
+
| 0.4301 | 5.9823 | 169 | 0.5053 | 0.8275 |
|
76 |
+
| 0.3716 | 6.9735 | 197 | 0.4885 | 0.83 |
|
77 |
+
| 0.3798 | 8.0 | 226 | 0.4639 | 0.8525 |
|
78 |
+
| 0.3123 | 8.9912 | 254 | 0.5282 | 0.825 |
|
79 |
+
| 0.3148 | 9.9823 | 282 | 0.4569 | 0.8475 |
|
80 |
+
| 0.2427 | 10.9735 | 310 | 0.4206 | 0.865 |
|
81 |
+
| 0.2198 | 12.0 | 339 | 0.4832 | 0.84 |
|
82 |
+
| 0.1995 | 12.9912 | 367 | 0.4468 | 0.865 |
|
83 |
+
| 0.1738 | 13.9823 | 395 | 0.5668 | 0.8425 |
|
84 |
+
| 0.1683 | 14.9735 | 423 | 0.4454 | 0.8725 |
|
85 |
+
| 0.1426 | 16.0 | 452 | 0.5118 | 0.8525 |
|
86 |
+
| 0.133 | 16.9912 | 480 | 0.4713 | 0.865 |
|
87 |
+
| 0.1148 | 17.9823 | 508 | 0.5226 | 0.855 |
|
88 |
+
| 0.1147 | 18.9735 | 536 | 0.5333 | 0.8425 |
|
89 |
+
| 0.1284 | 20.0 | 565 | 0.4399 | 0.8575 |
|
90 |
+
| 0.1035 | 20.9912 | 593 | 0.5194 | 0.8525 |
|
91 |
+
| 0.1054 | 21.9823 | 621 | 0.5140 | 0.845 |
|
92 |
+
| 0.1056 | 22.9735 | 649 | 0.5183 | 0.87 |
|
93 |
+
| 0.1224 | 24.0 | 678 | 0.5293 | 0.85 |
|
94 |
+
| 0.0956 | 24.9912 | 706 | 0.4985 | 0.87 |
|
95 |
+
| 0.0717 | 25.9823 | 734 | 0.5267 | 0.8625 |
|
96 |
+
| 0.0858 | 26.9735 | 762 | 0.5525 | 0.8575 |
|
97 |
+
| 0.097 | 28.0 | 791 | 0.5340 | 0.855 |
|
98 |
+
| 0.0914 | 28.9912 | 819 | 0.4830 | 0.87 |
|
99 |
+
| 0.0699 | 29.9823 | 847 | 0.4883 | 0.8725 |
|
100 |
+
| 0.0932 | 30.9735 | 875 | 0.6106 | 0.8575 |
|
101 |
+
| 0.0967 | 32.0 | 904 | 0.5614 | 0.855 |
|
102 |
+
| 0.101 | 32.9912 | 932 | 0.5947 | 0.8525 |
|
103 |
+
| 0.0734 | 33.9823 | 960 | 0.5388 | 0.87 |
|
104 |
+
| 0.0742 | 34.9735 | 988 | 0.5110 | 0.8725 |
|
105 |
+
| 0.0698 | 36.0 | 1017 | 0.5384 | 0.8525 |
|
106 |
+
| 0.0785 | 36.9912 | 1045 | 0.5407 | 0.8475 |
|
107 |
+
| 0.0718 | 37.9823 | 1073 | 0.5420 | 0.86 |
|
108 |
+
| 0.061 | 38.9735 | 1101 | 0.5747 | 0.8675 |
|
109 |
+
| 0.0695 | 40.0 | 1130 | 0.5829 | 0.8575 |
|
110 |
+
| 0.0611 | 40.9912 | 1158 | 0.6212 | 0.8525 |
|
111 |
+
| 0.0734 | 41.9823 | 1186 | 0.5035 | 0.875 |
|
112 |
+
| 0.0643 | 42.9735 | 1214 | 0.5345 | 0.8775 |
|
113 |
+
| 0.0625 | 44.0 | 1243 | 0.5208 | 0.8625 |
|
114 |
+
| 0.047 | 44.9912 | 1271 | 0.5635 | 0.8675 |
|
115 |
+
| 0.0612 | 45.9823 | 1299 | 0.4721 | 0.8775 |
|
116 |
+
| 0.0582 | 46.9735 | 1327 | 0.5683 | 0.855 |
|
117 |
+
| 0.0516 | 48.0 | 1356 | 0.5883 | 0.8625 |
|
118 |
+
| 0.0427 | 48.9912 | 1384 | 0.5757 | 0.8575 |
|
119 |
+
| 0.0601 | 49.9823 | 1412 | 0.5368 | 0.8625 |
|
120 |
+
| 0.0645 | 50.9735 | 1440 | 0.5608 | 0.84 |
|
121 |
+
| 0.054 | 52.0 | 1469 | 0.5380 | 0.87 |
|
122 |
+
| 0.0647 | 52.9912 | 1497 | 0.5490 | 0.8625 |
|
123 |
+
| 0.0539 | 53.9823 | 1525 | 0.5686 | 0.8625 |
|
124 |
+
| 0.0485 | 54.9735 | 1553 | 0.5474 | 0.8725 |
|
125 |
+
| 0.0649 | 56.0 | 1582 | 0.5938 | 0.86 |
|
126 |
+
| 0.0486 | 56.9912 | 1610 | 0.5642 | 0.86 |
|
127 |
+
| 0.0385 | 57.9823 | 1638 | 0.5390 | 0.8675 |
|
128 |
+
| 0.0404 | 58.9735 | 1666 | 0.5735 | 0.8775 |
|
129 |
+
| 0.0543 | 60.0 | 1695 | 0.5117 | 0.875 |
|
130 |
+
| 0.0506 | 60.9912 | 1723 | 0.5422 | 0.8725 |
|
131 |
+
| 0.0398 | 61.9823 | 1751 | 0.5473 | 0.87 |
|
132 |
+
| 0.0494 | 62.9735 | 1779 | 0.5333 | 0.8675 |
|
133 |
+
| 0.0472 | 64.0 | 1808 | 0.5650 | 0.8825 |
|
134 |
+
| 0.0504 | 64.9912 | 1836 | 0.5771 | 0.8575 |
|
135 |
+
| 0.044 | 65.9823 | 1864 | 0.5220 | 0.86 |
|
136 |
+
| 0.061 | 66.9735 | 1892 | 0.5622 | 0.8725 |
|
137 |
+
| 0.0459 | 68.0 | 1921 | 0.5864 | 0.8625 |
|
138 |
+
| 0.0294 | 68.9912 | 1949 | 0.6341 | 0.8625 |
|
139 |
+
| 0.0428 | 69.9823 | 1977 | 0.5696 | 0.8675 |
|
140 |
+
| 0.0317 | 70.9735 | 2005 | 0.6313 | 0.845 |
|
141 |
+
| 0.0453 | 72.0 | 2034 | 0.5955 | 0.875 |
|
142 |
+
| 0.0592 | 72.9912 | 2062 | 0.5844 | 0.8675 |
|
143 |
+
| 0.0408 | 73.9823 | 2090 | 0.5868 | 0.86 |
|
144 |
+
| 0.0358 | 74.9735 | 2118 | 0.6115 | 0.85 |
|
145 |
+
| 0.0412 | 76.0 | 2147 | 0.5940 | 0.865 |
|
146 |
+
| 0.0323 | 76.9912 | 2175 | 0.5752 | 0.8625 |
|
147 |
+
| 0.0378 | 77.9823 | 2203 | 0.5515 | 0.8725 |
|
148 |
+
| 0.0359 | 78.9735 | 2231 | 0.5910 | 0.8775 |
|
149 |
+
| 0.028 | 80.0 | 2260 | 0.6060 | 0.8725 |
|
150 |
+
| 0.032 | 80.9912 | 2288 | 0.6054 | 0.8775 |
|
151 |
+
| 0.032 | 81.9823 | 2316 | 0.6312 | 0.8725 |
|
152 |
+
| 0.0228 | 82.9735 | 2344 | 0.6153 | 0.87 |
|
153 |
+
| 0.0457 | 84.0 | 2373 | 0.6443 | 0.86 |
|
154 |
+
| 0.0248 | 84.9912 | 2401 | 0.5726 | 0.875 |
|
155 |
+
| 0.0405 | 85.9823 | 2429 | 0.6042 | 0.875 |
|
156 |
+
| 0.0203 | 86.9735 | 2457 | 0.6107 | 0.87 |
|
157 |
+
| 0.0557 | 88.0 | 2486 | 0.5890 | 0.88 |
|
158 |
+
| 0.0302 | 88.9912 | 2514 | 0.5778 | 0.8625 |
|
159 |
+
| 0.0268 | 89.9823 | 2542 | 0.6039 | 0.8625 |
|
160 |
+
| 0.0313 | 90.9735 | 2570 | 0.5608 | 0.885 |
|
161 |
+
| 0.0227 | 92.0 | 2599 | 0.6019 | 0.8625 |
|
162 |
+
| 0.0277 | 92.9912 | 2627 | 0.5949 | 0.8675 |
|
163 |
+
| 0.0378 | 93.9823 | 2655 | 0.5785 | 0.875 |
|
164 |
+
| 0.0381 | 94.9735 | 2683 | 0.5646 | 0.8825 |
|
165 |
+
| 0.0435 | 96.0 | 2712 | 0.5513 | 0.88 |
|
166 |
+
| 0.0264 | 96.9912 | 2740 | 0.5257 | 0.875 |
|
167 |
+
| 0.0362 | 97.9823 | 2768 | 0.5332 | 0.8825 |
|
168 |
+
| 0.0209 | 98.9735 | 2796 | 0.5777 | 0.855 |
|
169 |
+
| 0.0348 | 100.0 | 2825 | 0.5674 | 0.8675 |
|
170 |
+
| 0.02 | 100.9912 | 2853 | 0.5744 | 0.8625 |
|
171 |
+
| 0.0092 | 101.9823 | 2881 | 0.5852 | 0.8675 |
|
172 |
+
| 0.0343 | 102.9735 | 2909 | 0.5856 | 0.8675 |
|
173 |
+
| 0.0185 | 104.0 | 2938 | 0.5670 | 0.88 |
|
174 |
+
| 0.0198 | 104.9912 | 2966 | 0.5612 | 0.8775 |
|
175 |
+
| 0.016 | 105.9823 | 2994 | 0.5701 | 0.88 |
|
176 |
+
| 0.0369 | 106.9735 | 3022 | 0.5791 | 0.8825 |
|
177 |
+
| 0.0357 | 108.0 | 3051 | 0.5730 | 0.8725 |
|
178 |
+
| 0.0361 | 108.9912 | 3079 | 0.5627 | 0.8725 |
|
179 |
+
| 0.0438 | 109.9823 | 3107 | 0.5812 | 0.875 |
|
180 |
+
| 0.0243 | 110.9735 | 3135 | 0.5922 | 0.8725 |
|
181 |
+
| 0.0241 | 112.0 | 3164 | 0.5913 | 0.8775 |
|
182 |
+
| 0.0256 | 112.9912 | 3192 | 0.5862 | 0.8675 |
|
183 |
+
| 0.0247 | 113.9823 | 3220 | 0.5813 | 0.8675 |
|
184 |
+
| 0.028 | 114.9735 | 3248 | 0.5752 | 0.87 |
|
185 |
+
| 0.0177 | 116.0 | 3277 | 0.5742 | 0.87 |
|
186 |
+
| 0.0255 | 116.9912 | 3305 | 0.5795 | 0.87 |
|
187 |
+
| 0.0174 | 117.9823 | 3333 | 0.5803 | 0.875 |
|
188 |
+
| 0.0225 | 118.9381 | 3360 | 0.5797 | 0.875 |
|
189 |
+
|
190 |
+
|
191 |
+
### Framework versions
|
192 |
+
|
193 |
+
- Transformers 4.44.0
|
194 |
+
- Pytorch 2.4.0
|
195 |
+
- Datasets 2.21.0
|
196 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 118.93805309734513,
|
3 |
+
"eval_accuracy": 0.875,
|
4 |
+
"eval_loss": 0.5796830654144287,
|
5 |
+
"eval_runtime": 6.4807,
|
6 |
+
"eval_samples_per_second": 61.722,
|
7 |
+
"eval_steps_per_second": 2.006,
|
8 |
+
"total_flos": 1.0779764781475824e+19,
|
9 |
+
"train_loss": 0.10835168258996591,
|
10 |
+
"train_runtime": 10219.331,
|
11 |
+
"train_samples_per_second": 42.273,
|
12 |
+
"train_steps_per_second": 0.329
|
13 |
+
}
|
config.json
ADDED
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/convnextv2-tiny-1k-224",
|
3 |
+
"architectures": [
|
4 |
+
"ConvNextV2ForImageClassification"
|
5 |
+
],
|
6 |
+
"depths": [
|
7 |
+
3,
|
8 |
+
3,
|
9 |
+
9,
|
10 |
+
3
|
11 |
+
],
|
12 |
+
"drop_path_rate": 0.0,
|
13 |
+
"hidden_act": "gelu",
|
14 |
+
"hidden_sizes": [
|
15 |
+
96,
|
16 |
+
192,
|
17 |
+
384,
|
18 |
+
768
|
19 |
+
],
|
20 |
+
"id2label": {
|
21 |
+
"0": "Abstract_Pattern",
|
22 |
+
"1": "Check_Pattern",
|
23 |
+
"2": "Color_Block_Pattern",
|
24 |
+
"3": "Embellished_Pattern",
|
25 |
+
"4": "Embroider_Pattern",
|
26 |
+
"5": "Geometric_pattern",
|
27 |
+
"6": "Graphic_Pattern",
|
28 |
+
"7": "Lace_Pattern",
|
29 |
+
"8": "Polka_Dot_Pattern",
|
30 |
+
"9": "Stripes_Pattern"
|
31 |
+
},
|
32 |
+
"image_size": 224,
|
33 |
+
"initializer_range": 0.02,
|
34 |
+
"label2id": {
|
35 |
+
"Abstract_Pattern": 0,
|
36 |
+
"Check_Pattern": 1,
|
37 |
+
"Color_Block_Pattern": 2,
|
38 |
+
"Embellished_Pattern": 3,
|
39 |
+
"Embroider_Pattern": 4,
|
40 |
+
"Geometric_pattern": 5,
|
41 |
+
"Graphic_Pattern": 6,
|
42 |
+
"Lace_Pattern": 7,
|
43 |
+
"Polka_Dot_Pattern": 8,
|
44 |
+
"Stripes_Pattern": 9
|
45 |
+
},
|
46 |
+
"layer_norm_eps": 1e-12,
|
47 |
+
"model_type": "convnextv2",
|
48 |
+
"num_channels": 3,
|
49 |
+
"num_stages": 4,
|
50 |
+
"out_features": [
|
51 |
+
"stage4"
|
52 |
+
],
|
53 |
+
"out_indices": [
|
54 |
+
4
|
55 |
+
],
|
56 |
+
"patch_size": 4,
|
57 |
+
"problem_type": "single_label_classification",
|
58 |
+
"stage_names": [
|
59 |
+
"stem",
|
60 |
+
"stage1",
|
61 |
+
"stage2",
|
62 |
+
"stage3",
|
63 |
+
"stage4"
|
64 |
+
],
|
65 |
+
"torch_dtype": "float32",
|
66 |
+
"transformers_version": "4.44.0"
|
67 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 118.93805309734513,
|
3 |
+
"eval_accuracy": 0.875,
|
4 |
+
"eval_loss": 0.5796830654144287,
|
5 |
+
"eval_runtime": 6.4807,
|
6 |
+
"eval_samples_per_second": 61.722,
|
7 |
+
"eval_steps_per_second": 2.006
|
8 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9f7b9b0fd9beff1bfedf4c704237acd2522990e8dcd40a2d9d5f06a39844fc10
|
3 |
+
size 111520440
|
preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"crop_pct": 0.875,
|
3 |
+
"do_normalize": true,
|
4 |
+
"do_rescale": true,
|
5 |
+
"do_resize": true,
|
6 |
+
"image_mean": [
|
7 |
+
0.485,
|
8 |
+
0.456,
|
9 |
+
0.406
|
10 |
+
],
|
11 |
+
"image_processor_type": "ConvNextImageProcessor",
|
12 |
+
"image_std": [
|
13 |
+
0.229,
|
14 |
+
0.224,
|
15 |
+
0.225
|
16 |
+
],
|
17 |
+
"resample": 3,
|
18 |
+
"rescale_factor": 0.00392156862745098,
|
19 |
+
"size": {
|
20 |
+
"shortest_edge": 224
|
21 |
+
}
|
22 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 118.93805309734513,
|
3 |
+
"total_flos": 1.0779764781475824e+19,
|
4 |
+
"train_loss": 0.10835168258996591,
|
5 |
+
"train_runtime": 10219.331,
|
6 |
+
"train_samples_per_second": 42.273,
|
7 |
+
"train_steps_per_second": 0.329
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,3465 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 118.93805309734513,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 3360,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.35398230088495575,
|
13 |
+
"grad_norm": 38.51472473144531,
|
14 |
+
"learning_rate": 1.4880952380952381e-06,
|
15 |
+
"loss": 1.7357,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.7079646017699115,
|
20 |
+
"grad_norm": 38.40711212158203,
|
21 |
+
"learning_rate": 2.9761904761904763e-06,
|
22 |
+
"loss": 1.6954,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.9911504424778761,
|
27 |
+
"eval_accuracy": 0.58,
|
28 |
+
"eval_loss": 1.4482159614562988,
|
29 |
+
"eval_runtime": 8.0113,
|
30 |
+
"eval_samples_per_second": 49.929,
|
31 |
+
"eval_steps_per_second": 1.623,
|
32 |
+
"step": 28
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.0619469026548674,
|
36 |
+
"grad_norm": 32.50920867919922,
|
37 |
+
"learning_rate": 4.464285714285715e-06,
|
38 |
+
"loss": 1.4138,
|
39 |
+
"step": 30
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 1.415929203539823,
|
43 |
+
"grad_norm": 39.59659957885742,
|
44 |
+
"learning_rate": 5.9523809523809525e-06,
|
45 |
+
"loss": 1.1465,
|
46 |
+
"step": 40
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 1.7699115044247788,
|
50 |
+
"grad_norm": 26.62055778503418,
|
51 |
+
"learning_rate": 7.4404761904761905e-06,
|
52 |
+
"loss": 1.015,
|
53 |
+
"step": 50
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 1.9823008849557522,
|
57 |
+
"eval_accuracy": 0.7175,
|
58 |
+
"eval_loss": 0.9088323712348938,
|
59 |
+
"eval_runtime": 6.5891,
|
60 |
+
"eval_samples_per_second": 60.707,
|
61 |
+
"eval_steps_per_second": 1.973,
|
62 |
+
"step": 56
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.1238938053097347,
|
66 |
+
"grad_norm": 29.730579376220703,
|
67 |
+
"learning_rate": 8.92857142857143e-06,
|
68 |
+
"loss": 0.9611,
|
69 |
+
"step": 60
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 2.47787610619469,
|
73 |
+
"grad_norm": 30.00710678100586,
|
74 |
+
"learning_rate": 1.0416666666666668e-05,
|
75 |
+
"loss": 0.8009,
|
76 |
+
"step": 70
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 2.831858407079646,
|
80 |
+
"grad_norm": 21.186767578125,
|
81 |
+
"learning_rate": 1.1904761904761905e-05,
|
82 |
+
"loss": 0.7953,
|
83 |
+
"step": 80
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 2.9734513274336285,
|
87 |
+
"eval_accuracy": 0.7625,
|
88 |
+
"eval_loss": 0.7266157269477844,
|
89 |
+
"eval_runtime": 6.5536,
|
90 |
+
"eval_samples_per_second": 61.035,
|
91 |
+
"eval_steps_per_second": 1.984,
|
92 |
+
"step": 84
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 3.185840707964602,
|
96 |
+
"grad_norm": 32.94711685180664,
|
97 |
+
"learning_rate": 1.3392857142857144e-05,
|
98 |
+
"loss": 0.6467,
|
99 |
+
"step": 90
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 3.5398230088495577,
|
103 |
+
"grad_norm": 19.001646041870117,
|
104 |
+
"learning_rate": 1.4880952380952381e-05,
|
105 |
+
"loss": 0.6282,
|
106 |
+
"step": 100
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 3.893805309734513,
|
110 |
+
"grad_norm": 36.843204498291016,
|
111 |
+
"learning_rate": 1.636904761904762e-05,
|
112 |
+
"loss": 0.627,
|
113 |
+
"step": 110
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 4.0,
|
117 |
+
"eval_accuracy": 0.8,
|
118 |
+
"eval_loss": 0.5872101783752441,
|
119 |
+
"eval_runtime": 6.7124,
|
120 |
+
"eval_samples_per_second": 59.591,
|
121 |
+
"eval_steps_per_second": 1.937,
|
122 |
+
"step": 113
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 4.247787610619469,
|
126 |
+
"grad_norm": 16.53933334350586,
|
127 |
+
"learning_rate": 1.785714285714286e-05,
|
128 |
+
"loss": 0.5658,
|
129 |
+
"step": 120
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 4.601769911504425,
|
133 |
+
"grad_norm": 47.344696044921875,
|
134 |
+
"learning_rate": 1.9345238095238097e-05,
|
135 |
+
"loss": 0.5377,
|
136 |
+
"step": 130
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 4.95575221238938,
|
140 |
+
"grad_norm": 20.771940231323242,
|
141 |
+
"learning_rate": 2.0833333333333336e-05,
|
142 |
+
"loss": 0.4684,
|
143 |
+
"step": 140
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 4.991150442477876,
|
147 |
+
"eval_accuracy": 0.8175,
|
148 |
+
"eval_loss": 0.5533822774887085,
|
149 |
+
"eval_runtime": 6.5588,
|
150 |
+
"eval_samples_per_second": 60.987,
|
151 |
+
"eval_steps_per_second": 1.982,
|
152 |
+
"step": 141
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 5.3097345132743365,
|
156 |
+
"grad_norm": 21.18557357788086,
|
157 |
+
"learning_rate": 2.2321428571428575e-05,
|
158 |
+
"loss": 0.4512,
|
159 |
+
"step": 150
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 5.663716814159292,
|
163 |
+
"grad_norm": 23.0177059173584,
|
164 |
+
"learning_rate": 2.380952380952381e-05,
|
165 |
+
"loss": 0.4301,
|
166 |
+
"step": 160
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 5.982300884955752,
|
170 |
+
"eval_accuracy": 0.8275,
|
171 |
+
"eval_loss": 0.5052626132965088,
|
172 |
+
"eval_runtime": 6.4246,
|
173 |
+
"eval_samples_per_second": 62.261,
|
174 |
+
"eval_steps_per_second": 2.023,
|
175 |
+
"step": 169
|
176 |
+
},
|
177 |
+
{
|
178 |
+
"epoch": 6.017699115044247,
|
179 |
+
"grad_norm": 30.39218521118164,
|
180 |
+
"learning_rate": 2.529761904761905e-05,
|
181 |
+
"loss": 0.4105,
|
182 |
+
"step": 170
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 6.371681415929204,
|
186 |
+
"grad_norm": 19.97724723815918,
|
187 |
+
"learning_rate": 2.6785714285714288e-05,
|
188 |
+
"loss": 0.3878,
|
189 |
+
"step": 180
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 6.725663716814159,
|
193 |
+
"grad_norm": 29.496784210205078,
|
194 |
+
"learning_rate": 2.8273809523809523e-05,
|
195 |
+
"loss": 0.3716,
|
196 |
+
"step": 190
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 6.9734513274336285,
|
200 |
+
"eval_accuracy": 0.83,
|
201 |
+
"eval_loss": 0.4885025918483734,
|
202 |
+
"eval_runtime": 6.637,
|
203 |
+
"eval_samples_per_second": 60.268,
|
204 |
+
"eval_steps_per_second": 1.959,
|
205 |
+
"step": 197
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 7.079646017699115,
|
209 |
+
"grad_norm": 45.38138961791992,
|
210 |
+
"learning_rate": 2.9761904761904762e-05,
|
211 |
+
"loss": 0.3787,
|
212 |
+
"step": 200
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 7.433628318584071,
|
216 |
+
"grad_norm": 40.39886474609375,
|
217 |
+
"learning_rate": 3.125e-05,
|
218 |
+
"loss": 0.3277,
|
219 |
+
"step": 210
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 7.787610619469026,
|
223 |
+
"grad_norm": 16.372304916381836,
|
224 |
+
"learning_rate": 3.273809523809524e-05,
|
225 |
+
"loss": 0.3798,
|
226 |
+
"step": 220
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 8.0,
|
230 |
+
"eval_accuracy": 0.8525,
|
231 |
+
"eval_loss": 0.46391868591308594,
|
232 |
+
"eval_runtime": 6.339,
|
233 |
+
"eval_samples_per_second": 63.102,
|
234 |
+
"eval_steps_per_second": 2.051,
|
235 |
+
"step": 226
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 8.141592920353983,
|
239 |
+
"grad_norm": 17.45707130432129,
|
240 |
+
"learning_rate": 3.422619047619048e-05,
|
241 |
+
"loss": 0.2878,
|
242 |
+
"step": 230
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 8.495575221238939,
|
246 |
+
"grad_norm": 18.687721252441406,
|
247 |
+
"learning_rate": 3.571428571428572e-05,
|
248 |
+
"loss": 0.3108,
|
249 |
+
"step": 240
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 8.849557522123893,
|
253 |
+
"grad_norm": 16.21880531311035,
|
254 |
+
"learning_rate": 3.7202380952380956e-05,
|
255 |
+
"loss": 0.3123,
|
256 |
+
"step": 250
|
257 |
+
},
|
258 |
+
{
|
259 |
+
"epoch": 8.991150442477876,
|
260 |
+
"eval_accuracy": 0.825,
|
261 |
+
"eval_loss": 0.5282442569732666,
|
262 |
+
"eval_runtime": 6.6933,
|
263 |
+
"eval_samples_per_second": 59.761,
|
264 |
+
"eval_steps_per_second": 1.942,
|
265 |
+
"step": 254
|
266 |
+
},
|
267 |
+
{
|
268 |
+
"epoch": 9.20353982300885,
|
269 |
+
"grad_norm": 15.07292366027832,
|
270 |
+
"learning_rate": 3.8690476190476195e-05,
|
271 |
+
"loss": 0.2495,
|
272 |
+
"step": 260
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 9.557522123893806,
|
276 |
+
"grad_norm": 20.118724822998047,
|
277 |
+
"learning_rate": 4.017857142857143e-05,
|
278 |
+
"loss": 0.2457,
|
279 |
+
"step": 270
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 9.91150442477876,
|
283 |
+
"grad_norm": 14.054291725158691,
|
284 |
+
"learning_rate": 4.166666666666667e-05,
|
285 |
+
"loss": 0.3148,
|
286 |
+
"step": 280
|
287 |
+
},
|
288 |
+
{
|
289 |
+
"epoch": 9.982300884955752,
|
290 |
+
"eval_accuracy": 0.8475,
|
291 |
+
"eval_loss": 0.456900417804718,
|
292 |
+
"eval_runtime": 6.5738,
|
293 |
+
"eval_samples_per_second": 60.847,
|
294 |
+
"eval_steps_per_second": 1.978,
|
295 |
+
"step": 282
|
296 |
+
},
|
297 |
+
{
|
298 |
+
"epoch": 10.265486725663717,
|
299 |
+
"grad_norm": 36.96604919433594,
|
300 |
+
"learning_rate": 4.315476190476191e-05,
|
301 |
+
"loss": 0.2527,
|
302 |
+
"step": 290
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 10.619469026548673,
|
306 |
+
"grad_norm": 17.349655151367188,
|
307 |
+
"learning_rate": 4.464285714285715e-05,
|
308 |
+
"loss": 0.2425,
|
309 |
+
"step": 300
|
310 |
+
},
|
311 |
+
{
|
312 |
+
"epoch": 10.973451327433628,
|
313 |
+
"grad_norm": 14.201617240905762,
|
314 |
+
"learning_rate": 4.613095238095239e-05,
|
315 |
+
"loss": 0.2427,
|
316 |
+
"step": 310
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 10.973451327433628,
|
320 |
+
"eval_accuracy": 0.865,
|
321 |
+
"eval_loss": 0.42058929800987244,
|
322 |
+
"eval_runtime": 6.3979,
|
323 |
+
"eval_samples_per_second": 62.521,
|
324 |
+
"eval_steps_per_second": 2.032,
|
325 |
+
"step": 310
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 11.327433628318584,
|
329 |
+
"grad_norm": 13.164992332458496,
|
330 |
+
"learning_rate": 4.761904761904762e-05,
|
331 |
+
"loss": 0.2011,
|
332 |
+
"step": 320
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 11.68141592920354,
|
336 |
+
"grad_norm": 23.505887985229492,
|
337 |
+
"learning_rate": 4.910714285714286e-05,
|
338 |
+
"loss": 0.2198,
|
339 |
+
"step": 330
|
340 |
+
},
|
341 |
+
{
|
342 |
+
"epoch": 12.0,
|
343 |
+
"eval_accuracy": 0.84,
|
344 |
+
"eval_loss": 0.48323675990104675,
|
345 |
+
"eval_runtime": 6.7363,
|
346 |
+
"eval_samples_per_second": 59.38,
|
347 |
+
"eval_steps_per_second": 1.93,
|
348 |
+
"step": 339
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 12.035398230088495,
|
352 |
+
"grad_norm": 12.627903938293457,
|
353 |
+
"learning_rate": 4.993386243386244e-05,
|
354 |
+
"loss": 0.2127,
|
355 |
+
"step": 340
|
356 |
+
},
|
357 |
+
{
|
358 |
+
"epoch": 12.389380530973451,
|
359 |
+
"grad_norm": 10.761627197265625,
|
360 |
+
"learning_rate": 4.976851851851852e-05,
|
361 |
+
"loss": 0.2224,
|
362 |
+
"step": 350
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 12.743362831858407,
|
366 |
+
"grad_norm": 33.78993225097656,
|
367 |
+
"learning_rate": 4.960317460317461e-05,
|
368 |
+
"loss": 0.1995,
|
369 |
+
"step": 360
|
370 |
+
},
|
371 |
+
{
|
372 |
+
"epoch": 12.991150442477876,
|
373 |
+
"eval_accuracy": 0.865,
|
374 |
+
"eval_loss": 0.44683846831321716,
|
375 |
+
"eval_runtime": 6.6173,
|
376 |
+
"eval_samples_per_second": 60.448,
|
377 |
+
"eval_steps_per_second": 1.965,
|
378 |
+
"step": 367
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 13.097345132743364,
|
382 |
+
"grad_norm": 10.526817321777344,
|
383 |
+
"learning_rate": 4.943783068783069e-05,
|
384 |
+
"loss": 0.1715,
|
385 |
+
"step": 370
|
386 |
+
},
|
387 |
+
{
|
388 |
+
"epoch": 13.451327433628318,
|
389 |
+
"grad_norm": 15.434138298034668,
|
390 |
+
"learning_rate": 4.927248677248678e-05,
|
391 |
+
"loss": 0.1718,
|
392 |
+
"step": 380
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 13.805309734513274,
|
396 |
+
"grad_norm": 9.151437759399414,
|
397 |
+
"learning_rate": 4.910714285714286e-05,
|
398 |
+
"loss": 0.1738,
|
399 |
+
"step": 390
|
400 |
+
},
|
401 |
+
{
|
402 |
+
"epoch": 13.982300884955752,
|
403 |
+
"eval_accuracy": 0.8425,
|
404 |
+
"eval_loss": 0.5668337345123291,
|
405 |
+
"eval_runtime": 6.6205,
|
406 |
+
"eval_samples_per_second": 60.418,
|
407 |
+
"eval_steps_per_second": 1.964,
|
408 |
+
"step": 395
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 14.15929203539823,
|
412 |
+
"grad_norm": 11.179224014282227,
|
413 |
+
"learning_rate": 4.894179894179895e-05,
|
414 |
+
"loss": 0.1821,
|
415 |
+
"step": 400
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 14.513274336283185,
|
419 |
+
"grad_norm": 13.600383758544922,
|
420 |
+
"learning_rate": 4.8776455026455034e-05,
|
421 |
+
"loss": 0.1875,
|
422 |
+
"step": 410
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 14.867256637168142,
|
426 |
+
"grad_norm": 12.053109169006348,
|
427 |
+
"learning_rate": 4.8611111111111115e-05,
|
428 |
+
"loss": 0.1683,
|
429 |
+
"step": 420
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 14.973451327433628,
|
433 |
+
"eval_accuracy": 0.8725,
|
434 |
+
"eval_loss": 0.4454367756843567,
|
435 |
+
"eval_runtime": 6.5474,
|
436 |
+
"eval_samples_per_second": 61.093,
|
437 |
+
"eval_steps_per_second": 1.986,
|
438 |
+
"step": 423
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 15.221238938053098,
|
442 |
+
"grad_norm": 16.21660041809082,
|
443 |
+
"learning_rate": 4.84457671957672e-05,
|
444 |
+
"loss": 0.1299,
|
445 |
+
"step": 430
|
446 |
+
},
|
447 |
+
{
|
448 |
+
"epoch": 15.575221238938052,
|
449 |
+
"grad_norm": 14.533129692077637,
|
450 |
+
"learning_rate": 4.8280423280423284e-05,
|
451 |
+
"loss": 0.1353,
|
452 |
+
"step": 440
|
453 |
+
},
|
454 |
+
{
|
455 |
+
"epoch": 15.929203539823009,
|
456 |
+
"grad_norm": 24.27181625366211,
|
457 |
+
"learning_rate": 4.811507936507937e-05,
|
458 |
+
"loss": 0.1426,
|
459 |
+
"step": 450
|
460 |
+
},
|
461 |
+
{
|
462 |
+
"epoch": 16.0,
|
463 |
+
"eval_accuracy": 0.8525,
|
464 |
+
"eval_loss": 0.5118055939674377,
|
465 |
+
"eval_runtime": 6.8317,
|
466 |
+
"eval_samples_per_second": 58.551,
|
467 |
+
"eval_steps_per_second": 1.903,
|
468 |
+
"step": 452
|
469 |
+
},
|
470 |
+
{
|
471 |
+
"epoch": 16.283185840707965,
|
472 |
+
"grad_norm": 18.932626724243164,
|
473 |
+
"learning_rate": 4.794973544973545e-05,
|
474 |
+
"loss": 0.1276,
|
475 |
+
"step": 460
|
476 |
+
},
|
477 |
+
{
|
478 |
+
"epoch": 16.63716814159292,
|
479 |
+
"grad_norm": 13.03187084197998,
|
480 |
+
"learning_rate": 4.778439153439154e-05,
|
481 |
+
"loss": 0.1468,
|
482 |
+
"step": 470
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 16.991150442477878,
|
486 |
+
"grad_norm": 42.71265411376953,
|
487 |
+
"learning_rate": 4.761904761904762e-05,
|
488 |
+
"loss": 0.133,
|
489 |
+
"step": 480
|
490 |
+
},
|
491 |
+
{
|
492 |
+
"epoch": 16.991150442477878,
|
493 |
+
"eval_accuracy": 0.865,
|
494 |
+
"eval_loss": 0.4713393449783325,
|
495 |
+
"eval_runtime": 6.4565,
|
496 |
+
"eval_samples_per_second": 61.953,
|
497 |
+
"eval_steps_per_second": 2.013,
|
498 |
+
"step": 480
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 17.345132743362832,
|
502 |
+
"grad_norm": 8.917379379272461,
|
503 |
+
"learning_rate": 4.745370370370371e-05,
|
504 |
+
"loss": 0.1541,
|
505 |
+
"step": 490
|
506 |
+
},
|
507 |
+
{
|
508 |
+
"epoch": 17.699115044247787,
|
509 |
+
"grad_norm": 16.360355377197266,
|
510 |
+
"learning_rate": 4.7288359788359796e-05,
|
511 |
+
"loss": 0.1148,
|
512 |
+
"step": 500
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 17.98230088495575,
|
516 |
+
"eval_accuracy": 0.855,
|
517 |
+
"eval_loss": 0.5225958824157715,
|
518 |
+
"eval_runtime": 7.0446,
|
519 |
+
"eval_samples_per_second": 56.781,
|
520 |
+
"eval_steps_per_second": 1.845,
|
521 |
+
"step": 508
|
522 |
+
},
|
523 |
+
{
|
524 |
+
"epoch": 18.053097345132745,
|
525 |
+
"grad_norm": 14.729361534118652,
|
526 |
+
"learning_rate": 4.7123015873015876e-05,
|
527 |
+
"loss": 0.1592,
|
528 |
+
"step": 510
|
529 |
+
},
|
530 |
+
{
|
531 |
+
"epoch": 18.4070796460177,
|
532 |
+
"grad_norm": 8.965171813964844,
|
533 |
+
"learning_rate": 4.6957671957671964e-05,
|
534 |
+
"loss": 0.137,
|
535 |
+
"step": 520
|
536 |
+
},
|
537 |
+
{
|
538 |
+
"epoch": 18.761061946902654,
|
539 |
+
"grad_norm": 13.508126258850098,
|
540 |
+
"learning_rate": 4.6792328042328045e-05,
|
541 |
+
"loss": 0.1147,
|
542 |
+
"step": 530
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 18.97345132743363,
|
546 |
+
"eval_accuracy": 0.8425,
|
547 |
+
"eval_loss": 0.5332815051078796,
|
548 |
+
"eval_runtime": 6.5741,
|
549 |
+
"eval_samples_per_second": 60.844,
|
550 |
+
"eval_steps_per_second": 1.977,
|
551 |
+
"step": 536
|
552 |
+
},
|
553 |
+
{
|
554 |
+
"epoch": 19.115044247787612,
|
555 |
+
"grad_norm": 14.787744522094727,
|
556 |
+
"learning_rate": 4.662698412698413e-05,
|
557 |
+
"loss": 0.125,
|
558 |
+
"step": 540
|
559 |
+
},
|
560 |
+
{
|
561 |
+
"epoch": 19.469026548672566,
|
562 |
+
"grad_norm": 16.165084838867188,
|
563 |
+
"learning_rate": 4.646164021164021e-05,
|
564 |
+
"loss": 0.1157,
|
565 |
+
"step": 550
|
566 |
+
},
|
567 |
+
{
|
568 |
+
"epoch": 19.82300884955752,
|
569 |
+
"grad_norm": 17.71279525756836,
|
570 |
+
"learning_rate": 4.62962962962963e-05,
|
571 |
+
"loss": 0.1284,
|
572 |
+
"step": 560
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 20.0,
|
576 |
+
"eval_accuracy": 0.8575,
|
577 |
+
"eval_loss": 0.43993717432022095,
|
578 |
+
"eval_runtime": 6.5441,
|
579 |
+
"eval_samples_per_second": 61.123,
|
580 |
+
"eval_steps_per_second": 1.987,
|
581 |
+
"step": 565
|
582 |
+
},
|
583 |
+
{
|
584 |
+
"epoch": 20.17699115044248,
|
585 |
+
"grad_norm": 9.373887062072754,
|
586 |
+
"learning_rate": 4.613095238095239e-05,
|
587 |
+
"loss": 0.1113,
|
588 |
+
"step": 570
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 20.530973451327434,
|
592 |
+
"grad_norm": 21.481483459472656,
|
593 |
+
"learning_rate": 4.596560846560847e-05,
|
594 |
+
"loss": 0.1041,
|
595 |
+
"step": 580
|
596 |
+
},
|
597 |
+
{
|
598 |
+
"epoch": 20.884955752212388,
|
599 |
+
"grad_norm": 11.179847717285156,
|
600 |
+
"learning_rate": 4.580026455026456e-05,
|
601 |
+
"loss": 0.1035,
|
602 |
+
"step": 590
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 20.991150442477878,
|
606 |
+
"eval_accuracy": 0.8525,
|
607 |
+
"eval_loss": 0.5193604826927185,
|
608 |
+
"eval_runtime": 6.4413,
|
609 |
+
"eval_samples_per_second": 62.099,
|
610 |
+
"eval_steps_per_second": 2.018,
|
611 |
+
"step": 593
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 21.238938053097346,
|
615 |
+
"grad_norm": 11.193985939025879,
|
616 |
+
"learning_rate": 4.563492063492064e-05,
|
617 |
+
"loss": 0.0766,
|
618 |
+
"step": 600
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 21.5929203539823,
|
622 |
+
"grad_norm": 25.748003005981445,
|
623 |
+
"learning_rate": 4.5469576719576725e-05,
|
624 |
+
"loss": 0.1186,
|
625 |
+
"step": 610
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 21.946902654867255,
|
629 |
+
"grad_norm": 18.839889526367188,
|
630 |
+
"learning_rate": 4.5304232804232806e-05,
|
631 |
+
"loss": 0.1054,
|
632 |
+
"step": 620
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 21.98230088495575,
|
636 |
+
"eval_accuracy": 0.845,
|
637 |
+
"eval_loss": 0.5139681100845337,
|
638 |
+
"eval_runtime": 6.5429,
|
639 |
+
"eval_samples_per_second": 61.135,
|
640 |
+
"eval_steps_per_second": 1.987,
|
641 |
+
"step": 621
|
642 |
+
},
|
643 |
+
{
|
644 |
+
"epoch": 22.300884955752213,
|
645 |
+
"grad_norm": 23.48375701904297,
|
646 |
+
"learning_rate": 4.5138888888888894e-05,
|
647 |
+
"loss": 0.1108,
|
648 |
+
"step": 630
|
649 |
+
},
|
650 |
+
{
|
651 |
+
"epoch": 22.654867256637168,
|
652 |
+
"grad_norm": 14.143659591674805,
|
653 |
+
"learning_rate": 4.4973544973544974e-05,
|
654 |
+
"loss": 0.1056,
|
655 |
+
"step": 640
|
656 |
+
},
|
657 |
+
{
|
658 |
+
"epoch": 22.97345132743363,
|
659 |
+
"eval_accuracy": 0.87,
|
660 |
+
"eval_loss": 0.5183441638946533,
|
661 |
+
"eval_runtime": 6.505,
|
662 |
+
"eval_samples_per_second": 61.491,
|
663 |
+
"eval_steps_per_second": 1.998,
|
664 |
+
"step": 649
|
665 |
+
},
|
666 |
+
{
|
667 |
+
"epoch": 23.008849557522122,
|
668 |
+
"grad_norm": 18.987043380737305,
|
669 |
+
"learning_rate": 4.480820105820106e-05,
|
670 |
+
"loss": 0.1202,
|
671 |
+
"step": 650
|
672 |
+
},
|
673 |
+
{
|
674 |
+
"epoch": 23.36283185840708,
|
675 |
+
"grad_norm": 14.656883239746094,
|
676 |
+
"learning_rate": 4.464285714285715e-05,
|
677 |
+
"loss": 0.1047,
|
678 |
+
"step": 660
|
679 |
+
},
|
680 |
+
{
|
681 |
+
"epoch": 23.716814159292035,
|
682 |
+
"grad_norm": 9.784178733825684,
|
683 |
+
"learning_rate": 4.447751322751323e-05,
|
684 |
+
"loss": 0.1224,
|
685 |
+
"step": 670
|
686 |
+
},
|
687 |
+
{
|
688 |
+
"epoch": 24.0,
|
689 |
+
"eval_accuracy": 0.85,
|
690 |
+
"eval_loss": 0.5292716026306152,
|
691 |
+
"eval_runtime": 6.5244,
|
692 |
+
"eval_samples_per_second": 61.309,
|
693 |
+
"eval_steps_per_second": 1.993,
|
694 |
+
"step": 678
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 24.07079646017699,
|
698 |
+
"grad_norm": 6.749938011169434,
|
699 |
+
"learning_rate": 4.431216931216932e-05,
|
700 |
+
"loss": 0.0946,
|
701 |
+
"step": 680
|
702 |
+
},
|
703 |
+
{
|
704 |
+
"epoch": 24.424778761061948,
|
705 |
+
"grad_norm": 8.656756401062012,
|
706 |
+
"learning_rate": 4.41468253968254e-05,
|
707 |
+
"loss": 0.1011,
|
708 |
+
"step": 690
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 24.778761061946902,
|
712 |
+
"grad_norm": 12.24336051940918,
|
713 |
+
"learning_rate": 4.3981481481481486e-05,
|
714 |
+
"loss": 0.0956,
|
715 |
+
"step": 700
|
716 |
+
},
|
717 |
+
{
|
718 |
+
"epoch": 24.991150442477878,
|
719 |
+
"eval_accuracy": 0.87,
|
720 |
+
"eval_loss": 0.4984818398952484,
|
721 |
+
"eval_runtime": 6.8382,
|
722 |
+
"eval_samples_per_second": 58.495,
|
723 |
+
"eval_steps_per_second": 1.901,
|
724 |
+
"step": 706
|
725 |
+
},
|
726 |
+
{
|
727 |
+
"epoch": 25.13274336283186,
|
728 |
+
"grad_norm": 11.6083345413208,
|
729 |
+
"learning_rate": 4.381613756613757e-05,
|
730 |
+
"loss": 0.1002,
|
731 |
+
"step": 710
|
732 |
+
},
|
733 |
+
{
|
734 |
+
"epoch": 25.486725663716815,
|
735 |
+
"grad_norm": 7.939414024353027,
|
736 |
+
"learning_rate": 4.3650793650793655e-05,
|
737 |
+
"loss": 0.0846,
|
738 |
+
"step": 720
|
739 |
+
},
|
740 |
+
{
|
741 |
+
"epoch": 25.84070796460177,
|
742 |
+
"grad_norm": 17.587238311767578,
|
743 |
+
"learning_rate": 4.3485449735449736e-05,
|
744 |
+
"loss": 0.0717,
|
745 |
+
"step": 730
|
746 |
+
},
|
747 |
+
{
|
748 |
+
"epoch": 25.98230088495575,
|
749 |
+
"eval_accuracy": 0.8625,
|
750 |
+
"eval_loss": 0.5267429351806641,
|
751 |
+
"eval_runtime": 6.7375,
|
752 |
+
"eval_samples_per_second": 59.369,
|
753 |
+
"eval_steps_per_second": 1.93,
|
754 |
+
"step": 734
|
755 |
+
},
|
756 |
+
{
|
757 |
+
"epoch": 26.194690265486727,
|
758 |
+
"grad_norm": 6.991844177246094,
|
759 |
+
"learning_rate": 4.332010582010582e-05,
|
760 |
+
"loss": 0.0992,
|
761 |
+
"step": 740
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 26.548672566371682,
|
765 |
+
"grad_norm": 9.317466735839844,
|
766 |
+
"learning_rate": 4.315476190476191e-05,
|
767 |
+
"loss": 0.1031,
|
768 |
+
"step": 750
|
769 |
+
},
|
770 |
+
{
|
771 |
+
"epoch": 26.902654867256636,
|
772 |
+
"grad_norm": 10.530389785766602,
|
773 |
+
"learning_rate": 4.298941798941799e-05,
|
774 |
+
"loss": 0.0858,
|
775 |
+
"step": 760
|
776 |
+
},
|
777 |
+
{
|
778 |
+
"epoch": 26.97345132743363,
|
779 |
+
"eval_accuracy": 0.8575,
|
780 |
+
"eval_loss": 0.5524987578392029,
|
781 |
+
"eval_runtime": 6.5345,
|
782 |
+
"eval_samples_per_second": 61.214,
|
783 |
+
"eval_steps_per_second": 1.989,
|
784 |
+
"step": 762
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 27.256637168141594,
|
788 |
+
"grad_norm": 8.615869522094727,
|
789 |
+
"learning_rate": 4.282407407407408e-05,
|
790 |
+
"loss": 0.0926,
|
791 |
+
"step": 770
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 27.61061946902655,
|
795 |
+
"grad_norm": 12.212360382080078,
|
796 |
+
"learning_rate": 4.265873015873016e-05,
|
797 |
+
"loss": 0.0722,
|
798 |
+
"step": 780
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 27.964601769911503,
|
802 |
+
"grad_norm": 5.2418084144592285,
|
803 |
+
"learning_rate": 4.249338624338625e-05,
|
804 |
+
"loss": 0.097,
|
805 |
+
"step": 790
|
806 |
+
},
|
807 |
+
{
|
808 |
+
"epoch": 28.0,
|
809 |
+
"eval_accuracy": 0.855,
|
810 |
+
"eval_loss": 0.5340389609336853,
|
811 |
+
"eval_runtime": 6.686,
|
812 |
+
"eval_samples_per_second": 59.827,
|
813 |
+
"eval_steps_per_second": 1.944,
|
814 |
+
"step": 791
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 28.31858407079646,
|
818 |
+
"grad_norm": 14.163456916809082,
|
819 |
+
"learning_rate": 4.232804232804233e-05,
|
820 |
+
"loss": 0.0899,
|
821 |
+
"step": 800
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 28.672566371681416,
|
825 |
+
"grad_norm": 5.519562244415283,
|
826 |
+
"learning_rate": 4.2162698412698416e-05,
|
827 |
+
"loss": 0.0914,
|
828 |
+
"step": 810
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 28.991150442477878,
|
832 |
+
"eval_accuracy": 0.87,
|
833 |
+
"eval_loss": 0.4829975962638855,
|
834 |
+
"eval_runtime": 6.7064,
|
835 |
+
"eval_samples_per_second": 59.644,
|
836 |
+
"eval_steps_per_second": 1.938,
|
837 |
+
"step": 819
|
838 |
+
},
|
839 |
+
{
|
840 |
+
"epoch": 29.02654867256637,
|
841 |
+
"grad_norm": 9.859421730041504,
|
842 |
+
"learning_rate": 4.1997354497354504e-05,
|
843 |
+
"loss": 0.0763,
|
844 |
+
"step": 820
|
845 |
+
},
|
846 |
+
{
|
847 |
+
"epoch": 29.38053097345133,
|
848 |
+
"grad_norm": 8.416680335998535,
|
849 |
+
"learning_rate": 4.1832010582010584e-05,
|
850 |
+
"loss": 0.0776,
|
851 |
+
"step": 830
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 29.734513274336283,
|
855 |
+
"grad_norm": 7.984038829803467,
|
856 |
+
"learning_rate": 4.166666666666667e-05,
|
857 |
+
"loss": 0.0699,
|
858 |
+
"step": 840
|
859 |
+
},
|
860 |
+
{
|
861 |
+
"epoch": 29.98230088495575,
|
862 |
+
"eval_accuracy": 0.8725,
|
863 |
+
"eval_loss": 0.4882781207561493,
|
864 |
+
"eval_runtime": 6.8412,
|
865 |
+
"eval_samples_per_second": 58.47,
|
866 |
+
"eval_steps_per_second": 1.9,
|
867 |
+
"step": 847
|
868 |
+
},
|
869 |
+
{
|
870 |
+
"epoch": 30.088495575221238,
|
871 |
+
"grad_norm": 8.386868476867676,
|
872 |
+
"learning_rate": 4.150132275132275e-05,
|
873 |
+
"loss": 0.0715,
|
874 |
+
"step": 850
|
875 |
+
},
|
876 |
+
{
|
877 |
+
"epoch": 30.442477876106196,
|
878 |
+
"grad_norm": 9.529677391052246,
|
879 |
+
"learning_rate": 4.133597883597884e-05,
|
880 |
+
"loss": 0.0497,
|
881 |
+
"step": 860
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 30.79646017699115,
|
885 |
+
"grad_norm": 7.131481647491455,
|
886 |
+
"learning_rate": 4.117063492063492e-05,
|
887 |
+
"loss": 0.0932,
|
888 |
+
"step": 870
|
889 |
+
},
|
890 |
+
{
|
891 |
+
"epoch": 30.97345132743363,
|
892 |
+
"eval_accuracy": 0.8575,
|
893 |
+
"eval_loss": 0.6106424927711487,
|
894 |
+
"eval_runtime": 6.5389,
|
895 |
+
"eval_samples_per_second": 61.173,
|
896 |
+
"eval_steps_per_second": 1.988,
|
897 |
+
"step": 875
|
898 |
+
},
|
899 |
+
{
|
900 |
+
"epoch": 31.150442477876105,
|
901 |
+
"grad_norm": 8.48017692565918,
|
902 |
+
"learning_rate": 4.100529100529101e-05,
|
903 |
+
"loss": 0.0737,
|
904 |
+
"step": 880
|
905 |
+
},
|
906 |
+
{
|
907 |
+
"epoch": 31.504424778761063,
|
908 |
+
"grad_norm": 20.205638885498047,
|
909 |
+
"learning_rate": 4.083994708994709e-05,
|
910 |
+
"loss": 0.1046,
|
911 |
+
"step": 890
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 31.858407079646017,
|
915 |
+
"grad_norm": 7.488587856292725,
|
916 |
+
"learning_rate": 4.067460317460318e-05,
|
917 |
+
"loss": 0.0967,
|
918 |
+
"step": 900
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 32.0,
|
922 |
+
"eval_accuracy": 0.855,
|
923 |
+
"eval_loss": 0.5614070296287537,
|
924 |
+
"eval_runtime": 6.637,
|
925 |
+
"eval_samples_per_second": 60.268,
|
926 |
+
"eval_steps_per_second": 1.959,
|
927 |
+
"step": 904
|
928 |
+
},
|
929 |
+
{
|
930 |
+
"epoch": 32.21238938053097,
|
931 |
+
"grad_norm": 15.713277816772461,
|
932 |
+
"learning_rate": 4.0509259259259265e-05,
|
933 |
+
"loss": 0.0824,
|
934 |
+
"step": 910
|
935 |
+
},
|
936 |
+
{
|
937 |
+
"epoch": 32.56637168141593,
|
938 |
+
"grad_norm": 14.559676170349121,
|
939 |
+
"learning_rate": 4.0343915343915346e-05,
|
940 |
+
"loss": 0.08,
|
941 |
+
"step": 920
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 32.92035398230089,
|
945 |
+
"grad_norm": 10.259367942810059,
|
946 |
+
"learning_rate": 4.017857142857143e-05,
|
947 |
+
"loss": 0.101,
|
948 |
+
"step": 930
|
949 |
+
},
|
950 |
+
{
|
951 |
+
"epoch": 32.991150442477874,
|
952 |
+
"eval_accuracy": 0.8525,
|
953 |
+
"eval_loss": 0.5946634411811829,
|
954 |
+
"eval_runtime": 6.419,
|
955 |
+
"eval_samples_per_second": 62.315,
|
956 |
+
"eval_steps_per_second": 2.025,
|
957 |
+
"step": 932
|
958 |
+
},
|
959 |
+
{
|
960 |
+
"epoch": 33.27433628318584,
|
961 |
+
"grad_norm": 12.391620635986328,
|
962 |
+
"learning_rate": 4.0013227513227514e-05,
|
963 |
+
"loss": 0.0685,
|
964 |
+
"step": 940
|
965 |
+
},
|
966 |
+
{
|
967 |
+
"epoch": 33.6283185840708,
|
968 |
+
"grad_norm": 8.084507942199707,
|
969 |
+
"learning_rate": 3.98478835978836e-05,
|
970 |
+
"loss": 0.0638,
|
971 |
+
"step": 950
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 33.982300884955755,
|
975 |
+
"grad_norm": 13.834248542785645,
|
976 |
+
"learning_rate": 3.968253968253968e-05,
|
977 |
+
"loss": 0.0734,
|
978 |
+
"step": 960
|
979 |
+
},
|
980 |
+
{
|
981 |
+
"epoch": 33.982300884955755,
|
982 |
+
"eval_accuracy": 0.87,
|
983 |
+
"eval_loss": 0.5388474464416504,
|
984 |
+
"eval_runtime": 6.7338,
|
985 |
+
"eval_samples_per_second": 59.402,
|
986 |
+
"eval_steps_per_second": 1.931,
|
987 |
+
"step": 960
|
988 |
+
},
|
989 |
+
{
|
990 |
+
"epoch": 34.336283185840706,
|
991 |
+
"grad_norm": 12.779220581054688,
|
992 |
+
"learning_rate": 3.951719576719577e-05,
|
993 |
+
"loss": 0.0652,
|
994 |
+
"step": 970
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 34.690265486725664,
|
998 |
+
"grad_norm": 11.898571968078613,
|
999 |
+
"learning_rate": 3.935185185185186e-05,
|
1000 |
+
"loss": 0.0742,
|
1001 |
+
"step": 980
|
1002 |
+
},
|
1003 |
+
{
|
1004 |
+
"epoch": 34.97345132743363,
|
1005 |
+
"eval_accuracy": 0.8725,
|
1006 |
+
"eval_loss": 0.51104736328125,
|
1007 |
+
"eval_runtime": 6.5062,
|
1008 |
+
"eval_samples_per_second": 61.48,
|
1009 |
+
"eval_steps_per_second": 1.998,
|
1010 |
+
"step": 988
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 35.04424778761062,
|
1014 |
+
"grad_norm": 3.3031704425811768,
|
1015 |
+
"learning_rate": 3.918650793650794e-05,
|
1016 |
+
"loss": 0.0663,
|
1017 |
+
"step": 990
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 35.39823008849557,
|
1021 |
+
"grad_norm": 15.559179306030273,
|
1022 |
+
"learning_rate": 3.9021164021164026e-05,
|
1023 |
+
"loss": 0.0798,
|
1024 |
+
"step": 1000
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 35.75221238938053,
|
1028 |
+
"grad_norm": 6.134705543518066,
|
1029 |
+
"learning_rate": 3.885582010582011e-05,
|
1030 |
+
"loss": 0.0698,
|
1031 |
+
"step": 1010
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 36.0,
|
1035 |
+
"eval_accuracy": 0.8525,
|
1036 |
+
"eval_loss": 0.5383738875389099,
|
1037 |
+
"eval_runtime": 6.546,
|
1038 |
+
"eval_samples_per_second": 61.106,
|
1039 |
+
"eval_steps_per_second": 1.986,
|
1040 |
+
"step": 1017
|
1041 |
+
},
|
1042 |
+
{
|
1043 |
+
"epoch": 36.10619469026549,
|
1044 |
+
"grad_norm": 11.319952964782715,
|
1045 |
+
"learning_rate": 3.8690476190476195e-05,
|
1046 |
+
"loss": 0.0822,
|
1047 |
+
"step": 1020
|
1048 |
+
},
|
1049 |
+
{
|
1050 |
+
"epoch": 36.46017699115044,
|
1051 |
+
"grad_norm": 10.789596557617188,
|
1052 |
+
"learning_rate": 3.8525132275132275e-05,
|
1053 |
+
"loss": 0.073,
|
1054 |
+
"step": 1030
|
1055 |
+
},
|
1056 |
+
{
|
1057 |
+
"epoch": 36.8141592920354,
|
1058 |
+
"grad_norm": 8.706029891967773,
|
1059 |
+
"learning_rate": 3.835978835978836e-05,
|
1060 |
+
"loss": 0.0785,
|
1061 |
+
"step": 1040
|
1062 |
+
},
|
1063 |
+
{
|
1064 |
+
"epoch": 36.991150442477874,
|
1065 |
+
"eval_accuracy": 0.8475,
|
1066 |
+
"eval_loss": 0.5407121777534485,
|
1067 |
+
"eval_runtime": 6.5706,
|
1068 |
+
"eval_samples_per_second": 60.878,
|
1069 |
+
"eval_steps_per_second": 1.979,
|
1070 |
+
"step": 1045
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 37.16814159292036,
|
1074 |
+
"grad_norm": 6.673099040985107,
|
1075 |
+
"learning_rate": 3.8194444444444444e-05,
|
1076 |
+
"loss": 0.0852,
|
1077 |
+
"step": 1050
|
1078 |
+
},
|
1079 |
+
{
|
1080 |
+
"epoch": 37.52212389380531,
|
1081 |
+
"grad_norm": 7.36367654800415,
|
1082 |
+
"learning_rate": 3.802910052910053e-05,
|
1083 |
+
"loss": 0.0886,
|
1084 |
+
"step": 1060
|
1085 |
+
},
|
1086 |
+
{
|
1087 |
+
"epoch": 37.876106194690266,
|
1088 |
+
"grad_norm": 3.059283971786499,
|
1089 |
+
"learning_rate": 3.786375661375662e-05,
|
1090 |
+
"loss": 0.0718,
|
1091 |
+
"step": 1070
|
1092 |
+
},
|
1093 |
+
{
|
1094 |
+
"epoch": 37.982300884955755,
|
1095 |
+
"eval_accuracy": 0.86,
|
1096 |
+
"eval_loss": 0.5420335531234741,
|
1097 |
+
"eval_runtime": 6.4154,
|
1098 |
+
"eval_samples_per_second": 62.35,
|
1099 |
+
"eval_steps_per_second": 2.026,
|
1100 |
+
"step": 1073
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 38.230088495575224,
|
1104 |
+
"grad_norm": 7.506045818328857,
|
1105 |
+
"learning_rate": 3.76984126984127e-05,
|
1106 |
+
"loss": 0.076,
|
1107 |
+
"step": 1080
|
1108 |
+
},
|
1109 |
+
{
|
1110 |
+
"epoch": 38.584070796460175,
|
1111 |
+
"grad_norm": 8.148099899291992,
|
1112 |
+
"learning_rate": 3.753306878306879e-05,
|
1113 |
+
"loss": 0.0803,
|
1114 |
+
"step": 1090
|
1115 |
+
},
|
1116 |
+
{
|
1117 |
+
"epoch": 38.93805309734513,
|
1118 |
+
"grad_norm": 7.432723045349121,
|
1119 |
+
"learning_rate": 3.736772486772487e-05,
|
1120 |
+
"loss": 0.061,
|
1121 |
+
"step": 1100
|
1122 |
+
},
|
1123 |
+
{
|
1124 |
+
"epoch": 38.97345132743363,
|
1125 |
+
"eval_accuracy": 0.8675,
|
1126 |
+
"eval_loss": 0.574724018573761,
|
1127 |
+
"eval_runtime": 6.7069,
|
1128 |
+
"eval_samples_per_second": 59.64,
|
1129 |
+
"eval_steps_per_second": 1.938,
|
1130 |
+
"step": 1101
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 39.29203539823009,
|
1134 |
+
"grad_norm": 14.355220794677734,
|
1135 |
+
"learning_rate": 3.7202380952380956e-05,
|
1136 |
+
"loss": 0.0548,
|
1137 |
+
"step": 1110
|
1138 |
+
},
|
1139 |
+
{
|
1140 |
+
"epoch": 39.64601769911504,
|
1141 |
+
"grad_norm": 10.292501449584961,
|
1142 |
+
"learning_rate": 3.7037037037037037e-05,
|
1143 |
+
"loss": 0.0502,
|
1144 |
+
"step": 1120
|
1145 |
+
},
|
1146 |
+
{
|
1147 |
+
"epoch": 40.0,
|
1148 |
+
"grad_norm": 14.376887321472168,
|
1149 |
+
"learning_rate": 3.6871693121693124e-05,
|
1150 |
+
"loss": 0.0695,
|
1151 |
+
"step": 1130
|
1152 |
+
},
|
1153 |
+
{
|
1154 |
+
"epoch": 40.0,
|
1155 |
+
"eval_accuracy": 0.8575,
|
1156 |
+
"eval_loss": 0.5829032063484192,
|
1157 |
+
"eval_runtime": 6.5719,
|
1158 |
+
"eval_samples_per_second": 60.865,
|
1159 |
+
"eval_steps_per_second": 1.978,
|
1160 |
+
"step": 1130
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 40.35398230088496,
|
1164 |
+
"grad_norm": 18.114715576171875,
|
1165 |
+
"learning_rate": 3.6706349206349205e-05,
|
1166 |
+
"loss": 0.0797,
|
1167 |
+
"step": 1140
|
1168 |
+
},
|
1169 |
+
{
|
1170 |
+
"epoch": 40.70796460176991,
|
1171 |
+
"grad_norm": 10.844046592712402,
|
1172 |
+
"learning_rate": 3.654100529100529e-05,
|
1173 |
+
"loss": 0.0611,
|
1174 |
+
"step": 1150
|
1175 |
+
},
|
1176 |
+
{
|
1177 |
+
"epoch": 40.991150442477874,
|
1178 |
+
"eval_accuracy": 0.8525,
|
1179 |
+
"eval_loss": 0.6212093830108643,
|
1180 |
+
"eval_runtime": 6.8081,
|
1181 |
+
"eval_samples_per_second": 58.754,
|
1182 |
+
"eval_steps_per_second": 1.909,
|
1183 |
+
"step": 1158
|
1184 |
+
},
|
1185 |
+
{
|
1186 |
+
"epoch": 41.06194690265487,
|
1187 |
+
"grad_norm": 12.924333572387695,
|
1188 |
+
"learning_rate": 3.637566137566138e-05,
|
1189 |
+
"loss": 0.0728,
|
1190 |
+
"step": 1160
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 41.415929203539825,
|
1194 |
+
"grad_norm": 3.6837644577026367,
|
1195 |
+
"learning_rate": 3.621031746031746e-05,
|
1196 |
+
"loss": 0.0573,
|
1197 |
+
"step": 1170
|
1198 |
+
},
|
1199 |
+
{
|
1200 |
+
"epoch": 41.769911504424776,
|
1201 |
+
"grad_norm": 5.20457124710083,
|
1202 |
+
"learning_rate": 3.604497354497355e-05,
|
1203 |
+
"loss": 0.0734,
|
1204 |
+
"step": 1180
|
1205 |
+
},
|
1206 |
+
{
|
1207 |
+
"epoch": 41.982300884955755,
|
1208 |
+
"eval_accuracy": 0.875,
|
1209 |
+
"eval_loss": 0.5034931898117065,
|
1210 |
+
"eval_runtime": 6.557,
|
1211 |
+
"eval_samples_per_second": 61.003,
|
1212 |
+
"eval_steps_per_second": 1.983,
|
1213 |
+
"step": 1186
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 42.123893805309734,
|
1217 |
+
"grad_norm": 7.184965133666992,
|
1218 |
+
"learning_rate": 3.587962962962963e-05,
|
1219 |
+
"loss": 0.0647,
|
1220 |
+
"step": 1190
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 42.47787610619469,
|
1224 |
+
"grad_norm": 7.6348066329956055,
|
1225 |
+
"learning_rate": 3.571428571428572e-05,
|
1226 |
+
"loss": 0.0603,
|
1227 |
+
"step": 1200
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 42.83185840707964,
|
1231 |
+
"grad_norm": 12.752991676330566,
|
1232 |
+
"learning_rate": 3.55489417989418e-05,
|
1233 |
+
"loss": 0.0643,
|
1234 |
+
"step": 1210
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 42.97345132743363,
|
1238 |
+
"eval_accuracy": 0.8775,
|
1239 |
+
"eval_loss": 0.5344811677932739,
|
1240 |
+
"eval_runtime": 6.6219,
|
1241 |
+
"eval_samples_per_second": 60.405,
|
1242 |
+
"eval_steps_per_second": 1.963,
|
1243 |
+
"step": 1214
|
1244 |
+
},
|
1245 |
+
{
|
1246 |
+
"epoch": 43.1858407079646,
|
1247 |
+
"grad_norm": 5.510544776916504,
|
1248 |
+
"learning_rate": 3.5383597883597885e-05,
|
1249 |
+
"loss": 0.0662,
|
1250 |
+
"step": 1220
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 43.53982300884956,
|
1254 |
+
"grad_norm": 2.6528031826019287,
|
1255 |
+
"learning_rate": 3.521825396825397e-05,
|
1256 |
+
"loss": 0.0675,
|
1257 |
+
"step": 1230
|
1258 |
+
},
|
1259 |
+
{
|
1260 |
+
"epoch": 43.89380530973451,
|
1261 |
+
"grad_norm": 6.398866653442383,
|
1262 |
+
"learning_rate": 3.5052910052910054e-05,
|
1263 |
+
"loss": 0.0625,
|
1264 |
+
"step": 1240
|
1265 |
+
},
|
1266 |
+
{
|
1267 |
+
"epoch": 44.0,
|
1268 |
+
"eval_accuracy": 0.8625,
|
1269 |
+
"eval_loss": 0.5208341479301453,
|
1270 |
+
"eval_runtime": 6.5147,
|
1271 |
+
"eval_samples_per_second": 61.4,
|
1272 |
+
"eval_steps_per_second": 1.995,
|
1273 |
+
"step": 1243
|
1274 |
+
},
|
1275 |
+
{
|
1276 |
+
"epoch": 44.24778761061947,
|
1277 |
+
"grad_norm": 6.792928695678711,
|
1278 |
+
"learning_rate": 3.488756613756614e-05,
|
1279 |
+
"loss": 0.0483,
|
1280 |
+
"step": 1250
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 44.60176991150443,
|
1284 |
+
"grad_norm": 14.822004318237305,
|
1285 |
+
"learning_rate": 3.472222222222222e-05,
|
1286 |
+
"loss": 0.0677,
|
1287 |
+
"step": 1260
|
1288 |
+
},
|
1289 |
+
{
|
1290 |
+
"epoch": 44.95575221238938,
|
1291 |
+
"grad_norm": 11.953215599060059,
|
1292 |
+
"learning_rate": 3.455687830687831e-05,
|
1293 |
+
"loss": 0.047,
|
1294 |
+
"step": 1270
|
1295 |
+
},
|
1296 |
+
{
|
1297 |
+
"epoch": 44.991150442477874,
|
1298 |
+
"eval_accuracy": 0.8675,
|
1299 |
+
"eval_loss": 0.5635126233100891,
|
1300 |
+
"eval_runtime": 6.9112,
|
1301 |
+
"eval_samples_per_second": 57.877,
|
1302 |
+
"eval_steps_per_second": 1.881,
|
1303 |
+
"step": 1271
|
1304 |
+
},
|
1305 |
+
{
|
1306 |
+
"epoch": 45.309734513274336,
|
1307 |
+
"grad_norm": 3.814394950866699,
|
1308 |
+
"learning_rate": 3.439153439153439e-05,
|
1309 |
+
"loss": 0.0629,
|
1310 |
+
"step": 1280
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 45.663716814159294,
|
1314 |
+
"grad_norm": 11.087026596069336,
|
1315 |
+
"learning_rate": 3.422619047619048e-05,
|
1316 |
+
"loss": 0.0612,
|
1317 |
+
"step": 1290
|
1318 |
+
},
|
1319 |
+
{
|
1320 |
+
"epoch": 45.982300884955755,
|
1321 |
+
"eval_accuracy": 0.8775,
|
1322 |
+
"eval_loss": 0.4720584750175476,
|
1323 |
+
"eval_runtime": 6.6138,
|
1324 |
+
"eval_samples_per_second": 60.48,
|
1325 |
+
"eval_steps_per_second": 1.966,
|
1326 |
+
"step": 1299
|
1327 |
+
},
|
1328 |
+
{
|
1329 |
+
"epoch": 46.017699115044245,
|
1330 |
+
"grad_norm": 11.228111267089844,
|
1331 |
+
"learning_rate": 3.406084656084656e-05,
|
1332 |
+
"loss": 0.0503,
|
1333 |
+
"step": 1300
|
1334 |
+
},
|
1335 |
+
{
|
1336 |
+
"epoch": 46.3716814159292,
|
1337 |
+
"grad_norm": 5.904040813446045,
|
1338 |
+
"learning_rate": 3.3895502645502647e-05,
|
1339 |
+
"loss": 0.0598,
|
1340 |
+
"step": 1310
|
1341 |
+
},
|
1342 |
+
{
|
1343 |
+
"epoch": 46.72566371681416,
|
1344 |
+
"grad_norm": 4.406146049499512,
|
1345 |
+
"learning_rate": 3.3730158730158734e-05,
|
1346 |
+
"loss": 0.0582,
|
1347 |
+
"step": 1320
|
1348 |
+
},
|
1349 |
+
{
|
1350 |
+
"epoch": 46.97345132743363,
|
1351 |
+
"eval_accuracy": 0.855,
|
1352 |
+
"eval_loss": 0.5683026909828186,
|
1353 |
+
"eval_runtime": 6.5041,
|
1354 |
+
"eval_samples_per_second": 61.5,
|
1355 |
+
"eval_steps_per_second": 1.999,
|
1356 |
+
"step": 1327
|
1357 |
+
},
|
1358 |
+
{
|
1359 |
+
"epoch": 47.07964601769911,
|
1360 |
+
"grad_norm": 4.003131866455078,
|
1361 |
+
"learning_rate": 3.3564814814814815e-05,
|
1362 |
+
"loss": 0.0717,
|
1363 |
+
"step": 1330
|
1364 |
+
},
|
1365 |
+
{
|
1366 |
+
"epoch": 47.43362831858407,
|
1367 |
+
"grad_norm": 4.299389839172363,
|
1368 |
+
"learning_rate": 3.33994708994709e-05,
|
1369 |
+
"loss": 0.0387,
|
1370 |
+
"step": 1340
|
1371 |
+
},
|
1372 |
+
{
|
1373 |
+
"epoch": 47.78761061946903,
|
1374 |
+
"grad_norm": 2.899742364883423,
|
1375 |
+
"learning_rate": 3.3234126984126983e-05,
|
1376 |
+
"loss": 0.0516,
|
1377 |
+
"step": 1350
|
1378 |
+
},
|
1379 |
+
{
|
1380 |
+
"epoch": 48.0,
|
1381 |
+
"eval_accuracy": 0.8625,
|
1382 |
+
"eval_loss": 0.5882661938667297,
|
1383 |
+
"eval_runtime": 6.5446,
|
1384 |
+
"eval_samples_per_second": 61.119,
|
1385 |
+
"eval_steps_per_second": 1.986,
|
1386 |
+
"step": 1356
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 48.14159292035398,
|
1390 |
+
"grad_norm": 13.842363357543945,
|
1391 |
+
"learning_rate": 3.306878306878307e-05,
|
1392 |
+
"loss": 0.0612,
|
1393 |
+
"step": 1360
|
1394 |
+
},
|
1395 |
+
{
|
1396 |
+
"epoch": 48.49557522123894,
|
1397 |
+
"grad_norm": 8.8641939163208,
|
1398 |
+
"learning_rate": 3.290343915343915e-05,
|
1399 |
+
"loss": 0.0694,
|
1400 |
+
"step": 1370
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 48.849557522123895,
|
1404 |
+
"grad_norm": 1.9123202562332153,
|
1405 |
+
"learning_rate": 3.273809523809524e-05,
|
1406 |
+
"loss": 0.0427,
|
1407 |
+
"step": 1380
|
1408 |
+
},
|
1409 |
+
{
|
1410 |
+
"epoch": 48.991150442477874,
|
1411 |
+
"eval_accuracy": 0.8575,
|
1412 |
+
"eval_loss": 0.5756899118423462,
|
1413 |
+
"eval_runtime": 6.7885,
|
1414 |
+
"eval_samples_per_second": 58.923,
|
1415 |
+
"eval_steps_per_second": 1.915,
|
1416 |
+
"step": 1384
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 49.203539823008846,
|
1420 |
+
"grad_norm": 8.457569122314453,
|
1421 |
+
"learning_rate": 3.257275132275133e-05,
|
1422 |
+
"loss": 0.0541,
|
1423 |
+
"step": 1390
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 49.557522123893804,
|
1427 |
+
"grad_norm": 10.296213150024414,
|
1428 |
+
"learning_rate": 3.240740740740741e-05,
|
1429 |
+
"loss": 0.0693,
|
1430 |
+
"step": 1400
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 49.91150442477876,
|
1434 |
+
"grad_norm": 7.295350551605225,
|
1435 |
+
"learning_rate": 3.2242063492063495e-05,
|
1436 |
+
"loss": 0.0601,
|
1437 |
+
"step": 1410
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 49.982300884955755,
|
1441 |
+
"eval_accuracy": 0.8625,
|
1442 |
+
"eval_loss": 0.5367708802223206,
|
1443 |
+
"eval_runtime": 6.5402,
|
1444 |
+
"eval_samples_per_second": 61.16,
|
1445 |
+
"eval_steps_per_second": 1.988,
|
1446 |
+
"step": 1412
|
1447 |
+
},
|
1448 |
+
{
|
1449 |
+
"epoch": 50.26548672566372,
|
1450 |
+
"grad_norm": 10.658272743225098,
|
1451 |
+
"learning_rate": 3.2076719576719576e-05,
|
1452 |
+
"loss": 0.06,
|
1453 |
+
"step": 1420
|
1454 |
+
},
|
1455 |
+
{
|
1456 |
+
"epoch": 50.61946902654867,
|
1457 |
+
"grad_norm": 8.091354370117188,
|
1458 |
+
"learning_rate": 3.1911375661375664e-05,
|
1459 |
+
"loss": 0.0653,
|
1460 |
+
"step": 1430
|
1461 |
+
},
|
1462 |
+
{
|
1463 |
+
"epoch": 50.97345132743363,
|
1464 |
+
"grad_norm": 6.116772174835205,
|
1465 |
+
"learning_rate": 3.1746031746031745e-05,
|
1466 |
+
"loss": 0.0645,
|
1467 |
+
"step": 1440
|
1468 |
+
},
|
1469 |
+
{
|
1470 |
+
"epoch": 50.97345132743363,
|
1471 |
+
"eval_accuracy": 0.84,
|
1472 |
+
"eval_loss": 0.5608373284339905,
|
1473 |
+
"eval_runtime": 6.6192,
|
1474 |
+
"eval_samples_per_second": 60.43,
|
1475 |
+
"eval_steps_per_second": 1.964,
|
1476 |
+
"step": 1440
|
1477 |
+
},
|
1478 |
+
{
|
1479 |
+
"epoch": 51.32743362831859,
|
1480 |
+
"grad_norm": 10.381439208984375,
|
1481 |
+
"learning_rate": 3.158068783068783e-05,
|
1482 |
+
"loss": 0.0574,
|
1483 |
+
"step": 1450
|
1484 |
+
},
|
1485 |
+
{
|
1486 |
+
"epoch": 51.68141592920354,
|
1487 |
+
"grad_norm": 4.16278076171875,
|
1488 |
+
"learning_rate": 3.141534391534391e-05,
|
1489 |
+
"loss": 0.054,
|
1490 |
+
"step": 1460
|
1491 |
+
},
|
1492 |
+
{
|
1493 |
+
"epoch": 52.0,
|
1494 |
+
"eval_accuracy": 0.87,
|
1495 |
+
"eval_loss": 0.5379666686058044,
|
1496 |
+
"eval_runtime": 6.5723,
|
1497 |
+
"eval_samples_per_second": 60.862,
|
1498 |
+
"eval_steps_per_second": 1.978,
|
1499 |
+
"step": 1469
|
1500 |
+
},
|
1501 |
+
{
|
1502 |
+
"epoch": 52.0353982300885,
|
1503 |
+
"grad_norm": 6.869332313537598,
|
1504 |
+
"learning_rate": 3.125e-05,
|
1505 |
+
"loss": 0.0516,
|
1506 |
+
"step": 1470
|
1507 |
+
},
|
1508 |
+
{
|
1509 |
+
"epoch": 52.389380530973455,
|
1510 |
+
"grad_norm": 5.33891487121582,
|
1511 |
+
"learning_rate": 3.108465608465609e-05,
|
1512 |
+
"loss": 0.0548,
|
1513 |
+
"step": 1480
|
1514 |
+
},
|
1515 |
+
{
|
1516 |
+
"epoch": 52.743362831858406,
|
1517 |
+
"grad_norm": 8.20958137512207,
|
1518 |
+
"learning_rate": 3.091931216931217e-05,
|
1519 |
+
"loss": 0.0647,
|
1520 |
+
"step": 1490
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 52.991150442477874,
|
1524 |
+
"eval_accuracy": 0.8625,
|
1525 |
+
"eval_loss": 0.5490260124206543,
|
1526 |
+
"eval_runtime": 6.4131,
|
1527 |
+
"eval_samples_per_second": 62.372,
|
1528 |
+
"eval_steps_per_second": 2.027,
|
1529 |
+
"step": 1497
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 53.097345132743364,
|
1533 |
+
"grad_norm": 4.3137617111206055,
|
1534 |
+
"learning_rate": 3.075396825396826e-05,
|
1535 |
+
"loss": 0.0574,
|
1536 |
+
"step": 1500
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 53.45132743362832,
|
1540 |
+
"grad_norm": 8.59358024597168,
|
1541 |
+
"learning_rate": 3.058862433862434e-05,
|
1542 |
+
"loss": 0.0553,
|
1543 |
+
"step": 1510
|
1544 |
+
},
|
1545 |
+
{
|
1546 |
+
"epoch": 53.80530973451327,
|
1547 |
+
"grad_norm": 12.394618034362793,
|
1548 |
+
"learning_rate": 3.0423280423280425e-05,
|
1549 |
+
"loss": 0.0539,
|
1550 |
+
"step": 1520
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 53.982300884955755,
|
1554 |
+
"eval_accuracy": 0.8625,
|
1555 |
+
"eval_loss": 0.5685680508613586,
|
1556 |
+
"eval_runtime": 6.3501,
|
1557 |
+
"eval_samples_per_second": 62.991,
|
1558 |
+
"eval_steps_per_second": 2.047,
|
1559 |
+
"step": 1525
|
1560 |
+
},
|
1561 |
+
{
|
1562 |
+
"epoch": 54.15929203539823,
|
1563 |
+
"grad_norm": 5.349567890167236,
|
1564 |
+
"learning_rate": 3.0257936507936506e-05,
|
1565 |
+
"loss": 0.0658,
|
1566 |
+
"step": 1530
|
1567 |
+
},
|
1568 |
+
{
|
1569 |
+
"epoch": 54.51327433628319,
|
1570 |
+
"grad_norm": 9.836604118347168,
|
1571 |
+
"learning_rate": 3.0092592592592593e-05,
|
1572 |
+
"loss": 0.0587,
|
1573 |
+
"step": 1540
|
1574 |
+
},
|
1575 |
+
{
|
1576 |
+
"epoch": 54.86725663716814,
|
1577 |
+
"grad_norm": 3.192077159881592,
|
1578 |
+
"learning_rate": 2.9927248677248678e-05,
|
1579 |
+
"loss": 0.0485,
|
1580 |
+
"step": 1550
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 54.97345132743363,
|
1584 |
+
"eval_accuracy": 0.8725,
|
1585 |
+
"eval_loss": 0.5474461913108826,
|
1586 |
+
"eval_runtime": 6.299,
|
1587 |
+
"eval_samples_per_second": 63.502,
|
1588 |
+
"eval_steps_per_second": 2.064,
|
1589 |
+
"step": 1553
|
1590 |
+
},
|
1591 |
+
{
|
1592 |
+
"epoch": 55.2212389380531,
|
1593 |
+
"grad_norm": 12.98410415649414,
|
1594 |
+
"learning_rate": 2.9761904761904762e-05,
|
1595 |
+
"loss": 0.0585,
|
1596 |
+
"step": 1560
|
1597 |
+
},
|
1598 |
+
{
|
1599 |
+
"epoch": 55.575221238938056,
|
1600 |
+
"grad_norm": 5.569449424743652,
|
1601 |
+
"learning_rate": 2.959656084656085e-05,
|
1602 |
+
"loss": 0.0528,
|
1603 |
+
"step": 1570
|
1604 |
+
},
|
1605 |
+
{
|
1606 |
+
"epoch": 55.92920353982301,
|
1607 |
+
"grad_norm": 5.522889137268066,
|
1608 |
+
"learning_rate": 2.943121693121693e-05,
|
1609 |
+
"loss": 0.0649,
|
1610 |
+
"step": 1580
|
1611 |
+
},
|
1612 |
+
{
|
1613 |
+
"epoch": 56.0,
|
1614 |
+
"eval_accuracy": 0.86,
|
1615 |
+
"eval_loss": 0.5937778949737549,
|
1616 |
+
"eval_runtime": 6.5642,
|
1617 |
+
"eval_samples_per_second": 60.937,
|
1618 |
+
"eval_steps_per_second": 1.98,
|
1619 |
+
"step": 1582
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 56.283185840707965,
|
1623 |
+
"grad_norm": 5.637914180755615,
|
1624 |
+
"learning_rate": 2.9265873015873018e-05,
|
1625 |
+
"loss": 0.0401,
|
1626 |
+
"step": 1590
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 56.63716814159292,
|
1630 |
+
"grad_norm": 9.390466690063477,
|
1631 |
+
"learning_rate": 2.91005291005291e-05,
|
1632 |
+
"loss": 0.0645,
|
1633 |
+
"step": 1600
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 56.991150442477874,
|
1637 |
+
"grad_norm": 1.8010101318359375,
|
1638 |
+
"learning_rate": 2.8935185185185186e-05,
|
1639 |
+
"loss": 0.0486,
|
1640 |
+
"step": 1610
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 56.991150442477874,
|
1644 |
+
"eval_accuracy": 0.86,
|
1645 |
+
"eval_loss": 0.5641574263572693,
|
1646 |
+
"eval_runtime": 6.3881,
|
1647 |
+
"eval_samples_per_second": 62.617,
|
1648 |
+
"eval_steps_per_second": 2.035,
|
1649 |
+
"step": 1610
|
1650 |
+
},
|
1651 |
+
{
|
1652 |
+
"epoch": 57.34513274336283,
|
1653 |
+
"grad_norm": 4.3999924659729,
|
1654 |
+
"learning_rate": 2.876984126984127e-05,
|
1655 |
+
"loss": 0.0564,
|
1656 |
+
"step": 1620
|
1657 |
+
},
|
1658 |
+
{
|
1659 |
+
"epoch": 57.69911504424779,
|
1660 |
+
"grad_norm": 4.902792930603027,
|
1661 |
+
"learning_rate": 2.8604497354497355e-05,
|
1662 |
+
"loss": 0.0385,
|
1663 |
+
"step": 1630
|
1664 |
+
},
|
1665 |
+
{
|
1666 |
+
"epoch": 57.982300884955755,
|
1667 |
+
"eval_accuracy": 0.8675,
|
1668 |
+
"eval_loss": 0.5389891266822815,
|
1669 |
+
"eval_runtime": 6.5714,
|
1670 |
+
"eval_samples_per_second": 60.87,
|
1671 |
+
"eval_steps_per_second": 1.978,
|
1672 |
+
"step": 1638
|
1673 |
+
},
|
1674 |
+
{
|
1675 |
+
"epoch": 58.05309734513274,
|
1676 |
+
"grad_norm": 2.268341541290283,
|
1677 |
+
"learning_rate": 2.8439153439153442e-05,
|
1678 |
+
"loss": 0.0307,
|
1679 |
+
"step": 1640
|
1680 |
+
},
|
1681 |
+
{
|
1682 |
+
"epoch": 58.4070796460177,
|
1683 |
+
"grad_norm": 4.105015277862549,
|
1684 |
+
"learning_rate": 2.8273809523809523e-05,
|
1685 |
+
"loss": 0.0521,
|
1686 |
+
"step": 1650
|
1687 |
+
},
|
1688 |
+
{
|
1689 |
+
"epoch": 58.76106194690266,
|
1690 |
+
"grad_norm": 10.11693000793457,
|
1691 |
+
"learning_rate": 2.810846560846561e-05,
|
1692 |
+
"loss": 0.0404,
|
1693 |
+
"step": 1660
|
1694 |
+
},
|
1695 |
+
{
|
1696 |
+
"epoch": 58.97345132743363,
|
1697 |
+
"eval_accuracy": 0.8775,
|
1698 |
+
"eval_loss": 0.5734975934028625,
|
1699 |
+
"eval_runtime": 6.7005,
|
1700 |
+
"eval_samples_per_second": 59.697,
|
1701 |
+
"eval_steps_per_second": 1.94,
|
1702 |
+
"step": 1666
|
1703 |
+
},
|
1704 |
+
{
|
1705 |
+
"epoch": 59.11504424778761,
|
1706 |
+
"grad_norm": 6.96762228012085,
|
1707 |
+
"learning_rate": 2.7943121693121695e-05,
|
1708 |
+
"loss": 0.0536,
|
1709 |
+
"step": 1670
|
1710 |
+
},
|
1711 |
+
{
|
1712 |
+
"epoch": 59.469026548672566,
|
1713 |
+
"grad_norm": 5.11080265045166,
|
1714 |
+
"learning_rate": 2.777777777777778e-05,
|
1715 |
+
"loss": 0.0487,
|
1716 |
+
"step": 1680
|
1717 |
+
},
|
1718 |
+
{
|
1719 |
+
"epoch": 59.823008849557525,
|
1720 |
+
"grad_norm": 14.329901695251465,
|
1721 |
+
"learning_rate": 2.7612433862433863e-05,
|
1722 |
+
"loss": 0.0543,
|
1723 |
+
"step": 1690
|
1724 |
+
},
|
1725 |
+
{
|
1726 |
+
"epoch": 60.0,
|
1727 |
+
"eval_accuracy": 0.875,
|
1728 |
+
"eval_loss": 0.5116520524024963,
|
1729 |
+
"eval_runtime": 6.4229,
|
1730 |
+
"eval_samples_per_second": 62.277,
|
1731 |
+
"eval_steps_per_second": 2.024,
|
1732 |
+
"step": 1695
|
1733 |
+
},
|
1734 |
+
{
|
1735 |
+
"epoch": 60.176991150442475,
|
1736 |
+
"grad_norm": 5.092159271240234,
|
1737 |
+
"learning_rate": 2.7447089947089948e-05,
|
1738 |
+
"loss": 0.0512,
|
1739 |
+
"step": 1700
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 60.530973451327434,
|
1743 |
+
"grad_norm": 7.392366409301758,
|
1744 |
+
"learning_rate": 2.7281746031746032e-05,
|
1745 |
+
"loss": 0.051,
|
1746 |
+
"step": 1710
|
1747 |
+
},
|
1748 |
+
{
|
1749 |
+
"epoch": 60.88495575221239,
|
1750 |
+
"grad_norm": 5.99697732925415,
|
1751 |
+
"learning_rate": 2.7116402116402116e-05,
|
1752 |
+
"loss": 0.0506,
|
1753 |
+
"step": 1720
|
1754 |
+
},
|
1755 |
+
{
|
1756 |
+
"epoch": 60.991150442477874,
|
1757 |
+
"eval_accuracy": 0.8725,
|
1758 |
+
"eval_loss": 0.5422492027282715,
|
1759 |
+
"eval_runtime": 6.5064,
|
1760 |
+
"eval_samples_per_second": 61.478,
|
1761 |
+
"eval_steps_per_second": 1.998,
|
1762 |
+
"step": 1723
|
1763 |
+
},
|
1764 |
+
{
|
1765 |
+
"epoch": 61.23893805309734,
|
1766 |
+
"grad_norm": 7.004878044128418,
|
1767 |
+
"learning_rate": 2.6951058201058204e-05,
|
1768 |
+
"loss": 0.0435,
|
1769 |
+
"step": 1730
|
1770 |
+
},
|
1771 |
+
{
|
1772 |
+
"epoch": 61.5929203539823,
|
1773 |
+
"grad_norm": 10.655908584594727,
|
1774 |
+
"learning_rate": 2.6785714285714288e-05,
|
1775 |
+
"loss": 0.0522,
|
1776 |
+
"step": 1740
|
1777 |
+
},
|
1778 |
+
{
|
1779 |
+
"epoch": 61.94690265486726,
|
1780 |
+
"grad_norm": 2.694920778274536,
|
1781 |
+
"learning_rate": 2.6620370370370372e-05,
|
1782 |
+
"loss": 0.0398,
|
1783 |
+
"step": 1750
|
1784 |
+
},
|
1785 |
+
{
|
1786 |
+
"epoch": 61.982300884955755,
|
1787 |
+
"eval_accuracy": 0.87,
|
1788 |
+
"eval_loss": 0.5472509860992432,
|
1789 |
+
"eval_runtime": 6.5285,
|
1790 |
+
"eval_samples_per_second": 61.27,
|
1791 |
+
"eval_steps_per_second": 1.991,
|
1792 |
+
"step": 1751
|
1793 |
+
},
|
1794 |
+
{
|
1795 |
+
"epoch": 62.30088495575221,
|
1796 |
+
"grad_norm": 4.999312400817871,
|
1797 |
+
"learning_rate": 2.6455026455026456e-05,
|
1798 |
+
"loss": 0.0463,
|
1799 |
+
"step": 1760
|
1800 |
+
},
|
1801 |
+
{
|
1802 |
+
"epoch": 62.65486725663717,
|
1803 |
+
"grad_norm": 5.696354389190674,
|
1804 |
+
"learning_rate": 2.628968253968254e-05,
|
1805 |
+
"loss": 0.0494,
|
1806 |
+
"step": 1770
|
1807 |
+
},
|
1808 |
+
{
|
1809 |
+
"epoch": 62.97345132743363,
|
1810 |
+
"eval_accuracy": 0.8675,
|
1811 |
+
"eval_loss": 0.5332581996917725,
|
1812 |
+
"eval_runtime": 6.6059,
|
1813 |
+
"eval_samples_per_second": 60.552,
|
1814 |
+
"eval_steps_per_second": 1.968,
|
1815 |
+
"step": 1779
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 63.008849557522126,
|
1819 |
+
"grad_norm": 9.548038482666016,
|
1820 |
+
"learning_rate": 2.6124338624338625e-05,
|
1821 |
+
"loss": 0.0446,
|
1822 |
+
"step": 1780
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 63.36283185840708,
|
1826 |
+
"grad_norm": 7.306167125701904,
|
1827 |
+
"learning_rate": 2.5958994708994712e-05,
|
1828 |
+
"loss": 0.0541,
|
1829 |
+
"step": 1790
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 63.716814159292035,
|
1833 |
+
"grad_norm": 16.248004913330078,
|
1834 |
+
"learning_rate": 2.5793650793650796e-05,
|
1835 |
+
"loss": 0.0472,
|
1836 |
+
"step": 1800
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 64.0,
|
1840 |
+
"eval_accuracy": 0.8825,
|
1841 |
+
"eval_loss": 0.5650449395179749,
|
1842 |
+
"eval_runtime": 6.5122,
|
1843 |
+
"eval_samples_per_second": 61.423,
|
1844 |
+
"eval_steps_per_second": 1.996,
|
1845 |
+
"step": 1808
|
1846 |
+
},
|
1847 |
+
{
|
1848 |
+
"epoch": 64.070796460177,
|
1849 |
+
"grad_norm": 4.15922212600708,
|
1850 |
+
"learning_rate": 2.562830687830688e-05,
|
1851 |
+
"loss": 0.0564,
|
1852 |
+
"step": 1810
|
1853 |
+
},
|
1854 |
+
{
|
1855 |
+
"epoch": 64.42477876106194,
|
1856 |
+
"grad_norm": 4.988418102264404,
|
1857 |
+
"learning_rate": 2.5462962962962965e-05,
|
1858 |
+
"loss": 0.0426,
|
1859 |
+
"step": 1820
|
1860 |
+
},
|
1861 |
+
{
|
1862 |
+
"epoch": 64.77876106194691,
|
1863 |
+
"grad_norm": 8.311708450317383,
|
1864 |
+
"learning_rate": 2.529761904761905e-05,
|
1865 |
+
"loss": 0.0504,
|
1866 |
+
"step": 1830
|
1867 |
+
},
|
1868 |
+
{
|
1869 |
+
"epoch": 64.99115044247787,
|
1870 |
+
"eval_accuracy": 0.8575,
|
1871 |
+
"eval_loss": 0.577080488204956,
|
1872 |
+
"eval_runtime": 6.3693,
|
1873 |
+
"eval_samples_per_second": 62.801,
|
1874 |
+
"eval_steps_per_second": 2.041,
|
1875 |
+
"step": 1836
|
1876 |
+
},
|
1877 |
+
{
|
1878 |
+
"epoch": 65.13274336283186,
|
1879 |
+
"grad_norm": 7.197911739349365,
|
1880 |
+
"learning_rate": 2.5132275132275137e-05,
|
1881 |
+
"loss": 0.0531,
|
1882 |
+
"step": 1840
|
1883 |
+
},
|
1884 |
+
{
|
1885 |
+
"epoch": 65.48672566371681,
|
1886 |
+
"grad_norm": 3.72579026222229,
|
1887 |
+
"learning_rate": 2.496693121693122e-05,
|
1888 |
+
"loss": 0.052,
|
1889 |
+
"step": 1850
|
1890 |
+
},
|
1891 |
+
{
|
1892 |
+
"epoch": 65.84070796460178,
|
1893 |
+
"grad_norm": 5.802933216094971,
|
1894 |
+
"learning_rate": 2.4801587301587305e-05,
|
1895 |
+
"loss": 0.044,
|
1896 |
+
"step": 1860
|
1897 |
+
},
|
1898 |
+
{
|
1899 |
+
"epoch": 65.98230088495575,
|
1900 |
+
"eval_accuracy": 0.86,
|
1901 |
+
"eval_loss": 0.5219700336456299,
|
1902 |
+
"eval_runtime": 6.6071,
|
1903 |
+
"eval_samples_per_second": 60.541,
|
1904 |
+
"eval_steps_per_second": 1.968,
|
1905 |
+
"step": 1864
|
1906 |
+
},
|
1907 |
+
{
|
1908 |
+
"epoch": 66.19469026548673,
|
1909 |
+
"grad_norm": 3.172950267791748,
|
1910 |
+
"learning_rate": 2.463624338624339e-05,
|
1911 |
+
"loss": 0.0526,
|
1912 |
+
"step": 1870
|
1913 |
+
},
|
1914 |
+
{
|
1915 |
+
"epoch": 66.54867256637168,
|
1916 |
+
"grad_norm": 8.036627769470215,
|
1917 |
+
"learning_rate": 2.4470899470899473e-05,
|
1918 |
+
"loss": 0.0405,
|
1919 |
+
"step": 1880
|
1920 |
+
},
|
1921 |
+
{
|
1922 |
+
"epoch": 66.90265486725664,
|
1923 |
+
"grad_norm": 1.321056604385376,
|
1924 |
+
"learning_rate": 2.4305555555555558e-05,
|
1925 |
+
"loss": 0.061,
|
1926 |
+
"step": 1890
|
1927 |
+
},
|
1928 |
+
{
|
1929 |
+
"epoch": 66.97345132743362,
|
1930 |
+
"eval_accuracy": 0.8725,
|
1931 |
+
"eval_loss": 0.562214195728302,
|
1932 |
+
"eval_runtime": 6.4463,
|
1933 |
+
"eval_samples_per_second": 62.051,
|
1934 |
+
"eval_steps_per_second": 2.017,
|
1935 |
+
"step": 1892
|
1936 |
+
},
|
1937 |
+
{
|
1938 |
+
"epoch": 67.2566371681416,
|
1939 |
+
"grad_norm": 5.427750110626221,
|
1940 |
+
"learning_rate": 2.4140211640211642e-05,
|
1941 |
+
"loss": 0.0476,
|
1942 |
+
"step": 1900
|
1943 |
+
},
|
1944 |
+
{
|
1945 |
+
"epoch": 67.61061946902655,
|
1946 |
+
"grad_norm": 4.76095724105835,
|
1947 |
+
"learning_rate": 2.3974867724867726e-05,
|
1948 |
+
"loss": 0.0556,
|
1949 |
+
"step": 1910
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 67.96460176991151,
|
1953 |
+
"grad_norm": 9.611533164978027,
|
1954 |
+
"learning_rate": 2.380952380952381e-05,
|
1955 |
+
"loss": 0.0459,
|
1956 |
+
"step": 1920
|
1957 |
+
},
|
1958 |
+
{
|
1959 |
+
"epoch": 68.0,
|
1960 |
+
"eval_accuracy": 0.8625,
|
1961 |
+
"eval_loss": 0.5864116549491882,
|
1962 |
+
"eval_runtime": 6.4617,
|
1963 |
+
"eval_samples_per_second": 61.903,
|
1964 |
+
"eval_steps_per_second": 2.012,
|
1965 |
+
"step": 1921
|
1966 |
+
},
|
1967 |
+
{
|
1968 |
+
"epoch": 68.31858407079646,
|
1969 |
+
"grad_norm": 4.835655689239502,
|
1970 |
+
"learning_rate": 2.3644179894179898e-05,
|
1971 |
+
"loss": 0.0505,
|
1972 |
+
"step": 1930
|
1973 |
+
},
|
1974 |
+
{
|
1975 |
+
"epoch": 68.67256637168141,
|
1976 |
+
"grad_norm": 2.6347382068634033,
|
1977 |
+
"learning_rate": 2.3478835978835982e-05,
|
1978 |
+
"loss": 0.0294,
|
1979 |
+
"step": 1940
|
1980 |
+
},
|
1981 |
+
{
|
1982 |
+
"epoch": 68.99115044247787,
|
1983 |
+
"eval_accuracy": 0.8625,
|
1984 |
+
"eval_loss": 0.6341035962104797,
|
1985 |
+
"eval_runtime": 6.4243,
|
1986 |
+
"eval_samples_per_second": 62.263,
|
1987 |
+
"eval_steps_per_second": 2.024,
|
1988 |
+
"step": 1949
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 69.02654867256638,
|
1992 |
+
"grad_norm": 7.225272178649902,
|
1993 |
+
"learning_rate": 2.3313492063492066e-05,
|
1994 |
+
"loss": 0.0272,
|
1995 |
+
"step": 1950
|
1996 |
+
},
|
1997 |
+
{
|
1998 |
+
"epoch": 69.38053097345133,
|
1999 |
+
"grad_norm": 5.250880718231201,
|
2000 |
+
"learning_rate": 2.314814814814815e-05,
|
2001 |
+
"loss": 0.0485,
|
2002 |
+
"step": 1960
|
2003 |
+
},
|
2004 |
+
{
|
2005 |
+
"epoch": 69.73451327433628,
|
2006 |
+
"grad_norm": 7.59017276763916,
|
2007 |
+
"learning_rate": 2.2982804232804235e-05,
|
2008 |
+
"loss": 0.0428,
|
2009 |
+
"step": 1970
|
2010 |
+
},
|
2011 |
+
{
|
2012 |
+
"epoch": 69.98230088495575,
|
2013 |
+
"eval_accuracy": 0.8675,
|
2014 |
+
"eval_loss": 0.569638192653656,
|
2015 |
+
"eval_runtime": 6.761,
|
2016 |
+
"eval_samples_per_second": 59.163,
|
2017 |
+
"eval_steps_per_second": 1.923,
|
2018 |
+
"step": 1977
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 70.08849557522124,
|
2022 |
+
"grad_norm": 0.30605605244636536,
|
2023 |
+
"learning_rate": 2.281746031746032e-05,
|
2024 |
+
"loss": 0.0364,
|
2025 |
+
"step": 1980
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 70.4424778761062,
|
2029 |
+
"grad_norm": 5.189797878265381,
|
2030 |
+
"learning_rate": 2.2652116402116403e-05,
|
2031 |
+
"loss": 0.0313,
|
2032 |
+
"step": 1990
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 70.79646017699115,
|
2036 |
+
"grad_norm": 6.693019866943359,
|
2037 |
+
"learning_rate": 2.2486772486772487e-05,
|
2038 |
+
"loss": 0.0317,
|
2039 |
+
"step": 2000
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 70.97345132743362,
|
2043 |
+
"eval_accuracy": 0.845,
|
2044 |
+
"eval_loss": 0.6312776207923889,
|
2045 |
+
"eval_runtime": 6.5012,
|
2046 |
+
"eval_samples_per_second": 61.527,
|
2047 |
+
"eval_steps_per_second": 2.0,
|
2048 |
+
"step": 2005
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 71.15044247787611,
|
2052 |
+
"grad_norm": 4.359290599822998,
|
2053 |
+
"learning_rate": 2.2321428571428575e-05,
|
2054 |
+
"loss": 0.0403,
|
2055 |
+
"step": 2010
|
2056 |
+
},
|
2057 |
+
{
|
2058 |
+
"epoch": 71.50442477876106,
|
2059 |
+
"grad_norm": 4.60231351852417,
|
2060 |
+
"learning_rate": 2.215608465608466e-05,
|
2061 |
+
"loss": 0.0412,
|
2062 |
+
"step": 2020
|
2063 |
+
},
|
2064 |
+
{
|
2065 |
+
"epoch": 71.85840707964601,
|
2066 |
+
"grad_norm": 1.765071988105774,
|
2067 |
+
"learning_rate": 2.1990740740740743e-05,
|
2068 |
+
"loss": 0.0453,
|
2069 |
+
"step": 2030
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 72.0,
|
2073 |
+
"eval_accuracy": 0.875,
|
2074 |
+
"eval_loss": 0.5954553484916687,
|
2075 |
+
"eval_runtime": 6.4648,
|
2076 |
+
"eval_samples_per_second": 61.873,
|
2077 |
+
"eval_steps_per_second": 2.011,
|
2078 |
+
"step": 2034
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 72.21238938053098,
|
2082 |
+
"grad_norm": 11.668362617492676,
|
2083 |
+
"learning_rate": 2.1825396825396827e-05,
|
2084 |
+
"loss": 0.0486,
|
2085 |
+
"step": 2040
|
2086 |
+
},
|
2087 |
+
{
|
2088 |
+
"epoch": 72.56637168141593,
|
2089 |
+
"grad_norm": 7.9339704513549805,
|
2090 |
+
"learning_rate": 2.166005291005291e-05,
|
2091 |
+
"loss": 0.0439,
|
2092 |
+
"step": 2050
|
2093 |
+
},
|
2094 |
+
{
|
2095 |
+
"epoch": 72.92035398230088,
|
2096 |
+
"grad_norm": 19.436279296875,
|
2097 |
+
"learning_rate": 2.1494708994708996e-05,
|
2098 |
+
"loss": 0.0592,
|
2099 |
+
"step": 2060
|
2100 |
+
},
|
2101 |
+
{
|
2102 |
+
"epoch": 72.99115044247787,
|
2103 |
+
"eval_accuracy": 0.8675,
|
2104 |
+
"eval_loss": 0.5844297409057617,
|
2105 |
+
"eval_runtime": 6.4907,
|
2106 |
+
"eval_samples_per_second": 61.627,
|
2107 |
+
"eval_steps_per_second": 2.003,
|
2108 |
+
"step": 2062
|
2109 |
+
},
|
2110 |
+
{
|
2111 |
+
"epoch": 73.27433628318585,
|
2112 |
+
"grad_norm": 8.772773742675781,
|
2113 |
+
"learning_rate": 2.132936507936508e-05,
|
2114 |
+
"loss": 0.0515,
|
2115 |
+
"step": 2070
|
2116 |
+
},
|
2117 |
+
{
|
2118 |
+
"epoch": 73.6283185840708,
|
2119 |
+
"grad_norm": 5.5715651512146,
|
2120 |
+
"learning_rate": 2.1164021164021164e-05,
|
2121 |
+
"loss": 0.0381,
|
2122 |
+
"step": 2080
|
2123 |
+
},
|
2124 |
+
{
|
2125 |
+
"epoch": 73.98230088495575,
|
2126 |
+
"grad_norm": 8.322794914245605,
|
2127 |
+
"learning_rate": 2.0998677248677252e-05,
|
2128 |
+
"loss": 0.0408,
|
2129 |
+
"step": 2090
|
2130 |
+
},
|
2131 |
+
{
|
2132 |
+
"epoch": 73.98230088495575,
|
2133 |
+
"eval_accuracy": 0.86,
|
2134 |
+
"eval_loss": 0.58680659532547,
|
2135 |
+
"eval_runtime": 6.4145,
|
2136 |
+
"eval_samples_per_second": 62.359,
|
2137 |
+
"eval_steps_per_second": 2.027,
|
2138 |
+
"step": 2090
|
2139 |
+
},
|
2140 |
+
{
|
2141 |
+
"epoch": 74.33628318584071,
|
2142 |
+
"grad_norm": 5.4705657958984375,
|
2143 |
+
"learning_rate": 2.0833333333333336e-05,
|
2144 |
+
"loss": 0.0469,
|
2145 |
+
"step": 2100
|
2146 |
+
},
|
2147 |
+
{
|
2148 |
+
"epoch": 74.69026548672566,
|
2149 |
+
"grad_norm": 2.226810932159424,
|
2150 |
+
"learning_rate": 2.066798941798942e-05,
|
2151 |
+
"loss": 0.0358,
|
2152 |
+
"step": 2110
|
2153 |
+
},
|
2154 |
+
{
|
2155 |
+
"epoch": 74.97345132743362,
|
2156 |
+
"eval_accuracy": 0.85,
|
2157 |
+
"eval_loss": 0.6115119457244873,
|
2158 |
+
"eval_runtime": 6.4287,
|
2159 |
+
"eval_samples_per_second": 62.221,
|
2160 |
+
"eval_steps_per_second": 2.022,
|
2161 |
+
"step": 2118
|
2162 |
+
},
|
2163 |
+
{
|
2164 |
+
"epoch": 75.04424778761062,
|
2165 |
+
"grad_norm": 7.011991500854492,
|
2166 |
+
"learning_rate": 2.0502645502645504e-05,
|
2167 |
+
"loss": 0.0292,
|
2168 |
+
"step": 2120
|
2169 |
+
},
|
2170 |
+
{
|
2171 |
+
"epoch": 75.39823008849558,
|
2172 |
+
"grad_norm": 1.7847497463226318,
|
2173 |
+
"learning_rate": 2.033730158730159e-05,
|
2174 |
+
"loss": 0.0262,
|
2175 |
+
"step": 2130
|
2176 |
+
},
|
2177 |
+
{
|
2178 |
+
"epoch": 75.75221238938053,
|
2179 |
+
"grad_norm": 7.093685150146484,
|
2180 |
+
"learning_rate": 2.0171957671957673e-05,
|
2181 |
+
"loss": 0.0412,
|
2182 |
+
"step": 2140
|
2183 |
+
},
|
2184 |
+
{
|
2185 |
+
"epoch": 76.0,
|
2186 |
+
"eval_accuracy": 0.865,
|
2187 |
+
"eval_loss": 0.5940263271331787,
|
2188 |
+
"eval_runtime": 6.4173,
|
2189 |
+
"eval_samples_per_second": 62.331,
|
2190 |
+
"eval_steps_per_second": 2.026,
|
2191 |
+
"step": 2147
|
2192 |
+
},
|
2193 |
+
{
|
2194 |
+
"epoch": 76.10619469026548,
|
2195 |
+
"grad_norm": 6.225839138031006,
|
2196 |
+
"learning_rate": 2.0006613756613757e-05,
|
2197 |
+
"loss": 0.0417,
|
2198 |
+
"step": 2150
|
2199 |
+
},
|
2200 |
+
{
|
2201 |
+
"epoch": 76.46017699115045,
|
2202 |
+
"grad_norm": 2.1384568214416504,
|
2203 |
+
"learning_rate": 1.984126984126984e-05,
|
2204 |
+
"loss": 0.0434,
|
2205 |
+
"step": 2160
|
2206 |
+
},
|
2207 |
+
{
|
2208 |
+
"epoch": 76.8141592920354,
|
2209 |
+
"grad_norm": 5.255856037139893,
|
2210 |
+
"learning_rate": 1.967592592592593e-05,
|
2211 |
+
"loss": 0.0323,
|
2212 |
+
"step": 2170
|
2213 |
+
},
|
2214 |
+
{
|
2215 |
+
"epoch": 76.99115044247787,
|
2216 |
+
"eval_accuracy": 0.8625,
|
2217 |
+
"eval_loss": 0.5752249956130981,
|
2218 |
+
"eval_runtime": 6.4807,
|
2219 |
+
"eval_samples_per_second": 61.722,
|
2220 |
+
"eval_steps_per_second": 2.006,
|
2221 |
+
"step": 2175
|
2222 |
+
},
|
2223 |
+
{
|
2224 |
+
"epoch": 77.16814159292035,
|
2225 |
+
"grad_norm": 0.49754026532173157,
|
2226 |
+
"learning_rate": 1.9510582010582013e-05,
|
2227 |
+
"loss": 0.0221,
|
2228 |
+
"step": 2180
|
2229 |
+
},
|
2230 |
+
{
|
2231 |
+
"epoch": 77.52212389380531,
|
2232 |
+
"grad_norm": 12.235739707946777,
|
2233 |
+
"learning_rate": 1.9345238095238097e-05,
|
2234 |
+
"loss": 0.0538,
|
2235 |
+
"step": 2190
|
2236 |
+
},
|
2237 |
+
{
|
2238 |
+
"epoch": 77.87610619469027,
|
2239 |
+
"grad_norm": 2.086639404296875,
|
2240 |
+
"learning_rate": 1.917989417989418e-05,
|
2241 |
+
"loss": 0.0378,
|
2242 |
+
"step": 2200
|
2243 |
+
},
|
2244 |
+
{
|
2245 |
+
"epoch": 77.98230088495575,
|
2246 |
+
"eval_accuracy": 0.8725,
|
2247 |
+
"eval_loss": 0.5514724850654602,
|
2248 |
+
"eval_runtime": 6.5626,
|
2249 |
+
"eval_samples_per_second": 60.952,
|
2250 |
+
"eval_steps_per_second": 1.981,
|
2251 |
+
"step": 2203
|
2252 |
+
},
|
2253 |
+
{
|
2254 |
+
"epoch": 78.23008849557522,
|
2255 |
+
"grad_norm": 7.419439792633057,
|
2256 |
+
"learning_rate": 1.9014550264550266e-05,
|
2257 |
+
"loss": 0.0261,
|
2258 |
+
"step": 2210
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 78.58407079646018,
|
2262 |
+
"grad_norm": 10.227485656738281,
|
2263 |
+
"learning_rate": 1.884920634920635e-05,
|
2264 |
+
"loss": 0.036,
|
2265 |
+
"step": 2220
|
2266 |
+
},
|
2267 |
+
{
|
2268 |
+
"epoch": 78.93805309734513,
|
2269 |
+
"grad_norm": 2.4315621852874756,
|
2270 |
+
"learning_rate": 1.8683862433862434e-05,
|
2271 |
+
"loss": 0.0359,
|
2272 |
+
"step": 2230
|
2273 |
+
},
|
2274 |
+
{
|
2275 |
+
"epoch": 78.97345132743362,
|
2276 |
+
"eval_accuracy": 0.8775,
|
2277 |
+
"eval_loss": 0.5909903049468994,
|
2278 |
+
"eval_runtime": 6.4947,
|
2279 |
+
"eval_samples_per_second": 61.589,
|
2280 |
+
"eval_steps_per_second": 2.002,
|
2281 |
+
"step": 2231
|
2282 |
+
},
|
2283 |
+
{
|
2284 |
+
"epoch": 79.29203539823008,
|
2285 |
+
"grad_norm": 4.286372184753418,
|
2286 |
+
"learning_rate": 1.8518518518518518e-05,
|
2287 |
+
"loss": 0.0325,
|
2288 |
+
"step": 2240
|
2289 |
+
},
|
2290 |
+
{
|
2291 |
+
"epoch": 79.64601769911505,
|
2292 |
+
"grad_norm": 1.8956272602081299,
|
2293 |
+
"learning_rate": 1.8353174603174602e-05,
|
2294 |
+
"loss": 0.0374,
|
2295 |
+
"step": 2250
|
2296 |
+
},
|
2297 |
+
{
|
2298 |
+
"epoch": 80.0,
|
2299 |
+
"grad_norm": 7.963189125061035,
|
2300 |
+
"learning_rate": 1.818783068783069e-05,
|
2301 |
+
"loss": 0.028,
|
2302 |
+
"step": 2260
|
2303 |
+
},
|
2304 |
+
{
|
2305 |
+
"epoch": 80.0,
|
2306 |
+
"eval_accuracy": 0.8725,
|
2307 |
+
"eval_loss": 0.6060264110565186,
|
2308 |
+
"eval_runtime": 6.7895,
|
2309 |
+
"eval_samples_per_second": 58.914,
|
2310 |
+
"eval_steps_per_second": 1.915,
|
2311 |
+
"step": 2260
|
2312 |
+
},
|
2313 |
+
{
|
2314 |
+
"epoch": 80.35398230088495,
|
2315 |
+
"grad_norm": 0.9016655683517456,
|
2316 |
+
"learning_rate": 1.8022486772486774e-05,
|
2317 |
+
"loss": 0.0256,
|
2318 |
+
"step": 2270
|
2319 |
+
},
|
2320 |
+
{
|
2321 |
+
"epoch": 80.70796460176992,
|
2322 |
+
"grad_norm": 1.1370468139648438,
|
2323 |
+
"learning_rate": 1.785714285714286e-05,
|
2324 |
+
"loss": 0.032,
|
2325 |
+
"step": 2280
|
2326 |
+
},
|
2327 |
+
{
|
2328 |
+
"epoch": 80.99115044247787,
|
2329 |
+
"eval_accuracy": 0.8775,
|
2330 |
+
"eval_loss": 0.6053940057754517,
|
2331 |
+
"eval_runtime": 6.5102,
|
2332 |
+
"eval_samples_per_second": 61.442,
|
2333 |
+
"eval_steps_per_second": 1.997,
|
2334 |
+
"step": 2288
|
2335 |
+
},
|
2336 |
+
{
|
2337 |
+
"epoch": 81.06194690265487,
|
2338 |
+
"grad_norm": 1.3163762092590332,
|
2339 |
+
"learning_rate": 1.7691798941798943e-05,
|
2340 |
+
"loss": 0.0392,
|
2341 |
+
"step": 2290
|
2342 |
+
},
|
2343 |
+
{
|
2344 |
+
"epoch": 81.41592920353982,
|
2345 |
+
"grad_norm": 5.895244598388672,
|
2346 |
+
"learning_rate": 1.7526455026455027e-05,
|
2347 |
+
"loss": 0.032,
|
2348 |
+
"step": 2300
|
2349 |
+
},
|
2350 |
+
{
|
2351 |
+
"epoch": 81.76991150442478,
|
2352 |
+
"grad_norm": 2.7035818099975586,
|
2353 |
+
"learning_rate": 1.736111111111111e-05,
|
2354 |
+
"loss": 0.032,
|
2355 |
+
"step": 2310
|
2356 |
+
},
|
2357 |
+
{
|
2358 |
+
"epoch": 81.98230088495575,
|
2359 |
+
"eval_accuracy": 0.8725,
|
2360 |
+
"eval_loss": 0.6311793923377991,
|
2361 |
+
"eval_runtime": 6.4019,
|
2362 |
+
"eval_samples_per_second": 62.482,
|
2363 |
+
"eval_steps_per_second": 2.031,
|
2364 |
+
"step": 2316
|
2365 |
+
},
|
2366 |
+
{
|
2367 |
+
"epoch": 82.12389380530973,
|
2368 |
+
"grad_norm": 1.024576187133789,
|
2369 |
+
"learning_rate": 1.7195767195767195e-05,
|
2370 |
+
"loss": 0.0224,
|
2371 |
+
"step": 2320
|
2372 |
+
},
|
2373 |
+
{
|
2374 |
+
"epoch": 82.47787610619469,
|
2375 |
+
"grad_norm": 5.964217662811279,
|
2376 |
+
"learning_rate": 1.703042328042328e-05,
|
2377 |
+
"loss": 0.0328,
|
2378 |
+
"step": 2330
|
2379 |
+
},
|
2380 |
+
{
|
2381 |
+
"epoch": 82.83185840707965,
|
2382 |
+
"grad_norm": 1.945139765739441,
|
2383 |
+
"learning_rate": 1.6865079365079367e-05,
|
2384 |
+
"loss": 0.0228,
|
2385 |
+
"step": 2340
|
2386 |
+
},
|
2387 |
+
{
|
2388 |
+
"epoch": 82.97345132743362,
|
2389 |
+
"eval_accuracy": 0.87,
|
2390 |
+
"eval_loss": 0.6153484582901001,
|
2391 |
+
"eval_runtime": 6.5056,
|
2392 |
+
"eval_samples_per_second": 61.485,
|
2393 |
+
"eval_steps_per_second": 1.998,
|
2394 |
+
"step": 2344
|
2395 |
+
},
|
2396 |
+
{
|
2397 |
+
"epoch": 83.1858407079646,
|
2398 |
+
"grad_norm": 3.917052745819092,
|
2399 |
+
"learning_rate": 1.669973544973545e-05,
|
2400 |
+
"loss": 0.0274,
|
2401 |
+
"step": 2350
|
2402 |
+
},
|
2403 |
+
{
|
2404 |
+
"epoch": 83.53982300884955,
|
2405 |
+
"grad_norm": 3.854186773300171,
|
2406 |
+
"learning_rate": 1.6534391534391536e-05,
|
2407 |
+
"loss": 0.0374,
|
2408 |
+
"step": 2360
|
2409 |
+
},
|
2410 |
+
{
|
2411 |
+
"epoch": 83.89380530973452,
|
2412 |
+
"grad_norm": 5.1352152824401855,
|
2413 |
+
"learning_rate": 1.636904761904762e-05,
|
2414 |
+
"loss": 0.0457,
|
2415 |
+
"step": 2370
|
2416 |
+
},
|
2417 |
+
{
|
2418 |
+
"epoch": 84.0,
|
2419 |
+
"eval_accuracy": 0.86,
|
2420 |
+
"eval_loss": 0.6442777514457703,
|
2421 |
+
"eval_runtime": 6.6007,
|
2422 |
+
"eval_samples_per_second": 60.6,
|
2423 |
+
"eval_steps_per_second": 1.969,
|
2424 |
+
"step": 2373
|
2425 |
+
},
|
2426 |
+
{
|
2427 |
+
"epoch": 84.24778761061947,
|
2428 |
+
"grad_norm": 7.320106029510498,
|
2429 |
+
"learning_rate": 1.6203703703703704e-05,
|
2430 |
+
"loss": 0.046,
|
2431 |
+
"step": 2380
|
2432 |
+
},
|
2433 |
+
{
|
2434 |
+
"epoch": 84.60176991150442,
|
2435 |
+
"grad_norm": 4.170300006866455,
|
2436 |
+
"learning_rate": 1.6038359788359788e-05,
|
2437 |
+
"loss": 0.0253,
|
2438 |
+
"step": 2390
|
2439 |
+
},
|
2440 |
+
{
|
2441 |
+
"epoch": 84.95575221238938,
|
2442 |
+
"grad_norm": 4.205738544464111,
|
2443 |
+
"learning_rate": 1.5873015873015872e-05,
|
2444 |
+
"loss": 0.0248,
|
2445 |
+
"step": 2400
|
2446 |
+
},
|
2447 |
+
{
|
2448 |
+
"epoch": 84.99115044247787,
|
2449 |
+
"eval_accuracy": 0.875,
|
2450 |
+
"eval_loss": 0.5725922584533691,
|
2451 |
+
"eval_runtime": 6.3567,
|
2452 |
+
"eval_samples_per_second": 62.926,
|
2453 |
+
"eval_steps_per_second": 2.045,
|
2454 |
+
"step": 2401
|
2455 |
+
},
|
2456 |
+
{
|
2457 |
+
"epoch": 85.30973451327434,
|
2458 |
+
"grad_norm": 8.318975448608398,
|
2459 |
+
"learning_rate": 1.5707671957671957e-05,
|
2460 |
+
"loss": 0.0447,
|
2461 |
+
"step": 2410
|
2462 |
+
},
|
2463 |
+
{
|
2464 |
+
"epoch": 85.66371681415929,
|
2465 |
+
"grad_norm": 4.4113311767578125,
|
2466 |
+
"learning_rate": 1.5542328042328044e-05,
|
2467 |
+
"loss": 0.0405,
|
2468 |
+
"step": 2420
|
2469 |
+
},
|
2470 |
+
{
|
2471 |
+
"epoch": 85.98230088495575,
|
2472 |
+
"eval_accuracy": 0.875,
|
2473 |
+
"eval_loss": 0.6042489409446716,
|
2474 |
+
"eval_runtime": 6.4207,
|
2475 |
+
"eval_samples_per_second": 62.298,
|
2476 |
+
"eval_steps_per_second": 2.025,
|
2477 |
+
"step": 2429
|
2478 |
+
},
|
2479 |
+
{
|
2480 |
+
"epoch": 86.01769911504425,
|
2481 |
+
"grad_norm": 5.3457159996032715,
|
2482 |
+
"learning_rate": 1.537698412698413e-05,
|
2483 |
+
"loss": 0.0326,
|
2484 |
+
"step": 2430
|
2485 |
+
},
|
2486 |
+
{
|
2487 |
+
"epoch": 86.3716814159292,
|
2488 |
+
"grad_norm": 6.311764717102051,
|
2489 |
+
"learning_rate": 1.5211640211640213e-05,
|
2490 |
+
"loss": 0.0263,
|
2491 |
+
"step": 2440
|
2492 |
+
},
|
2493 |
+
{
|
2494 |
+
"epoch": 86.72566371681415,
|
2495 |
+
"grad_norm": 5.146406650543213,
|
2496 |
+
"learning_rate": 1.5046296296296297e-05,
|
2497 |
+
"loss": 0.0203,
|
2498 |
+
"step": 2450
|
2499 |
+
},
|
2500 |
+
{
|
2501 |
+
"epoch": 86.97345132743362,
|
2502 |
+
"eval_accuracy": 0.87,
|
2503 |
+
"eval_loss": 0.6107016801834106,
|
2504 |
+
"eval_runtime": 6.5468,
|
2505 |
+
"eval_samples_per_second": 61.098,
|
2506 |
+
"eval_steps_per_second": 1.986,
|
2507 |
+
"step": 2457
|
2508 |
+
},
|
2509 |
+
{
|
2510 |
+
"epoch": 87.07964601769912,
|
2511 |
+
"grad_norm": 1.7579360008239746,
|
2512 |
+
"learning_rate": 1.4880952380952381e-05,
|
2513 |
+
"loss": 0.0358,
|
2514 |
+
"step": 2460
|
2515 |
+
},
|
2516 |
+
{
|
2517 |
+
"epoch": 87.43362831858407,
|
2518 |
+
"grad_norm": 4.512058258056641,
|
2519 |
+
"learning_rate": 1.4715608465608465e-05,
|
2520 |
+
"loss": 0.0485,
|
2521 |
+
"step": 2470
|
2522 |
+
},
|
2523 |
+
{
|
2524 |
+
"epoch": 87.78761061946902,
|
2525 |
+
"grad_norm": 11.72947883605957,
|
2526 |
+
"learning_rate": 1.455026455026455e-05,
|
2527 |
+
"loss": 0.0557,
|
2528 |
+
"step": 2480
|
2529 |
+
},
|
2530 |
+
{
|
2531 |
+
"epoch": 88.0,
|
2532 |
+
"eval_accuracy": 0.88,
|
2533 |
+
"eval_loss": 0.5889678001403809,
|
2534 |
+
"eval_runtime": 6.3973,
|
2535 |
+
"eval_samples_per_second": 62.527,
|
2536 |
+
"eval_steps_per_second": 2.032,
|
2537 |
+
"step": 2486
|
2538 |
+
},
|
2539 |
+
{
|
2540 |
+
"epoch": 88.14159292035399,
|
2541 |
+
"grad_norm": 15.040348052978516,
|
2542 |
+
"learning_rate": 1.4384920634920635e-05,
|
2543 |
+
"loss": 0.0306,
|
2544 |
+
"step": 2490
|
2545 |
+
},
|
2546 |
+
{
|
2547 |
+
"epoch": 88.49557522123894,
|
2548 |
+
"grad_norm": 6.331680774688721,
|
2549 |
+
"learning_rate": 1.4219576719576721e-05,
|
2550 |
+
"loss": 0.0359,
|
2551 |
+
"step": 2500
|
2552 |
+
},
|
2553 |
+
{
|
2554 |
+
"epoch": 88.84955752212389,
|
2555 |
+
"grad_norm": 5.7502827644348145,
|
2556 |
+
"learning_rate": 1.4054232804232805e-05,
|
2557 |
+
"loss": 0.0302,
|
2558 |
+
"step": 2510
|
2559 |
+
},
|
2560 |
+
{
|
2561 |
+
"epoch": 88.99115044247787,
|
2562 |
+
"eval_accuracy": 0.8625,
|
2563 |
+
"eval_loss": 0.5778038501739502,
|
2564 |
+
"eval_runtime": 6.345,
|
2565 |
+
"eval_samples_per_second": 63.042,
|
2566 |
+
"eval_steps_per_second": 2.049,
|
2567 |
+
"step": 2514
|
2568 |
+
},
|
2569 |
+
{
|
2570 |
+
"epoch": 89.20353982300885,
|
2571 |
+
"grad_norm": 8.19079303741455,
|
2572 |
+
"learning_rate": 1.388888888888889e-05,
|
2573 |
+
"loss": 0.0338,
|
2574 |
+
"step": 2520
|
2575 |
+
},
|
2576 |
+
{
|
2577 |
+
"epoch": 89.5575221238938,
|
2578 |
+
"grad_norm": 3.228179693222046,
|
2579 |
+
"learning_rate": 1.3723544973544974e-05,
|
2580 |
+
"loss": 0.0292,
|
2581 |
+
"step": 2530
|
2582 |
+
},
|
2583 |
+
{
|
2584 |
+
"epoch": 89.91150442477876,
|
2585 |
+
"grad_norm": 6.094960689544678,
|
2586 |
+
"learning_rate": 1.3558201058201058e-05,
|
2587 |
+
"loss": 0.0268,
|
2588 |
+
"step": 2540
|
2589 |
+
},
|
2590 |
+
{
|
2591 |
+
"epoch": 89.98230088495575,
|
2592 |
+
"eval_accuracy": 0.8625,
|
2593 |
+
"eval_loss": 0.6038649082183838,
|
2594 |
+
"eval_runtime": 6.4685,
|
2595 |
+
"eval_samples_per_second": 61.838,
|
2596 |
+
"eval_steps_per_second": 2.01,
|
2597 |
+
"step": 2542
|
2598 |
+
},
|
2599 |
+
{
|
2600 |
+
"epoch": 90.26548672566372,
|
2601 |
+
"grad_norm": 5.123661994934082,
|
2602 |
+
"learning_rate": 1.3392857142857144e-05,
|
2603 |
+
"loss": 0.0239,
|
2604 |
+
"step": 2550
|
2605 |
+
},
|
2606 |
+
{
|
2607 |
+
"epoch": 90.61946902654867,
|
2608 |
+
"grad_norm": 6.286872386932373,
|
2609 |
+
"learning_rate": 1.3227513227513228e-05,
|
2610 |
+
"loss": 0.0295,
|
2611 |
+
"step": 2560
|
2612 |
+
},
|
2613 |
+
{
|
2614 |
+
"epoch": 90.97345132743362,
|
2615 |
+
"grad_norm": 8.014678955078125,
|
2616 |
+
"learning_rate": 1.3062169312169312e-05,
|
2617 |
+
"loss": 0.0313,
|
2618 |
+
"step": 2570
|
2619 |
+
},
|
2620 |
+
{
|
2621 |
+
"epoch": 90.97345132743362,
|
2622 |
+
"eval_accuracy": 0.885,
|
2623 |
+
"eval_loss": 0.5608097314834595,
|
2624 |
+
"eval_runtime": 6.3306,
|
2625 |
+
"eval_samples_per_second": 63.185,
|
2626 |
+
"eval_steps_per_second": 2.054,
|
2627 |
+
"step": 2570
|
2628 |
+
},
|
2629 |
+
{
|
2630 |
+
"epoch": 91.32743362831859,
|
2631 |
+
"grad_norm": 8.690397262573242,
|
2632 |
+
"learning_rate": 1.2896825396825398e-05,
|
2633 |
+
"loss": 0.028,
|
2634 |
+
"step": 2580
|
2635 |
+
},
|
2636 |
+
{
|
2637 |
+
"epoch": 91.68141592920354,
|
2638 |
+
"grad_norm": 4.610723972320557,
|
2639 |
+
"learning_rate": 1.2731481481481482e-05,
|
2640 |
+
"loss": 0.0227,
|
2641 |
+
"step": 2590
|
2642 |
+
},
|
2643 |
+
{
|
2644 |
+
"epoch": 92.0,
|
2645 |
+
"eval_accuracy": 0.8625,
|
2646 |
+
"eval_loss": 0.6019301414489746,
|
2647 |
+
"eval_runtime": 6.3561,
|
2648 |
+
"eval_samples_per_second": 62.931,
|
2649 |
+
"eval_steps_per_second": 2.045,
|
2650 |
+
"step": 2599
|
2651 |
+
},
|
2652 |
+
{
|
2653 |
+
"epoch": 92.03539823008849,
|
2654 |
+
"grad_norm": 33.1154899597168,
|
2655 |
+
"learning_rate": 1.2566137566137568e-05,
|
2656 |
+
"loss": 0.034,
|
2657 |
+
"step": 2600
|
2658 |
+
},
|
2659 |
+
{
|
2660 |
+
"epoch": 92.38938053097345,
|
2661 |
+
"grad_norm": 5.2956013679504395,
|
2662 |
+
"learning_rate": 1.2400793650793652e-05,
|
2663 |
+
"loss": 0.0393,
|
2664 |
+
"step": 2610
|
2665 |
+
},
|
2666 |
+
{
|
2667 |
+
"epoch": 92.7433628318584,
|
2668 |
+
"grad_norm": 3.9011313915252686,
|
2669 |
+
"learning_rate": 1.2235449735449737e-05,
|
2670 |
+
"loss": 0.0277,
|
2671 |
+
"step": 2620
|
2672 |
+
},
|
2673 |
+
{
|
2674 |
+
"epoch": 92.99115044247787,
|
2675 |
+
"eval_accuracy": 0.8675,
|
2676 |
+
"eval_loss": 0.5949091911315918,
|
2677 |
+
"eval_runtime": 6.4133,
|
2678 |
+
"eval_samples_per_second": 62.37,
|
2679 |
+
"eval_steps_per_second": 2.027,
|
2680 |
+
"step": 2627
|
2681 |
+
},
|
2682 |
+
{
|
2683 |
+
"epoch": 93.09734513274336,
|
2684 |
+
"grad_norm": 7.499249458312988,
|
2685 |
+
"learning_rate": 1.2070105820105821e-05,
|
2686 |
+
"loss": 0.0338,
|
2687 |
+
"step": 2630
|
2688 |
+
},
|
2689 |
+
{
|
2690 |
+
"epoch": 93.45132743362832,
|
2691 |
+
"grad_norm": 10.214336395263672,
|
2692 |
+
"learning_rate": 1.1904761904761905e-05,
|
2693 |
+
"loss": 0.0219,
|
2694 |
+
"step": 2640
|
2695 |
+
},
|
2696 |
+
{
|
2697 |
+
"epoch": 93.80530973451327,
|
2698 |
+
"grad_norm": 11.573188781738281,
|
2699 |
+
"learning_rate": 1.1739417989417991e-05,
|
2700 |
+
"loss": 0.0378,
|
2701 |
+
"step": 2650
|
2702 |
+
},
|
2703 |
+
{
|
2704 |
+
"epoch": 93.98230088495575,
|
2705 |
+
"eval_accuracy": 0.875,
|
2706 |
+
"eval_loss": 0.578478217124939,
|
2707 |
+
"eval_runtime": 6.319,
|
2708 |
+
"eval_samples_per_second": 63.301,
|
2709 |
+
"eval_steps_per_second": 2.057,
|
2710 |
+
"step": 2655
|
2711 |
+
},
|
2712 |
+
{
|
2713 |
+
"epoch": 94.15929203539822,
|
2714 |
+
"grad_norm": 4.489903926849365,
|
2715 |
+
"learning_rate": 1.1574074074074075e-05,
|
2716 |
+
"loss": 0.0254,
|
2717 |
+
"step": 2660
|
2718 |
+
},
|
2719 |
+
{
|
2720 |
+
"epoch": 94.51327433628319,
|
2721 |
+
"grad_norm": 3.9934630393981934,
|
2722 |
+
"learning_rate": 1.140873015873016e-05,
|
2723 |
+
"loss": 0.0288,
|
2724 |
+
"step": 2670
|
2725 |
+
},
|
2726 |
+
{
|
2727 |
+
"epoch": 94.86725663716814,
|
2728 |
+
"grad_norm": 7.336540699005127,
|
2729 |
+
"learning_rate": 1.1243386243386244e-05,
|
2730 |
+
"loss": 0.0381,
|
2731 |
+
"step": 2680
|
2732 |
+
},
|
2733 |
+
{
|
2734 |
+
"epoch": 94.97345132743362,
|
2735 |
+
"eval_accuracy": 0.8825,
|
2736 |
+
"eval_loss": 0.5645653009414673,
|
2737 |
+
"eval_runtime": 6.4399,
|
2738 |
+
"eval_samples_per_second": 62.113,
|
2739 |
+
"eval_steps_per_second": 2.019,
|
2740 |
+
"step": 2683
|
2741 |
+
},
|
2742 |
+
{
|
2743 |
+
"epoch": 95.22123893805309,
|
2744 |
+
"grad_norm": 5.17459774017334,
|
2745 |
+
"learning_rate": 1.107804232804233e-05,
|
2746 |
+
"loss": 0.0347,
|
2747 |
+
"step": 2690
|
2748 |
+
},
|
2749 |
+
{
|
2750 |
+
"epoch": 95.57522123893806,
|
2751 |
+
"grad_norm": 10.617063522338867,
|
2752 |
+
"learning_rate": 1.0912698412698414e-05,
|
2753 |
+
"loss": 0.0338,
|
2754 |
+
"step": 2700
|
2755 |
+
},
|
2756 |
+
{
|
2757 |
+
"epoch": 95.929203539823,
|
2758 |
+
"grad_norm": 8.908598899841309,
|
2759 |
+
"learning_rate": 1.0747354497354498e-05,
|
2760 |
+
"loss": 0.0435,
|
2761 |
+
"step": 2710
|
2762 |
+
},
|
2763 |
+
{
|
2764 |
+
"epoch": 96.0,
|
2765 |
+
"eval_accuracy": 0.88,
|
2766 |
+
"eval_loss": 0.5512508153915405,
|
2767 |
+
"eval_runtime": 6.387,
|
2768 |
+
"eval_samples_per_second": 62.627,
|
2769 |
+
"eval_steps_per_second": 2.035,
|
2770 |
+
"step": 2712
|
2771 |
+
},
|
2772 |
+
{
|
2773 |
+
"epoch": 96.28318584070796,
|
2774 |
+
"grad_norm": 3.1755924224853516,
|
2775 |
+
"learning_rate": 1.0582010582010582e-05,
|
2776 |
+
"loss": 0.0417,
|
2777 |
+
"step": 2720
|
2778 |
+
},
|
2779 |
+
{
|
2780 |
+
"epoch": 96.63716814159292,
|
2781 |
+
"grad_norm": 6.078367233276367,
|
2782 |
+
"learning_rate": 1.0416666666666668e-05,
|
2783 |
+
"loss": 0.0268,
|
2784 |
+
"step": 2730
|
2785 |
+
},
|
2786 |
+
{
|
2787 |
+
"epoch": 96.99115044247787,
|
2788 |
+
"grad_norm": 11.641453742980957,
|
2789 |
+
"learning_rate": 1.0251322751322752e-05,
|
2790 |
+
"loss": 0.0264,
|
2791 |
+
"step": 2740
|
2792 |
+
},
|
2793 |
+
{
|
2794 |
+
"epoch": 96.99115044247787,
|
2795 |
+
"eval_accuracy": 0.875,
|
2796 |
+
"eval_loss": 0.5257237553596497,
|
2797 |
+
"eval_runtime": 6.678,
|
2798 |
+
"eval_samples_per_second": 59.898,
|
2799 |
+
"eval_steps_per_second": 1.947,
|
2800 |
+
"step": 2740
|
2801 |
+
},
|
2802 |
+
{
|
2803 |
+
"epoch": 97.34513274336283,
|
2804 |
+
"grad_norm": 14.086902618408203,
|
2805 |
+
"learning_rate": 1.0085978835978836e-05,
|
2806 |
+
"loss": 0.0489,
|
2807 |
+
"step": 2750
|
2808 |
+
},
|
2809 |
+
{
|
2810 |
+
"epoch": 97.69911504424779,
|
2811 |
+
"grad_norm": 2.605114698410034,
|
2812 |
+
"learning_rate": 9.92063492063492e-06,
|
2813 |
+
"loss": 0.0362,
|
2814 |
+
"step": 2760
|
2815 |
+
},
|
2816 |
+
{
|
2817 |
+
"epoch": 97.98230088495575,
|
2818 |
+
"eval_accuracy": 0.8825,
|
2819 |
+
"eval_loss": 0.5332143306732178,
|
2820 |
+
"eval_runtime": 6.5062,
|
2821 |
+
"eval_samples_per_second": 61.48,
|
2822 |
+
"eval_steps_per_second": 1.998,
|
2823 |
+
"step": 2768
|
2824 |
+
},
|
2825 |
+
{
|
2826 |
+
"epoch": 98.05309734513274,
|
2827 |
+
"grad_norm": 5.3540730476379395,
|
2828 |
+
"learning_rate": 9.755291005291007e-06,
|
2829 |
+
"loss": 0.0286,
|
2830 |
+
"step": 2770
|
2831 |
+
},
|
2832 |
+
{
|
2833 |
+
"epoch": 98.40707964601769,
|
2834 |
+
"grad_norm": 8.65832233428955,
|
2835 |
+
"learning_rate": 9.58994708994709e-06,
|
2836 |
+
"loss": 0.0345,
|
2837 |
+
"step": 2780
|
2838 |
+
},
|
2839 |
+
{
|
2840 |
+
"epoch": 98.76106194690266,
|
2841 |
+
"grad_norm": 1.0023494958877563,
|
2842 |
+
"learning_rate": 9.424603174603175e-06,
|
2843 |
+
"loss": 0.0209,
|
2844 |
+
"step": 2790
|
2845 |
+
},
|
2846 |
+
{
|
2847 |
+
"epoch": 98.97345132743362,
|
2848 |
+
"eval_accuracy": 0.855,
|
2849 |
+
"eval_loss": 0.577680766582489,
|
2850 |
+
"eval_runtime": 6.6535,
|
2851 |
+
"eval_samples_per_second": 60.119,
|
2852 |
+
"eval_steps_per_second": 1.954,
|
2853 |
+
"step": 2796
|
2854 |
+
},
|
2855 |
+
{
|
2856 |
+
"epoch": 99.11504424778761,
|
2857 |
+
"grad_norm": 5.147085666656494,
|
2858 |
+
"learning_rate": 9.259259259259259e-06,
|
2859 |
+
"loss": 0.0245,
|
2860 |
+
"step": 2800
|
2861 |
+
},
|
2862 |
+
{
|
2863 |
+
"epoch": 99.46902654867256,
|
2864 |
+
"grad_norm": 14.565306663513184,
|
2865 |
+
"learning_rate": 9.093915343915345e-06,
|
2866 |
+
"loss": 0.0282,
|
2867 |
+
"step": 2810
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 99.82300884955752,
|
2871 |
+
"grad_norm": 4.94295072555542,
|
2872 |
+
"learning_rate": 8.92857142857143e-06,
|
2873 |
+
"loss": 0.0348,
|
2874 |
+
"step": 2820
|
2875 |
+
},
|
2876 |
+
{
|
2877 |
+
"epoch": 100.0,
|
2878 |
+
"eval_accuracy": 0.8675,
|
2879 |
+
"eval_loss": 0.5673981308937073,
|
2880 |
+
"eval_runtime": 6.516,
|
2881 |
+
"eval_samples_per_second": 61.388,
|
2882 |
+
"eval_steps_per_second": 1.995,
|
2883 |
+
"step": 2825
|
2884 |
+
},
|
2885 |
+
{
|
2886 |
+
"epoch": 100.17699115044248,
|
2887 |
+
"grad_norm": 3.3961915969848633,
|
2888 |
+
"learning_rate": 8.763227513227513e-06,
|
2889 |
+
"loss": 0.0287,
|
2890 |
+
"step": 2830
|
2891 |
+
},
|
2892 |
+
{
|
2893 |
+
"epoch": 100.53097345132744,
|
2894 |
+
"grad_norm": 2.0917205810546875,
|
2895 |
+
"learning_rate": 8.597883597883598e-06,
|
2896 |
+
"loss": 0.0278,
|
2897 |
+
"step": 2840
|
2898 |
+
},
|
2899 |
+
{
|
2900 |
+
"epoch": 100.88495575221239,
|
2901 |
+
"grad_norm": 6.237668991088867,
|
2902 |
+
"learning_rate": 8.432539682539684e-06,
|
2903 |
+
"loss": 0.02,
|
2904 |
+
"step": 2850
|
2905 |
+
},
|
2906 |
+
{
|
2907 |
+
"epoch": 100.99115044247787,
|
2908 |
+
"eval_accuracy": 0.8625,
|
2909 |
+
"eval_loss": 0.5743553638458252,
|
2910 |
+
"eval_runtime": 6.4632,
|
2911 |
+
"eval_samples_per_second": 61.889,
|
2912 |
+
"eval_steps_per_second": 2.011,
|
2913 |
+
"step": 2853
|
2914 |
+
},
|
2915 |
+
{
|
2916 |
+
"epoch": 101.23893805309734,
|
2917 |
+
"grad_norm": 3.717299222946167,
|
2918 |
+
"learning_rate": 8.267195767195768e-06,
|
2919 |
+
"loss": 0.0431,
|
2920 |
+
"step": 2860
|
2921 |
+
},
|
2922 |
+
{
|
2923 |
+
"epoch": 101.59292035398231,
|
2924 |
+
"grad_norm": 4.616144180297852,
|
2925 |
+
"learning_rate": 8.101851851851852e-06,
|
2926 |
+
"loss": 0.0206,
|
2927 |
+
"step": 2870
|
2928 |
+
},
|
2929 |
+
{
|
2930 |
+
"epoch": 101.94690265486726,
|
2931 |
+
"grad_norm": 3.6342599391937256,
|
2932 |
+
"learning_rate": 7.936507936507936e-06,
|
2933 |
+
"loss": 0.0092,
|
2934 |
+
"step": 2880
|
2935 |
+
},
|
2936 |
+
{
|
2937 |
+
"epoch": 101.98230088495575,
|
2938 |
+
"eval_accuracy": 0.8675,
|
2939 |
+
"eval_loss": 0.5852380990982056,
|
2940 |
+
"eval_runtime": 6.5504,
|
2941 |
+
"eval_samples_per_second": 61.065,
|
2942 |
+
"eval_steps_per_second": 1.985,
|
2943 |
+
"step": 2881
|
2944 |
+
},
|
2945 |
+
{
|
2946 |
+
"epoch": 102.30088495575221,
|
2947 |
+
"grad_norm": 3.989046573638916,
|
2948 |
+
"learning_rate": 7.771164021164022e-06,
|
2949 |
+
"loss": 0.0359,
|
2950 |
+
"step": 2890
|
2951 |
+
},
|
2952 |
+
{
|
2953 |
+
"epoch": 102.65486725663717,
|
2954 |
+
"grad_norm": 5.654636859893799,
|
2955 |
+
"learning_rate": 7.605820105820106e-06,
|
2956 |
+
"loss": 0.0343,
|
2957 |
+
"step": 2900
|
2958 |
+
},
|
2959 |
+
{
|
2960 |
+
"epoch": 102.97345132743362,
|
2961 |
+
"eval_accuracy": 0.8675,
|
2962 |
+
"eval_loss": 0.5855699181556702,
|
2963 |
+
"eval_runtime": 6.4985,
|
2964 |
+
"eval_samples_per_second": 61.553,
|
2965 |
+
"eval_steps_per_second": 2.0,
|
2966 |
+
"step": 2909
|
2967 |
+
},
|
2968 |
+
{
|
2969 |
+
"epoch": 103.00884955752213,
|
2970 |
+
"grad_norm": 10.393845558166504,
|
2971 |
+
"learning_rate": 7.4404761904761905e-06,
|
2972 |
+
"loss": 0.0307,
|
2973 |
+
"step": 2910
|
2974 |
+
},
|
2975 |
+
{
|
2976 |
+
"epoch": 103.36283185840708,
|
2977 |
+
"grad_norm": 12.575658798217773,
|
2978 |
+
"learning_rate": 7.275132275132275e-06,
|
2979 |
+
"loss": 0.0435,
|
2980 |
+
"step": 2920
|
2981 |
+
},
|
2982 |
+
{
|
2983 |
+
"epoch": 103.71681415929204,
|
2984 |
+
"grad_norm": 2.3426296710968018,
|
2985 |
+
"learning_rate": 7.1097883597883606e-06,
|
2986 |
+
"loss": 0.0185,
|
2987 |
+
"step": 2930
|
2988 |
+
},
|
2989 |
+
{
|
2990 |
+
"epoch": 104.0,
|
2991 |
+
"eval_accuracy": 0.88,
|
2992 |
+
"eval_loss": 0.566969633102417,
|
2993 |
+
"eval_runtime": 6.7641,
|
2994 |
+
"eval_samples_per_second": 59.135,
|
2995 |
+
"eval_steps_per_second": 1.922,
|
2996 |
+
"step": 2938
|
2997 |
+
},
|
2998 |
+
{
|
2999 |
+
"epoch": 104.070796460177,
|
3000 |
+
"grad_norm": 1.3525067567825317,
|
3001 |
+
"learning_rate": 6.944444444444445e-06,
|
3002 |
+
"loss": 0.0287,
|
3003 |
+
"step": 2940
|
3004 |
+
},
|
3005 |
+
{
|
3006 |
+
"epoch": 104.42477876106194,
|
3007 |
+
"grad_norm": 2.2476863861083984,
|
3008 |
+
"learning_rate": 6.779100529100529e-06,
|
3009 |
+
"loss": 0.021,
|
3010 |
+
"step": 2950
|
3011 |
+
},
|
3012 |
+
{
|
3013 |
+
"epoch": 104.77876106194691,
|
3014 |
+
"grad_norm": 4.498152732849121,
|
3015 |
+
"learning_rate": 6.613756613756614e-06,
|
3016 |
+
"loss": 0.0198,
|
3017 |
+
"step": 2960
|
3018 |
+
},
|
3019 |
+
{
|
3020 |
+
"epoch": 104.99115044247787,
|
3021 |
+
"eval_accuracy": 0.8775,
|
3022 |
+
"eval_loss": 0.5611503720283508,
|
3023 |
+
"eval_runtime": 6.4775,
|
3024 |
+
"eval_samples_per_second": 61.752,
|
3025 |
+
"eval_steps_per_second": 2.007,
|
3026 |
+
"step": 2966
|
3027 |
+
},
|
3028 |
+
{
|
3029 |
+
"epoch": 105.13274336283186,
|
3030 |
+
"grad_norm": 9.938004493713379,
|
3031 |
+
"learning_rate": 6.448412698412699e-06,
|
3032 |
+
"loss": 0.0255,
|
3033 |
+
"step": 2970
|
3034 |
+
},
|
3035 |
+
{
|
3036 |
+
"epoch": 105.48672566371681,
|
3037 |
+
"grad_norm": 1.6436090469360352,
|
3038 |
+
"learning_rate": 6.283068783068784e-06,
|
3039 |
+
"loss": 0.0218,
|
3040 |
+
"step": 2980
|
3041 |
+
},
|
3042 |
+
{
|
3043 |
+
"epoch": 105.84070796460178,
|
3044 |
+
"grad_norm": 3.246063232421875,
|
3045 |
+
"learning_rate": 6.117724867724868e-06,
|
3046 |
+
"loss": 0.016,
|
3047 |
+
"step": 2990
|
3048 |
+
},
|
3049 |
+
{
|
3050 |
+
"epoch": 105.98230088495575,
|
3051 |
+
"eval_accuracy": 0.88,
|
3052 |
+
"eval_loss": 0.5701327323913574,
|
3053 |
+
"eval_runtime": 6.4574,
|
3054 |
+
"eval_samples_per_second": 61.944,
|
3055 |
+
"eval_steps_per_second": 2.013,
|
3056 |
+
"step": 2994
|
3057 |
+
},
|
3058 |
+
{
|
3059 |
+
"epoch": 106.19469026548673,
|
3060 |
+
"grad_norm": 0.46645891666412354,
|
3061 |
+
"learning_rate": 5.9523809523809525e-06,
|
3062 |
+
"loss": 0.0389,
|
3063 |
+
"step": 3000
|
3064 |
+
},
|
3065 |
+
{
|
3066 |
+
"epoch": 106.54867256637168,
|
3067 |
+
"grad_norm": 4.597782135009766,
|
3068 |
+
"learning_rate": 5.787037037037038e-06,
|
3069 |
+
"loss": 0.0216,
|
3070 |
+
"step": 3010
|
3071 |
+
},
|
3072 |
+
{
|
3073 |
+
"epoch": 106.90265486725664,
|
3074 |
+
"grad_norm": 9.084229469299316,
|
3075 |
+
"learning_rate": 5.621693121693122e-06,
|
3076 |
+
"loss": 0.0369,
|
3077 |
+
"step": 3020
|
3078 |
+
},
|
3079 |
+
{
|
3080 |
+
"epoch": 106.97345132743362,
|
3081 |
+
"eval_accuracy": 0.8825,
|
3082 |
+
"eval_loss": 0.5790844559669495,
|
3083 |
+
"eval_runtime": 6.5374,
|
3084 |
+
"eval_samples_per_second": 61.187,
|
3085 |
+
"eval_steps_per_second": 1.989,
|
3086 |
+
"step": 3022
|
3087 |
+
},
|
3088 |
+
{
|
3089 |
+
"epoch": 107.2566371681416,
|
3090 |
+
"grad_norm": 2.540560007095337,
|
3091 |
+
"learning_rate": 5.456349206349207e-06,
|
3092 |
+
"loss": 0.0371,
|
3093 |
+
"step": 3030
|
3094 |
+
},
|
3095 |
+
{
|
3096 |
+
"epoch": 107.61061946902655,
|
3097 |
+
"grad_norm": 1.518036127090454,
|
3098 |
+
"learning_rate": 5.291005291005291e-06,
|
3099 |
+
"loss": 0.0263,
|
3100 |
+
"step": 3040
|
3101 |
+
},
|
3102 |
+
{
|
3103 |
+
"epoch": 107.96460176991151,
|
3104 |
+
"grad_norm": 1.1195813417434692,
|
3105 |
+
"learning_rate": 5.125661375661376e-06,
|
3106 |
+
"loss": 0.0357,
|
3107 |
+
"step": 3050
|
3108 |
+
},
|
3109 |
+
{
|
3110 |
+
"epoch": 108.0,
|
3111 |
+
"eval_accuracy": 0.8725,
|
3112 |
+
"eval_loss": 0.5730476975440979,
|
3113 |
+
"eval_runtime": 8.1599,
|
3114 |
+
"eval_samples_per_second": 49.02,
|
3115 |
+
"eval_steps_per_second": 1.593,
|
3116 |
+
"step": 3051
|
3117 |
+
},
|
3118 |
+
{
|
3119 |
+
"epoch": 108.31858407079646,
|
3120 |
+
"grad_norm": 1.3580831289291382,
|
3121 |
+
"learning_rate": 4.96031746031746e-06,
|
3122 |
+
"loss": 0.0277,
|
3123 |
+
"step": 3060
|
3124 |
+
},
|
3125 |
+
{
|
3126 |
+
"epoch": 108.67256637168141,
|
3127 |
+
"grad_norm": 7.92542839050293,
|
3128 |
+
"learning_rate": 4.794973544973545e-06,
|
3129 |
+
"loss": 0.0361,
|
3130 |
+
"step": 3070
|
3131 |
+
},
|
3132 |
+
{
|
3133 |
+
"epoch": 108.99115044247787,
|
3134 |
+
"eval_accuracy": 0.8725,
|
3135 |
+
"eval_loss": 0.5626590847969055,
|
3136 |
+
"eval_runtime": 6.6363,
|
3137 |
+
"eval_samples_per_second": 60.275,
|
3138 |
+
"eval_steps_per_second": 1.959,
|
3139 |
+
"step": 3079
|
3140 |
+
},
|
3141 |
+
{
|
3142 |
+
"epoch": 109.02654867256638,
|
3143 |
+
"grad_norm": 3.268057107925415,
|
3144 |
+
"learning_rate": 4.6296296296296296e-06,
|
3145 |
+
"loss": 0.0286,
|
3146 |
+
"step": 3080
|
3147 |
+
},
|
3148 |
+
{
|
3149 |
+
"epoch": 109.38053097345133,
|
3150 |
+
"grad_norm": 2.402693510055542,
|
3151 |
+
"learning_rate": 4.464285714285715e-06,
|
3152 |
+
"loss": 0.0252,
|
3153 |
+
"step": 3090
|
3154 |
+
},
|
3155 |
+
{
|
3156 |
+
"epoch": 109.73451327433628,
|
3157 |
+
"grad_norm": 2.8348002433776855,
|
3158 |
+
"learning_rate": 4.298941798941799e-06,
|
3159 |
+
"loss": 0.0438,
|
3160 |
+
"step": 3100
|
3161 |
+
},
|
3162 |
+
{
|
3163 |
+
"epoch": 109.98230088495575,
|
3164 |
+
"eval_accuracy": 0.875,
|
3165 |
+
"eval_loss": 0.5811652541160583,
|
3166 |
+
"eval_runtime": 6.45,
|
3167 |
+
"eval_samples_per_second": 62.016,
|
3168 |
+
"eval_steps_per_second": 2.016,
|
3169 |
+
"step": 3107
|
3170 |
+
},
|
3171 |
+
{
|
3172 |
+
"epoch": 110.08849557522124,
|
3173 |
+
"grad_norm": 1.9933407306671143,
|
3174 |
+
"learning_rate": 4.133597883597884e-06,
|
3175 |
+
"loss": 0.0369,
|
3176 |
+
"step": 3110
|
3177 |
+
},
|
3178 |
+
{
|
3179 |
+
"epoch": 110.4424778761062,
|
3180 |
+
"grad_norm": 5.182060241699219,
|
3181 |
+
"learning_rate": 3.968253968253968e-06,
|
3182 |
+
"loss": 0.0297,
|
3183 |
+
"step": 3120
|
3184 |
+
},
|
3185 |
+
{
|
3186 |
+
"epoch": 110.79646017699115,
|
3187 |
+
"grad_norm": 1.6837078332901,
|
3188 |
+
"learning_rate": 3.802910052910053e-06,
|
3189 |
+
"loss": 0.0243,
|
3190 |
+
"step": 3130
|
3191 |
+
},
|
3192 |
+
{
|
3193 |
+
"epoch": 110.97345132743362,
|
3194 |
+
"eval_accuracy": 0.8725,
|
3195 |
+
"eval_loss": 0.5921982526779175,
|
3196 |
+
"eval_runtime": 6.5644,
|
3197 |
+
"eval_samples_per_second": 60.935,
|
3198 |
+
"eval_steps_per_second": 1.98,
|
3199 |
+
"step": 3135
|
3200 |
+
},
|
3201 |
+
{
|
3202 |
+
"epoch": 111.15044247787611,
|
3203 |
+
"grad_norm": 8.855575561523438,
|
3204 |
+
"learning_rate": 3.6375661375661373e-06,
|
3205 |
+
"loss": 0.0231,
|
3206 |
+
"step": 3140
|
3207 |
+
},
|
3208 |
+
{
|
3209 |
+
"epoch": 111.50442477876106,
|
3210 |
+
"grad_norm": 4.877293586730957,
|
3211 |
+
"learning_rate": 3.4722222222222224e-06,
|
3212 |
+
"loss": 0.0216,
|
3213 |
+
"step": 3150
|
3214 |
+
},
|
3215 |
+
{
|
3216 |
+
"epoch": 111.85840707964601,
|
3217 |
+
"grad_norm": 0.5238684415817261,
|
3218 |
+
"learning_rate": 3.306878306878307e-06,
|
3219 |
+
"loss": 0.0241,
|
3220 |
+
"step": 3160
|
3221 |
+
},
|
3222 |
+
{
|
3223 |
+
"epoch": 112.0,
|
3224 |
+
"eval_accuracy": 0.8775,
|
3225 |
+
"eval_loss": 0.5913088917732239,
|
3226 |
+
"eval_runtime": 6.441,
|
3227 |
+
"eval_samples_per_second": 62.102,
|
3228 |
+
"eval_steps_per_second": 2.018,
|
3229 |
+
"step": 3164
|
3230 |
+
},
|
3231 |
+
{
|
3232 |
+
"epoch": 112.21238938053098,
|
3233 |
+
"grad_norm": 2.535374879837036,
|
3234 |
+
"learning_rate": 3.141534391534392e-06,
|
3235 |
+
"loss": 0.0211,
|
3236 |
+
"step": 3170
|
3237 |
+
},
|
3238 |
+
{
|
3239 |
+
"epoch": 112.56637168141593,
|
3240 |
+
"grad_norm": 2.7834455966949463,
|
3241 |
+
"learning_rate": 2.9761904761904763e-06,
|
3242 |
+
"loss": 0.0262,
|
3243 |
+
"step": 3180
|
3244 |
+
},
|
3245 |
+
{
|
3246 |
+
"epoch": 112.92035398230088,
|
3247 |
+
"grad_norm": 10.206039428710938,
|
3248 |
+
"learning_rate": 2.810846560846561e-06,
|
3249 |
+
"loss": 0.0256,
|
3250 |
+
"step": 3190
|
3251 |
+
},
|
3252 |
+
{
|
3253 |
+
"epoch": 112.99115044247787,
|
3254 |
+
"eval_accuracy": 0.8675,
|
3255 |
+
"eval_loss": 0.5861657857894897,
|
3256 |
+
"eval_runtime": 6.5388,
|
3257 |
+
"eval_samples_per_second": 61.173,
|
3258 |
+
"eval_steps_per_second": 1.988,
|
3259 |
+
"step": 3192
|
3260 |
+
},
|
3261 |
+
{
|
3262 |
+
"epoch": 113.27433628318585,
|
3263 |
+
"grad_norm": 1.7272131443023682,
|
3264 |
+
"learning_rate": 2.6455026455026455e-06,
|
3265 |
+
"loss": 0.0222,
|
3266 |
+
"step": 3200
|
3267 |
+
},
|
3268 |
+
{
|
3269 |
+
"epoch": 113.6283185840708,
|
3270 |
+
"grad_norm": 0.809811532497406,
|
3271 |
+
"learning_rate": 2.48015873015873e-06,
|
3272 |
+
"loss": 0.0196,
|
3273 |
+
"step": 3210
|
3274 |
+
},
|
3275 |
+
{
|
3276 |
+
"epoch": 113.98230088495575,
|
3277 |
+
"grad_norm": 2.7351064682006836,
|
3278 |
+
"learning_rate": 2.3148148148148148e-06,
|
3279 |
+
"loss": 0.0247,
|
3280 |
+
"step": 3220
|
3281 |
+
},
|
3282 |
+
{
|
3283 |
+
"epoch": 113.98230088495575,
|
3284 |
+
"eval_accuracy": 0.8675,
|
3285 |
+
"eval_loss": 0.5812863111495972,
|
3286 |
+
"eval_runtime": 6.4622,
|
3287 |
+
"eval_samples_per_second": 61.899,
|
3288 |
+
"eval_steps_per_second": 2.012,
|
3289 |
+
"step": 3220
|
3290 |
+
},
|
3291 |
+
{
|
3292 |
+
"epoch": 114.33628318584071,
|
3293 |
+
"grad_norm": 0.49953562021255493,
|
3294 |
+
"learning_rate": 2.1494708994708994e-06,
|
3295 |
+
"loss": 0.0193,
|
3296 |
+
"step": 3230
|
3297 |
+
},
|
3298 |
+
{
|
3299 |
+
"epoch": 114.69026548672566,
|
3300 |
+
"grad_norm": 4.690150260925293,
|
3301 |
+
"learning_rate": 1.984126984126984e-06,
|
3302 |
+
"loss": 0.028,
|
3303 |
+
"step": 3240
|
3304 |
+
},
|
3305 |
+
{
|
3306 |
+
"epoch": 114.97345132743362,
|
3307 |
+
"eval_accuracy": 0.87,
|
3308 |
+
"eval_loss": 0.5752473473548889,
|
3309 |
+
"eval_runtime": 6.3377,
|
3310 |
+
"eval_samples_per_second": 63.114,
|
3311 |
+
"eval_steps_per_second": 2.051,
|
3312 |
+
"step": 3248
|
3313 |
+
},
|
3314 |
+
{
|
3315 |
+
"epoch": 115.04424778761062,
|
3316 |
+
"grad_norm": 3.4690709114074707,
|
3317 |
+
"learning_rate": 1.8187830687830687e-06,
|
3318 |
+
"loss": 0.039,
|
3319 |
+
"step": 3250
|
3320 |
+
},
|
3321 |
+
{
|
3322 |
+
"epoch": 115.39823008849558,
|
3323 |
+
"grad_norm": 7.698827743530273,
|
3324 |
+
"learning_rate": 1.6534391534391535e-06,
|
3325 |
+
"loss": 0.0261,
|
3326 |
+
"step": 3260
|
3327 |
+
},
|
3328 |
+
{
|
3329 |
+
"epoch": 115.75221238938053,
|
3330 |
+
"grad_norm": 2.016017198562622,
|
3331 |
+
"learning_rate": 1.4880952380952381e-06,
|
3332 |
+
"loss": 0.0177,
|
3333 |
+
"step": 3270
|
3334 |
+
},
|
3335 |
+
{
|
3336 |
+
"epoch": 116.0,
|
3337 |
+
"eval_accuracy": 0.87,
|
3338 |
+
"eval_loss": 0.574239194393158,
|
3339 |
+
"eval_runtime": 6.3905,
|
3340 |
+
"eval_samples_per_second": 62.593,
|
3341 |
+
"eval_steps_per_second": 2.034,
|
3342 |
+
"step": 3277
|
3343 |
+
},
|
3344 |
+
{
|
3345 |
+
"epoch": 116.10619469026548,
|
3346 |
+
"grad_norm": 5.834602355957031,
|
3347 |
+
"learning_rate": 1.3227513227513228e-06,
|
3348 |
+
"loss": 0.0129,
|
3349 |
+
"step": 3280
|
3350 |
+
},
|
3351 |
+
{
|
3352 |
+
"epoch": 116.46017699115045,
|
3353 |
+
"grad_norm": 3.8041961193084717,
|
3354 |
+
"learning_rate": 1.1574074074074074e-06,
|
3355 |
+
"loss": 0.0336,
|
3356 |
+
"step": 3290
|
3357 |
+
},
|
3358 |
+
{
|
3359 |
+
"epoch": 116.8141592920354,
|
3360 |
+
"grad_norm": 4.338891983032227,
|
3361 |
+
"learning_rate": 9.92063492063492e-07,
|
3362 |
+
"loss": 0.0255,
|
3363 |
+
"step": 3300
|
3364 |
+
},
|
3365 |
+
{
|
3366 |
+
"epoch": 116.99115044247787,
|
3367 |
+
"eval_accuracy": 0.87,
|
3368 |
+
"eval_loss": 0.5795088410377502,
|
3369 |
+
"eval_runtime": 6.4316,
|
3370 |
+
"eval_samples_per_second": 62.193,
|
3371 |
+
"eval_steps_per_second": 2.021,
|
3372 |
+
"step": 3305
|
3373 |
+
},
|
3374 |
+
{
|
3375 |
+
"epoch": 117.16814159292035,
|
3376 |
+
"grad_norm": 0.9658322930335999,
|
3377 |
+
"learning_rate": 8.267195767195768e-07,
|
3378 |
+
"loss": 0.0206,
|
3379 |
+
"step": 3310
|
3380 |
+
},
|
3381 |
+
{
|
3382 |
+
"epoch": 117.52212389380531,
|
3383 |
+
"grad_norm": 2.4178810119628906,
|
3384 |
+
"learning_rate": 6.613756613756614e-07,
|
3385 |
+
"loss": 0.029,
|
3386 |
+
"step": 3320
|
3387 |
+
},
|
3388 |
+
{
|
3389 |
+
"epoch": 117.87610619469027,
|
3390 |
+
"grad_norm": 0.15479572117328644,
|
3391 |
+
"learning_rate": 4.96031746031746e-07,
|
3392 |
+
"loss": 0.0174,
|
3393 |
+
"step": 3330
|
3394 |
+
},
|
3395 |
+
{
|
3396 |
+
"epoch": 117.98230088495575,
|
3397 |
+
"eval_accuracy": 0.875,
|
3398 |
+
"eval_loss": 0.5803083777427673,
|
3399 |
+
"eval_runtime": 6.7884,
|
3400 |
+
"eval_samples_per_second": 58.924,
|
3401 |
+
"eval_steps_per_second": 1.915,
|
3402 |
+
"step": 3333
|
3403 |
+
},
|
3404 |
+
{
|
3405 |
+
"epoch": 118.23008849557522,
|
3406 |
+
"grad_norm": 3.2203192710876465,
|
3407 |
+
"learning_rate": 3.306878306878307e-07,
|
3408 |
+
"loss": 0.0223,
|
3409 |
+
"step": 3340
|
3410 |
+
},
|
3411 |
+
{
|
3412 |
+
"epoch": 118.58407079646018,
|
3413 |
+
"grad_norm": 1.2153400182724,
|
3414 |
+
"learning_rate": 1.6534391534391535e-07,
|
3415 |
+
"loss": 0.0236,
|
3416 |
+
"step": 3350
|
3417 |
+
},
|
3418 |
+
{
|
3419 |
+
"epoch": 118.93805309734513,
|
3420 |
+
"grad_norm": 3.442626714706421,
|
3421 |
+
"learning_rate": 0.0,
|
3422 |
+
"loss": 0.0225,
|
3423 |
+
"step": 3360
|
3424 |
+
},
|
3425 |
+
{
|
3426 |
+
"epoch": 118.93805309734513,
|
3427 |
+
"eval_accuracy": 0.875,
|
3428 |
+
"eval_loss": 0.5796830654144287,
|
3429 |
+
"eval_runtime": 6.4364,
|
3430 |
+
"eval_samples_per_second": 62.147,
|
3431 |
+
"eval_steps_per_second": 2.02,
|
3432 |
+
"step": 3360
|
3433 |
+
},
|
3434 |
+
{
|
3435 |
+
"epoch": 118.93805309734513,
|
3436 |
+
"step": 3360,
|
3437 |
+
"total_flos": 1.0779764781475824e+19,
|
3438 |
+
"train_loss": 0.10835168258996591,
|
3439 |
+
"train_runtime": 10219.331,
|
3440 |
+
"train_samples_per_second": 42.273,
|
3441 |
+
"train_steps_per_second": 0.329
|
3442 |
+
}
|
3443 |
+
],
|
3444 |
+
"logging_steps": 10,
|
3445 |
+
"max_steps": 3360,
|
3446 |
+
"num_input_tokens_seen": 0,
|
3447 |
+
"num_train_epochs": 120,
|
3448 |
+
"save_steps": 500,
|
3449 |
+
"stateful_callbacks": {
|
3450 |
+
"TrainerControl": {
|
3451 |
+
"args": {
|
3452 |
+
"should_epoch_stop": false,
|
3453 |
+
"should_evaluate": false,
|
3454 |
+
"should_log": false,
|
3455 |
+
"should_save": false,
|
3456 |
+
"should_training_stop": false
|
3457 |
+
},
|
3458 |
+
"attributes": {}
|
3459 |
+
}
|
3460 |
+
},
|
3461 |
+
"total_flos": 1.0779764781475824e+19,
|
3462 |
+
"train_batch_size": 32,
|
3463 |
+
"trial_name": null,
|
3464 |
+
"trial_params": null
|
3465 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:76963b8a9ccad918f481032b34fe88f03ead6ac29edf721aa097e20abeffa2e4
|
3 |
+
size 5240
|