feat: add tnn
Browse files- README.md +28 -39
- all_results.json +10 -10
- config.json +10 -6
- eval_results.json +5 -5
- pytorch_model.bin +2 -2
- train_results.json +5 -5
- trainer_state.json +896 -1613
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -23,7 +23,7 @@ model-index:
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
-
value: 0.
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
-
- Accuracy: 0.
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
@@ -54,7 +54,7 @@ More information needed
|
|
| 54 |
|
| 55 |
The following hyperparameters were used during training:
|
| 56 |
- learning_rate: 0.0002
|
| 57 |
-
- train_batch_size:
|
| 58 |
- eval_batch_size: 8
|
| 59 |
- seed: 42
|
| 60 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
@@ -65,41 +65,30 @@ The following hyperparameters were used during training:
|
|
| 65 |
|
| 66 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 67 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
-
| 0.
|
| 78 |
-
| 0.
|
| 79 |
-
| 0.
|
| 80 |
-
| 0.
|
| 81 |
-
| 0.
|
| 82 |
-
| 0.
|
| 83 |
-
| 0.
|
| 84 |
-
| 0.
|
| 85 |
-
| 0.
|
| 86 |
-
| 0.
|
| 87 |
-
| 0.
|
| 88 |
-
| 0.
|
| 89 |
-
| 0.
|
| 90 |
-
| 0.
|
| 91 |
-
| 0.
|
| 92 |
-
| 0.1669 | 2.86 | 2500 | 0.3132 | 0.8854 |
|
| 93 |
-
| 0.1461 | 2.97 | 2600 | 0.3039 | 0.8883 |
|
| 94 |
-
| 0.07 | 3.09 | 2700 | 0.3402 | 0.8921 |
|
| 95 |
-
| 0.0637 | 3.2 | 2800 | 0.3446 | 0.8944 |
|
| 96 |
-
| 0.0807 | 3.31 | 2900 | 0.3425 | 0.8947 |
|
| 97 |
-
| 0.0637 | 3.43 | 3000 | 0.3396 | 0.8964 |
|
| 98 |
-
| 0.0535 | 3.54 | 3100 | 0.3407 | 0.8971 |
|
| 99 |
-
| 0.064 | 3.66 | 3200 | 0.3420 | 0.9002 |
|
| 100 |
-
| 0.0707 | 3.77 | 3300 | 0.3314 | 0.8995 |
|
| 101 |
-
| 0.058 | 3.89 | 3400 | 0.3286 | 0.9002 |
|
| 102 |
-
| 0.048 | 4.0 | 3500 | 0.3263 | 0.9013 |
|
| 103 |
|
| 104 |
|
| 105 |
### Framework versions
|
|
|
|
| 23 |
metrics:
|
| 24 |
- name: Accuracy
|
| 25 |
type: accuracy
|
| 26 |
+
value: 0.8962788114412663
|
| 27 |
---
|
| 28 |
|
| 29 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.2805
|
| 37 |
+
- Accuracy: 0.8963
|
| 38 |
|
| 39 |
## Model description
|
| 40 |
|
|
|
|
| 54 |
|
| 55 |
The following hyperparameters were used during training:
|
| 56 |
- learning_rate: 0.0002
|
| 57 |
+
- train_batch_size: 256
|
| 58 |
- eval_batch_size: 8
|
| 59 |
- seed: 42
|
| 60 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
|
|
|
| 65 |
|
| 66 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 67 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 68 |
+
| 0.7945 | 0.16 | 100 | 0.7726 | 0.6916 |
|
| 69 |
+
| 0.6502 | 0.32 | 200 | 0.6581 | 0.7327 |
|
| 70 |
+
| 0.5671 | 0.49 | 300 | 0.5892 | 0.7589 |
|
| 71 |
+
| 0.5625 | 0.65 | 400 | 0.5424 | 0.7754 |
|
| 72 |
+
| 0.5115 | 0.81 | 500 | 0.4990 | 0.7931 |
|
| 73 |
+
| 0.4643 | 0.97 | 600 | 0.4710 | 0.8040 |
|
| 74 |
+
| 0.3586 | 1.13 | 700 | 0.4337 | 0.8221 |
|
| 75 |
+
| 0.3421 | 1.29 | 800 | 0.4097 | 0.8337 |
|
| 76 |
+
| 0.3478 | 1.46 | 900 | 0.3817 | 0.8446 |
|
| 77 |
+
| 0.2965 | 1.62 | 1000 | 0.3754 | 0.8457 |
|
| 78 |
+
| 0.2986 | 1.78 | 1100 | 0.3548 | 0.8550 |
|
| 79 |
+
| 0.2932 | 1.94 | 1200 | 0.3387 | 0.8633 |
|
| 80 |
+
| 0.1701 | 2.1 | 1300 | 0.3415 | 0.8677 |
|
| 81 |
+
| 0.1891 | 2.27 | 1400 | 0.3260 | 0.8766 |
|
| 82 |
+
| 0.1741 | 2.43 | 1500 | 0.3103 | 0.8818 |
|
| 83 |
+
| 0.1542 | 2.59 | 1600 | 0.3061 | 0.8869 |
|
| 84 |
+
| 0.172 | 2.75 | 1700 | 0.2925 | 0.8888 |
|
| 85 |
+
| 0.1575 | 2.91 | 1800 | 0.2805 | 0.8963 |
|
| 86 |
+
| 0.0698 | 3.07 | 1900 | 0.3031 | 0.8984 |
|
| 87 |
+
| 0.0671 | 3.24 | 2000 | 0.3075 | 0.9009 |
|
| 88 |
+
| 0.0576 | 3.4 | 2100 | 0.3051 | 0.9029 |
|
| 89 |
+
| 0.0519 | 3.56 | 2200 | 0.2982 | 0.9066 |
|
| 90 |
+
| 0.0527 | 3.72 | 2300 | 0.2974 | 0.9072 |
|
| 91 |
+
| 0.0561 | 3.88 | 2400 | 0.2912 | 0.9091 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 92 |
|
| 93 |
|
| 94 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 16.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.8962788114412663,
|
| 4 |
+
"eval_loss": 0.2805336117744446,
|
| 5 |
+
"eval_runtime": 164.7032,
|
| 6 |
+
"eval_samples_per_second": 131.181,
|
| 7 |
+
"eval_steps_per_second": 16.399,
|
| 8 |
+
"total_flos": 4.902492055502632e+19,
|
| 9 |
+
"train_loss": 0.2963483939957006,
|
| 10 |
+
"train_runtime": 9311.4957,
|
| 11 |
+
"train_samples_per_second": 67.941,
|
| 12 |
+
"train_steps_per_second": 0.265
|
| 13 |
}
|
config.json
CHANGED
|
@@ -9,17 +9,21 @@
|
|
| 9 |
"hidden_dropout_prob": 0.0,
|
| 10 |
"hidden_size": 768,
|
| 11 |
"id2label": {
|
| 12 |
-
"0": "
|
| 13 |
-
"1": "
|
| 14 |
-
"2": "up
|
|
|
|
|
|
|
| 15 |
},
|
| 16 |
"image_size": 224,
|
| 17 |
"initializer_range": 0.02,
|
| 18 |
"intermediate_size": 3072,
|
| 19 |
"label2id": {
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"up
|
|
|
|
|
|
|
| 23 |
},
|
| 24 |
"layer_norm_eps": 1e-12,
|
| 25 |
"model_type": "vit",
|
|
|
|
| 9 |
"hidden_dropout_prob": 0.0,
|
| 10 |
"hidden_size": 768,
|
| 11 |
"id2label": {
|
| 12 |
+
"0": "left",
|
| 13 |
+
"1": "right",
|
| 14 |
+
"2": "up",
|
| 15 |
+
"3": "up-left",
|
| 16 |
+
"4": "up-right"
|
| 17 |
},
|
| 18 |
"image_size": 224,
|
| 19 |
"initializer_range": 0.02,
|
| 20 |
"intermediate_size": 3072,
|
| 21 |
"label2id": {
|
| 22 |
+
"left": "0",
|
| 23 |
+
"right": "1",
|
| 24 |
+
"up": "2",
|
| 25 |
+
"up-left": "3",
|
| 26 |
+
"up-right": "4"
|
| 27 |
},
|
| 28 |
"layer_norm_eps": 1e-12,
|
| 29 |
"model_type": "vit",
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 16.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.8962788114412663,
|
| 4 |
+
"eval_loss": 0.2805336117744446,
|
| 5 |
+
"eval_runtime": 164.7032,
|
| 6 |
+
"eval_samples_per_second": 131.181,
|
| 7 |
+
"eval_steps_per_second": 16.399
|
| 8 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5dc773afc8ca6694e1f7ef11e739981e312e5348256fdc8ab652e1a9b7b44f8
|
| 3 |
+
size 343277933
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 4.0,
|
| 3 |
+
"total_flos": 4.902492055502632e+19,
|
| 4 |
+
"train_loss": 0.2963483939957006,
|
| 5 |
+
"train_runtime": 9311.4957,
|
| 6 |
+
"train_samples_per_second": 67.941,
|
| 7 |
+
"train_steps_per_second": 0.265
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,2443 +1,1726 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "./vit-base-tm/checkpoint-
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 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.
|
| 13 |
-
"learning_rate": 0.
|
| 14 |
-
"loss": 1.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 0.
|
| 20 |
-
"loss": 1.
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 0.
|
| 26 |
-
"loss": 1.
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch": 0.
|
| 31 |
-
"learning_rate": 0.
|
| 32 |
-
"loss": 0.
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch": 0.
|
| 37 |
-
"learning_rate": 0.
|
| 38 |
-
"loss": 0.
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch": 0.
|
| 43 |
-
"learning_rate": 0.
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 0.
|
| 49 |
-
"learning_rate": 0.
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"learning_rate": 0.
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch": 0.
|
| 61 |
-
"learning_rate": 0.
|
| 62 |
-
"loss": 0.
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"epoch": 0.
|
| 67 |
-
"learning_rate": 0.
|
| 68 |
-
"loss": 0.
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
-
"epoch": 0.
|
| 73 |
-
"eval_accuracy": 0.
|
| 74 |
-
"eval_loss": 0.
|
| 75 |
-
"eval_runtime":
|
| 76 |
-
"eval_samples_per_second":
|
| 77 |
-
"eval_steps_per_second":
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
-
"epoch": 0.
|
| 82 |
-
"learning_rate": 0.
|
| 83 |
-
"loss": 0.
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
-
"epoch": 0.
|
| 88 |
-
"learning_rate": 0.
|
| 89 |
-
"loss": 0.
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"epoch": 0.
|
| 94 |
-
"learning_rate": 0.
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
-
"epoch": 0.
|
| 100 |
-
"learning_rate": 0.
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch": 0.
|
| 106 |
-
"learning_rate": 0.
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
-
"epoch": 0.
|
| 112 |
-
"learning_rate": 0.
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"learning_rate": 0.
|
| 119 |
-
"loss": 0.
|
| 120 |
"step": 170
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"epoch": 0.
|
| 124 |
-
"learning_rate": 0.
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 180
|
| 127 |
},
|
| 128 |
{
|
| 129 |
-
"epoch": 0.
|
| 130 |
-
"learning_rate": 0.
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch": 0.
|
| 136 |
-
"learning_rate": 0.
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch": 0.
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 0.
|
| 144 |
-
"eval_runtime":
|
| 145 |
-
"eval_samples_per_second":
|
| 146 |
-
"eval_steps_per_second":
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch": 0.
|
| 151 |
-
"learning_rate": 0.
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch": 0.
|
| 157 |
-
"learning_rate": 0.
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"epoch": 0.
|
| 163 |
-
"learning_rate": 0.
|
| 164 |
-
"loss": 0.
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"epoch": 0.
|
| 169 |
-
"learning_rate": 0.
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 0.
|
| 175 |
-
"learning_rate": 0.
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 250
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch": 0.
|
| 181 |
-
"learning_rate": 0.
|
| 182 |
-
"loss": 0.
|
| 183 |
"step": 260
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch": 0.
|
| 187 |
-
"learning_rate": 0.
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 270
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch": 0.
|
| 193 |
-
"learning_rate": 0.
|
| 194 |
-
"loss": 0.
|
| 195 |
"step": 280
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch": 0.
|
| 199 |
-
"learning_rate": 0.
|
| 200 |
-
"loss": 0.
|
| 201 |
"step": 290
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch": 0.
|
| 205 |
-
"learning_rate": 0.
|
| 206 |
-
"loss": 0.
|
| 207 |
"step": 300
|
| 208 |
},
|
| 209 |
{
|
| 210 |
-
"epoch": 0.
|
| 211 |
-
"eval_accuracy": 0.
|
| 212 |
-
"eval_loss": 0.
|
| 213 |
-
"eval_runtime":
|
| 214 |
-
"eval_samples_per_second":
|
| 215 |
-
"eval_steps_per_second":
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
-
"epoch": 0.
|
| 220 |
-
"learning_rate": 0.
|
| 221 |
-
"loss": 0.
|
| 222 |
"step": 310
|
| 223 |
},
|
| 224 |
{
|
| 225 |
-
"epoch": 0.
|
| 226 |
-
"learning_rate": 0.
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 320
|
| 229 |
},
|
| 230 |
{
|
| 231 |
-
"epoch": 0.
|
| 232 |
-
"learning_rate": 0.
|
| 233 |
-
"loss": 0.
|
| 234 |
"step": 330
|
| 235 |
},
|
| 236 |
{
|
| 237 |
-
"epoch": 0.
|
| 238 |
-
"learning_rate": 0.
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 340
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch": 0.
|
| 244 |
-
"learning_rate": 0.
|
| 245 |
-
"loss": 0.
|
| 246 |
"step": 350
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch": 0.
|
| 250 |
-
"learning_rate": 0.
|
| 251 |
-
"loss": 0.
|
| 252 |
"step": 360
|
| 253 |
},
|
| 254 |
{
|
| 255 |
-
"epoch": 0.
|
| 256 |
-
"learning_rate": 0.
|
| 257 |
-
"loss": 0.
|
| 258 |
"step": 370
|
| 259 |
},
|
| 260 |
{
|
| 261 |
-
"epoch": 0.
|
| 262 |
-
"learning_rate": 0.
|
| 263 |
-
"loss": 0.
|
| 264 |
"step": 380
|
| 265 |
},
|
| 266 |
{
|
| 267 |
-
"epoch": 0.
|
| 268 |
-
"learning_rate": 0.
|
| 269 |
-
"loss": 0.
|
| 270 |
"step": 390
|
| 271 |
},
|
| 272 |
{
|
| 273 |
-
"epoch": 0.
|
| 274 |
-
"learning_rate": 0.
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 400
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch": 0.
|
| 280 |
-
"eval_accuracy": 0.
|
| 281 |
-
"eval_loss": 0.
|
| 282 |
-
"eval_runtime":
|
| 283 |
-
"eval_samples_per_second":
|
| 284 |
-
"eval_steps_per_second":
|
| 285 |
"step": 400
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch": 0.
|
| 289 |
-
"learning_rate": 0.
|
| 290 |
-
"loss": 0.
|
| 291 |
"step": 410
|
| 292 |
},
|
| 293 |
{
|
| 294 |
-
"epoch": 0.
|
| 295 |
-
"learning_rate": 0.
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 420
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch": 0.
|
| 301 |
-
"learning_rate": 0.
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 430
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"epoch": 0.
|
| 307 |
-
"learning_rate": 0.
|
| 308 |
-
"loss": 0.
|
| 309 |
"step": 440
|
| 310 |
},
|
| 311 |
{
|
| 312 |
-
"epoch": 0.
|
| 313 |
-
"learning_rate": 0.
|
| 314 |
-
"loss": 0.
|
| 315 |
"step": 450
|
| 316 |
},
|
| 317 |
{
|
| 318 |
-
"epoch": 0.
|
| 319 |
-
"learning_rate": 0.
|
| 320 |
-
"loss": 0.
|
| 321 |
"step": 460
|
| 322 |
},
|
| 323 |
{
|
| 324 |
-
"epoch": 0.
|
| 325 |
-
"learning_rate": 0.
|
| 326 |
-
"loss": 0.
|
| 327 |
"step": 470
|
| 328 |
},
|
| 329 |
{
|
| 330 |
-
"epoch": 0.
|
| 331 |
-
"learning_rate": 0.
|
| 332 |
-
"loss": 0.
|
| 333 |
"step": 480
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"epoch": 0.
|
| 337 |
-
"learning_rate": 0.
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 490
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch": 0.
|
| 343 |
-
"learning_rate": 0.
|
| 344 |
-
"loss": 0.
|
| 345 |
"step": 500
|
| 346 |
},
|
| 347 |
{
|
| 348 |
-
"epoch": 0.
|
| 349 |
-
"eval_accuracy": 0.
|
| 350 |
-
"eval_loss": 0.
|
| 351 |
-
"eval_runtime":
|
| 352 |
-
"eval_samples_per_second":
|
| 353 |
-
"eval_steps_per_second": 18.
|
| 354 |
"step": 500
|
| 355 |
},
|
| 356 |
{
|
| 357 |
-
"epoch": 0.
|
| 358 |
-
"learning_rate": 0.
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 510
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch": 0.
|
| 364 |
-
"learning_rate": 0.
|
| 365 |
-
"loss": 0.
|
| 366 |
"step": 520
|
| 367 |
},
|
| 368 |
{
|
| 369 |
-
"epoch": 0.
|
| 370 |
-
"learning_rate": 0.
|
| 371 |
-
"loss": 0.
|
| 372 |
"step": 530
|
| 373 |
},
|
| 374 |
{
|
| 375 |
-
"epoch": 0.
|
| 376 |
-
"learning_rate": 0.
|
| 377 |
-
"loss": 0.
|
| 378 |
"step": 540
|
| 379 |
},
|
| 380 |
{
|
| 381 |
-
"epoch": 0.
|
| 382 |
-
"learning_rate": 0.
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 550
|
| 385 |
},
|
| 386 |
{
|
| 387 |
-
"epoch": 0.
|
| 388 |
-
"learning_rate": 0.
|
| 389 |
-
"loss": 0.
|
| 390 |
"step": 560
|
| 391 |
},
|
| 392 |
{
|
| 393 |
-
"epoch": 0.
|
| 394 |
-
"learning_rate": 0.
|
| 395 |
-
"loss": 0.
|
| 396 |
"step": 570
|
| 397 |
},
|
| 398 |
{
|
| 399 |
-
"epoch": 0.
|
| 400 |
-
"learning_rate": 0.
|
| 401 |
-
"loss": 0.
|
| 402 |
"step": 580
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch": 0.
|
| 406 |
-
"learning_rate": 0.
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 590
|
| 409 |
},
|
| 410 |
{
|
| 411 |
-
"epoch": 0.
|
| 412 |
-
"learning_rate": 0.
|
| 413 |
-
"loss": 0.
|
| 414 |
"step": 600
|
| 415 |
},
|
| 416 |
{
|
| 417 |
-
"epoch": 0.
|
| 418 |
-
"eval_accuracy": 0.
|
| 419 |
-
"eval_loss": 0.
|
| 420 |
-
"eval_runtime":
|
| 421 |
-
"eval_samples_per_second":
|
| 422 |
-
"eval_steps_per_second": 18.
|
| 423 |
"step": 600
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch": 0.
|
| 427 |
-
"learning_rate": 0.
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 610
|
| 430 |
},
|
| 431 |
{
|
| 432 |
-
"epoch":
|
| 433 |
-
"learning_rate": 0.
|
| 434 |
-
"loss": 0.
|
| 435 |
"step": 620
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch":
|
| 439 |
-
"learning_rate": 0.
|
| 440 |
-
"loss": 0.
|
| 441 |
"step": 630
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch":
|
| 445 |
-
"learning_rate": 0.
|
| 446 |
-
"loss": 0.
|
| 447 |
"step": 640
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch":
|
| 451 |
-
"learning_rate": 0.
|
| 452 |
-
"loss": 0.
|
| 453 |
"step": 650
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch":
|
| 457 |
-
"learning_rate": 0.
|
| 458 |
-
"loss": 0.
|
| 459 |
"step": 660
|
| 460 |
},
|
| 461 |
{
|
| 462 |
-
"epoch":
|
| 463 |
-
"learning_rate": 0.
|
| 464 |
-
"loss": 0.
|
| 465 |
"step": 670
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch":
|
| 469 |
-
"learning_rate": 0.
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 680
|
| 472 |
},
|
| 473 |
{
|
| 474 |
-
"epoch":
|
| 475 |
-
"learning_rate": 0.
|
| 476 |
-
"loss": 0.
|
| 477 |
"step": 690
|
| 478 |
},
|
| 479 |
{
|
| 480 |
-
"epoch":
|
| 481 |
-
"learning_rate": 0.
|
| 482 |
-
"loss": 0.
|
| 483 |
"step": 700
|
| 484 |
},
|
| 485 |
{
|
| 486 |
-
"epoch":
|
| 487 |
-
"eval_accuracy": 0.
|
| 488 |
-
"eval_loss": 0.
|
| 489 |
-
"eval_runtime":
|
| 490 |
-
"eval_samples_per_second":
|
| 491 |
-
"eval_steps_per_second":
|
| 492 |
"step": 700
|
| 493 |
},
|
| 494 |
{
|
| 495 |
-
"epoch":
|
| 496 |
-
"learning_rate": 0.
|
| 497 |
-
"loss": 0.
|
| 498 |
"step": 710
|
| 499 |
},
|
| 500 |
{
|
| 501 |
-
"epoch":
|
| 502 |
-
"learning_rate": 0.
|
| 503 |
-
"loss": 0.
|
| 504 |
"step": 720
|
| 505 |
},
|
| 506 |
{
|
| 507 |
-
"epoch":
|
| 508 |
-
"learning_rate": 0.
|
| 509 |
-
"loss": 0.
|
| 510 |
"step": 730
|
| 511 |
},
|
| 512 |
{
|
| 513 |
-
"epoch":
|
| 514 |
-
"learning_rate": 0.
|
| 515 |
-
"loss": 0.
|
| 516 |
"step": 740
|
| 517 |
},
|
| 518 |
{
|
| 519 |
-
"epoch":
|
| 520 |
-
"learning_rate": 0.
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 750
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"epoch":
|
| 526 |
-
"learning_rate": 0.
|
| 527 |
-
"loss": 0.
|
| 528 |
"step": 760
|
| 529 |
},
|
| 530 |
{
|
| 531 |
-
"epoch":
|
| 532 |
-
"learning_rate": 0.
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 770
|
| 535 |
},
|
| 536 |
{
|
| 537 |
-
"epoch":
|
| 538 |
-
"learning_rate": 0.
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 780
|
| 541 |
},
|
| 542 |
{
|
| 543 |
-
"epoch":
|
| 544 |
-
"learning_rate": 0.
|
| 545 |
-
"loss": 0.
|
| 546 |
"step": 790
|
| 547 |
},
|
| 548 |
{
|
| 549 |
-
"epoch":
|
| 550 |
-
"learning_rate": 0.
|
| 551 |
-
"loss": 0.
|
| 552 |
"step": 800
|
| 553 |
},
|
| 554 |
{
|
| 555 |
-
"epoch":
|
| 556 |
-
"eval_accuracy": 0.
|
| 557 |
-
"eval_loss": 0.
|
| 558 |
-
"eval_runtime":
|
| 559 |
-
"eval_samples_per_second":
|
| 560 |
-
"eval_steps_per_second":
|
| 561 |
"step": 800
|
| 562 |
},
|
| 563 |
{
|
| 564 |
-
"epoch":
|
| 565 |
-
"learning_rate": 0.
|
| 566 |
-
"loss": 0.
|
| 567 |
"step": 810
|
| 568 |
},
|
| 569 |
{
|
| 570 |
-
"epoch":
|
| 571 |
-
"learning_rate": 0.
|
| 572 |
-
"loss": 0.
|
| 573 |
"step": 820
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"epoch":
|
| 577 |
-
"learning_rate": 0.
|
| 578 |
-
"loss": 0.
|
| 579 |
"step": 830
|
| 580 |
},
|
| 581 |
{
|
| 582 |
-
"epoch":
|
| 583 |
-
"learning_rate": 0.
|
| 584 |
-
"loss": 0.
|
| 585 |
"step": 840
|
| 586 |
},
|
| 587 |
{
|
| 588 |
-
"epoch":
|
| 589 |
-
"learning_rate": 0.
|
| 590 |
-
"loss": 0.
|
| 591 |
"step": 850
|
| 592 |
},
|
| 593 |
{
|
| 594 |
-
"epoch":
|
| 595 |
-
"learning_rate": 0.
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 860
|
| 598 |
},
|
| 599 |
{
|
| 600 |
-
"epoch":
|
| 601 |
-
"learning_rate": 0.
|
| 602 |
-
"loss": 0.
|
| 603 |
"step": 870
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch": 1.
|
| 607 |
-
"learning_rate": 0.
|
| 608 |
-
"loss": 0.
|
| 609 |
"step": 880
|
| 610 |
},
|
| 611 |
{
|
| 612 |
-
"epoch": 1.
|
| 613 |
-
"learning_rate": 0.
|
| 614 |
-
"loss": 0.
|
| 615 |
"step": 890
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch": 1.
|
| 619 |
-
"learning_rate": 0.
|
| 620 |
-
"loss": 0.
|
| 621 |
"step": 900
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch": 1.
|
| 625 |
-
"eval_accuracy": 0.
|
| 626 |
-
"eval_loss": 0.
|
| 627 |
-
"eval_runtime":
|
| 628 |
-
"eval_samples_per_second":
|
| 629 |
-
"eval_steps_per_second": 17.
|
| 630 |
"step": 900
|
| 631 |
},
|
| 632 |
{
|
| 633 |
-
"epoch": 1.
|
| 634 |
-
"learning_rate": 0.
|
| 635 |
-
"loss": 0.
|
| 636 |
"step": 910
|
| 637 |
},
|
| 638 |
{
|
| 639 |
-
"epoch": 1.
|
| 640 |
-
"learning_rate": 0.
|
| 641 |
-
"loss": 0.
|
| 642 |
"step": 920
|
| 643 |
},
|
| 644 |
{
|
| 645 |
-
"epoch": 1.
|
| 646 |
-
"learning_rate": 0.
|
| 647 |
-
"loss": 0.
|
| 648 |
"step": 930
|
| 649 |
},
|
| 650 |
{
|
| 651 |
-
"epoch": 1.
|
| 652 |
-
"learning_rate": 0.
|
| 653 |
-
"loss": 0.
|
| 654 |
"step": 940
|
| 655 |
},
|
| 656 |
{
|
| 657 |
-
"epoch": 1.
|
| 658 |
-
"learning_rate": 0.
|
| 659 |
-
"loss": 0.
|
| 660 |
"step": 950
|
| 661 |
},
|
| 662 |
{
|
| 663 |
-
"epoch": 1.
|
| 664 |
-
"learning_rate": 0.
|
| 665 |
-
"loss": 0.
|
| 666 |
"step": 960
|
| 667 |
},
|
| 668 |
{
|
| 669 |
-
"epoch": 1.
|
| 670 |
-
"learning_rate": 0.
|
| 671 |
-
"loss": 0.
|
| 672 |
"step": 970
|
| 673 |
},
|
| 674 |
{
|
| 675 |
-
"epoch": 1.
|
| 676 |
-
"learning_rate": 0.
|
| 677 |
-
"loss": 0.
|
| 678 |
"step": 980
|
| 679 |
},
|
| 680 |
{
|
| 681 |
-
"epoch": 1.
|
| 682 |
-
"learning_rate": 0.
|
| 683 |
-
"loss": 0.
|
| 684 |
"step": 990
|
| 685 |
},
|
| 686 |
{
|
| 687 |
-
"epoch": 1.
|
| 688 |
-
"learning_rate": 0.
|
| 689 |
-
"loss": 0.
|
| 690 |
"step": 1000
|
| 691 |
},
|
| 692 |
{
|
| 693 |
-
"epoch": 1.
|
| 694 |
-
"eval_accuracy": 0.
|
| 695 |
-
"eval_loss": 0.
|
| 696 |
-
"eval_runtime":
|
| 697 |
-
"eval_samples_per_second":
|
| 698 |
-
"eval_steps_per_second": 17.
|
| 699 |
"step": 1000
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch": 1.
|
| 703 |
-
"learning_rate": 0.
|
| 704 |
-
"loss": 0.
|
| 705 |
"step": 1010
|
| 706 |
},
|
| 707 |
{
|
| 708 |
-
"epoch": 1.
|
| 709 |
-
"learning_rate": 0.
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 1020
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"epoch": 1.
|
| 715 |
-
"learning_rate": 0.
|
| 716 |
-
"loss": 0.
|
| 717 |
"step": 1030
|
| 718 |
},
|
| 719 |
{
|
| 720 |
-
"epoch": 1.
|
| 721 |
-
"learning_rate": 0.
|
| 722 |
-
"loss": 0.
|
| 723 |
"step": 1040
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"epoch": 1.
|
| 727 |
-
"learning_rate": 0.
|
| 728 |
-
"loss": 0.
|
| 729 |
"step": 1050
|
| 730 |
},
|
| 731 |
{
|
| 732 |
-
"epoch": 1.
|
| 733 |
-
"learning_rate": 0.
|
| 734 |
-
"loss": 0.
|
| 735 |
"step": 1060
|
| 736 |
},
|
| 737 |
{
|
| 738 |
-
"epoch": 1.
|
| 739 |
-
"learning_rate": 0.
|
| 740 |
-
"loss": 0.
|
| 741 |
"step": 1070
|
| 742 |
},
|
| 743 |
{
|
| 744 |
-
"epoch": 1.
|
| 745 |
-
"learning_rate": 0.
|
| 746 |
-
"loss": 0.
|
| 747 |
"step": 1080
|
| 748 |
},
|
| 749 |
{
|
| 750 |
-
"epoch": 1.
|
| 751 |
-
"learning_rate": 0.
|
| 752 |
-
"loss": 0.
|
| 753 |
"step": 1090
|
| 754 |
},
|
| 755 |
{
|
| 756 |
-
"epoch": 1.
|
| 757 |
-
"learning_rate": 0.
|
| 758 |
-
"loss": 0.
|
| 759 |
"step": 1100
|
| 760 |
},
|
| 761 |
{
|
| 762 |
-
"epoch": 1.
|
| 763 |
-
"eval_accuracy": 0.
|
| 764 |
-
"eval_loss": 0.
|
| 765 |
-
"eval_runtime":
|
| 766 |
-
"eval_samples_per_second":
|
| 767 |
-
"eval_steps_per_second":
|
| 768 |
"step": 1100
|
| 769 |
},
|
| 770 |
{
|
| 771 |
-
"epoch": 1.
|
| 772 |
-
"learning_rate": 0.
|
| 773 |
-
"loss": 0.
|
| 774 |
"step": 1110
|
| 775 |
},
|
| 776 |
{
|
| 777 |
-
"epoch": 1.
|
| 778 |
-
"learning_rate": 0.
|
| 779 |
-
"loss": 0.
|
| 780 |
"step": 1120
|
| 781 |
},
|
| 782 |
{
|
| 783 |
-
"epoch": 1.
|
| 784 |
-
"learning_rate": 0.
|
| 785 |
-
"loss": 0.
|
| 786 |
"step": 1130
|
| 787 |
},
|
| 788 |
{
|
| 789 |
-
"epoch": 1.
|
| 790 |
-
"learning_rate": 0.
|
| 791 |
-
"loss": 0.
|
| 792 |
"step": 1140
|
| 793 |
},
|
| 794 |
{
|
| 795 |
-
"epoch": 1.
|
| 796 |
-
"learning_rate": 0.
|
| 797 |
-
"loss": 0.
|
| 798 |
"step": 1150
|
| 799 |
},
|
| 800 |
{
|
| 801 |
-
"epoch": 1.
|
| 802 |
-
"learning_rate": 0.
|
| 803 |
-
"loss": 0.
|
| 804 |
"step": 1160
|
| 805 |
},
|
| 806 |
{
|
| 807 |
-
"epoch": 1.
|
| 808 |
-
"learning_rate": 0.
|
| 809 |
-
"loss": 0.
|
| 810 |
"step": 1170
|
| 811 |
},
|
| 812 |
{
|
| 813 |
-
"epoch": 1.
|
| 814 |
-
"learning_rate": 0.
|
| 815 |
-
"loss": 0.
|
| 816 |
"step": 1180
|
| 817 |
},
|
| 818 |
{
|
| 819 |
-
"epoch": 1.
|
| 820 |
-
"learning_rate": 0.
|
| 821 |
-
"loss": 0.
|
| 822 |
"step": 1190
|
| 823 |
},
|
| 824 |
{
|
| 825 |
-
"epoch": 1.
|
| 826 |
-
"learning_rate": 0.
|
| 827 |
-
"loss": 0.
|
| 828 |
"step": 1200
|
| 829 |
},
|
| 830 |
{
|
| 831 |
-
"epoch": 1.
|
| 832 |
-
"eval_accuracy": 0.
|
| 833 |
-
"eval_loss": 0.
|
| 834 |
-
"eval_runtime":
|
| 835 |
-
"eval_samples_per_second":
|
| 836 |
-
"eval_steps_per_second":
|
| 837 |
"step": 1200
|
| 838 |
},
|
| 839 |
{
|
| 840 |
-
"epoch": 1.
|
| 841 |
-
"learning_rate": 0.
|
| 842 |
-
"loss": 0.
|
| 843 |
"step": 1210
|
| 844 |
},
|
| 845 |
{
|
| 846 |
-
"epoch": 1.
|
| 847 |
-
"learning_rate": 0.
|
| 848 |
-
"loss": 0.
|
| 849 |
"step": 1220
|
| 850 |
},
|
| 851 |
{
|
| 852 |
-
"epoch": 1.
|
| 853 |
-
"learning_rate": 0.
|
| 854 |
-
"loss": 0.
|
| 855 |
"step": 1230
|
| 856 |
},
|
| 857 |
{
|
| 858 |
-
"epoch":
|
| 859 |
-
"learning_rate":
|
| 860 |
-
"loss": 0.
|
| 861 |
"step": 1240
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch":
|
| 865 |
-
"learning_rate":
|
| 866 |
-
"loss": 0.
|
| 867 |
"step": 1250
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"learning_rate":
|
| 872 |
-
"loss": 0.
|
| 873 |
"step": 1260
|
| 874 |
},
|
| 875 |
{
|
| 876 |
-
"epoch":
|
| 877 |
-
"learning_rate":
|
| 878 |
-
"loss": 0.
|
| 879 |
"step": 1270
|
| 880 |
},
|
| 881 |
{
|
| 882 |
-
"epoch":
|
| 883 |
-
"learning_rate":
|
| 884 |
-
"loss": 0.
|
| 885 |
"step": 1280
|
| 886 |
},
|
| 887 |
{
|
| 888 |
-
"epoch":
|
| 889 |
-
"learning_rate":
|
| 890 |
-
"loss": 0.
|
| 891 |
"step": 1290
|
| 892 |
},
|
| 893 |
{
|
| 894 |
-
"epoch":
|
| 895 |
-
"learning_rate":
|
| 896 |
-
"loss": 0.
|
| 897 |
"step": 1300
|
| 898 |
},
|
| 899 |
{
|
| 900 |
-
"epoch":
|
| 901 |
-
"eval_accuracy": 0.
|
| 902 |
-
"eval_loss": 0.
|
| 903 |
-
"eval_runtime":
|
| 904 |
-
"eval_samples_per_second":
|
| 905 |
-
"eval_steps_per_second":
|
| 906 |
"step": 1300
|
| 907 |
},
|
| 908 |
{
|
| 909 |
-
"epoch":
|
| 910 |
-
"learning_rate":
|
| 911 |
-
"loss": 0.
|
| 912 |
"step": 1310
|
| 913 |
},
|
| 914 |
{
|
| 915 |
-
"epoch":
|
| 916 |
-
"learning_rate":
|
| 917 |
-
"loss": 0.
|
| 918 |
"step": 1320
|
| 919 |
},
|
| 920 |
{
|
| 921 |
-
"epoch":
|
| 922 |
-
"learning_rate":
|
| 923 |
-
"loss": 0.
|
| 924 |
"step": 1330
|
| 925 |
},
|
| 926 |
{
|
| 927 |
-
"epoch":
|
| 928 |
-
"learning_rate":
|
| 929 |
-
"loss": 0.
|
| 930 |
"step": 1340
|
| 931 |
},
|
| 932 |
{
|
| 933 |
-
"epoch":
|
| 934 |
-
"learning_rate":
|
| 935 |
-
"loss": 0.
|
| 936 |
"step": 1350
|
| 937 |
},
|
| 938 |
{
|
| 939 |
-
"epoch":
|
| 940 |
-
"learning_rate":
|
| 941 |
-
"loss": 0.
|
| 942 |
"step": 1360
|
| 943 |
},
|
| 944 |
{
|
| 945 |
-
"epoch":
|
| 946 |
-
"learning_rate":
|
| 947 |
-
"loss": 0.
|
| 948 |
"step": 1370
|
| 949 |
},
|
| 950 |
{
|
| 951 |
-
"epoch":
|
| 952 |
-
"learning_rate":
|
| 953 |
-
"loss": 0.
|
| 954 |
"step": 1380
|
| 955 |
},
|
| 956 |
{
|
| 957 |
-
"epoch":
|
| 958 |
-
"learning_rate":
|
| 959 |
-
"loss": 0.
|
| 960 |
"step": 1390
|
| 961 |
},
|
| 962 |
{
|
| 963 |
-
"epoch":
|
| 964 |
-
"learning_rate":
|
| 965 |
-
"loss": 0.
|
| 966 |
"step": 1400
|
| 967 |
},
|
| 968 |
{
|
| 969 |
-
"epoch":
|
| 970 |
-
"eval_accuracy": 0.
|
| 971 |
-
"eval_loss": 0.
|
| 972 |
-
"eval_runtime":
|
| 973 |
-
"eval_samples_per_second":
|
| 974 |
-
"eval_steps_per_second":
|
| 975 |
"step": 1400
|
| 976 |
},
|
| 977 |
{
|
| 978 |
-
"epoch":
|
| 979 |
-
"learning_rate":
|
| 980 |
-
"loss": 0.
|
| 981 |
"step": 1410
|
| 982 |
},
|
| 983 |
{
|
| 984 |
-
"epoch":
|
| 985 |
-
"learning_rate":
|
| 986 |
-
"loss": 0.
|
| 987 |
"step": 1420
|
| 988 |
},
|
| 989 |
{
|
| 990 |
-
"epoch":
|
| 991 |
-
"learning_rate":
|
| 992 |
-
"loss": 0.
|
| 993 |
"step": 1430
|
| 994 |
},
|
| 995 |
{
|
| 996 |
-
"epoch":
|
| 997 |
-
"learning_rate":
|
| 998 |
-
"loss": 0.
|
| 999 |
"step": 1440
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
-
"epoch":
|
| 1003 |
-
"learning_rate":
|
| 1004 |
-
"loss": 0.
|
| 1005 |
"step": 1450
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
-
"epoch":
|
| 1009 |
-
"learning_rate":
|
| 1010 |
-
"loss": 0.
|
| 1011 |
"step": 1460
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
-
"epoch":
|
| 1015 |
-
"learning_rate":
|
| 1016 |
-
"loss": 0.
|
| 1017 |
"step": 1470
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
-
"epoch":
|
| 1021 |
-
"learning_rate":
|
| 1022 |
-
"loss": 0.
|
| 1023 |
"step": 1480
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch":
|
| 1027 |
-
"learning_rate":
|
| 1028 |
-
"loss": 0.
|
| 1029 |
"step": 1490
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
-
"epoch":
|
| 1033 |
-
"learning_rate":
|
| 1034 |
-
"loss": 0.
|
| 1035 |
"step": 1500
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
-
"epoch":
|
| 1039 |
-
"eval_accuracy": 0.
|
| 1040 |
-
"eval_loss": 0.
|
| 1041 |
-
"eval_runtime":
|
| 1042 |
-
"eval_samples_per_second":
|
| 1043 |
-
"eval_steps_per_second":
|
| 1044 |
"step": 1500
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
-
"epoch":
|
| 1048 |
-
"learning_rate":
|
| 1049 |
-
"loss": 0.
|
| 1050 |
"step": 1510
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
-
"epoch":
|
| 1054 |
-
"learning_rate":
|
| 1055 |
-
"loss": 0.
|
| 1056 |
"step": 1520
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
-
"epoch":
|
| 1060 |
-
"learning_rate":
|
| 1061 |
-
"loss": 0.
|
| 1062 |
"step": 1530
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
-
"epoch":
|
| 1066 |
-
"learning_rate":
|
| 1067 |
-
"loss": 0.
|
| 1068 |
"step": 1540
|
| 1069 |
},
|
| 1070 |
{
|
| 1071 |
-
"epoch":
|
| 1072 |
-
"learning_rate":
|
| 1073 |
-
"loss": 0.
|
| 1074 |
"step": 1550
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
-
"epoch":
|
| 1078 |
-
"learning_rate":
|
| 1079 |
-
"loss": 0.
|
| 1080 |
"step": 1560
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
-
"epoch":
|
| 1084 |
-
"learning_rate":
|
| 1085 |
-
"loss": 0.
|
| 1086 |
"step": 1570
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
-
"epoch":
|
| 1090 |
-
"learning_rate":
|
| 1091 |
-
"loss": 0.
|
| 1092 |
"step": 1580
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
-
"epoch":
|
| 1096 |
-
"learning_rate":
|
| 1097 |
-
"loss": 0.
|
| 1098 |
"step": 1590
|
| 1099 |
},
|
| 1100 |
{
|
| 1101 |
-
"epoch":
|
| 1102 |
-
"learning_rate":
|
| 1103 |
-
"loss": 0.
|
| 1104 |
"step": 1600
|
| 1105 |
},
|
| 1106 |
{
|
| 1107 |
-
"epoch":
|
| 1108 |
-
"eval_accuracy": 0.
|
| 1109 |
-
"eval_loss": 0.
|
| 1110 |
-
"eval_runtime":
|
| 1111 |
-
"eval_samples_per_second":
|
| 1112 |
-
"eval_steps_per_second":
|
| 1113 |
"step": 1600
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
-
"epoch":
|
| 1117 |
-
"learning_rate":
|
| 1118 |
-
"loss": 0.
|
| 1119 |
"step": 1610
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
-
"epoch":
|
| 1123 |
-
"learning_rate":
|
| 1124 |
-
"loss": 0.
|
| 1125 |
"step": 1620
|
| 1126 |
},
|
| 1127 |
{
|
| 1128 |
-
"epoch":
|
| 1129 |
-
"learning_rate":
|
| 1130 |
-
"loss": 0.
|
| 1131 |
"step": 1630
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
-
"epoch":
|
| 1135 |
-
"learning_rate":
|
| 1136 |
-
"loss": 0.
|
| 1137 |
"step": 1640
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
-
"epoch":
|
| 1141 |
-
"learning_rate":
|
| 1142 |
-
"loss": 0.
|
| 1143 |
"step": 1650
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
-
"epoch":
|
| 1147 |
-
"learning_rate":
|
| 1148 |
-
"loss": 0.
|
| 1149 |
"step": 1660
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
-
"epoch":
|
| 1153 |
-
"learning_rate":
|
| 1154 |
-
"loss": 0.
|
| 1155 |
"step": 1670
|
| 1156 |
},
|
| 1157 |
{
|
| 1158 |
-
"epoch":
|
| 1159 |
-
"learning_rate":
|
| 1160 |
-
"loss": 0.
|
| 1161 |
"step": 1680
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
-
"epoch":
|
| 1165 |
-
"learning_rate":
|
| 1166 |
-
"loss": 0.
|
| 1167 |
"step": 1690
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
-
"epoch":
|
| 1171 |
-
"learning_rate":
|
| 1172 |
-
"loss": 0.
|
| 1173 |
"step": 1700
|
| 1174 |
},
|
| 1175 |
{
|
| 1176 |
-
"epoch":
|
| 1177 |
-
"eval_accuracy": 0.
|
| 1178 |
-
"eval_loss": 0.
|
| 1179 |
-
"eval_runtime":
|
| 1180 |
-
"eval_samples_per_second":
|
| 1181 |
-
"eval_steps_per_second":
|
| 1182 |
"step": 1700
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
-
"epoch":
|
| 1186 |
-
"learning_rate":
|
| 1187 |
-
"loss": 0.
|
| 1188 |
"step": 1710
|
| 1189 |
},
|
| 1190 |
{
|
| 1191 |
-
"epoch":
|
| 1192 |
-
"learning_rate":
|
| 1193 |
-
"loss": 0.
|
| 1194 |
"step": 1720
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
-
"epoch":
|
| 1198 |
-
"learning_rate":
|
| 1199 |
-
"loss": 0.
|
| 1200 |
"step": 1730
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
-
"epoch":
|
| 1204 |
-
"learning_rate":
|
| 1205 |
-
"loss": 0.
|
| 1206 |
"step": 1740
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
-
"epoch": 2.
|
| 1210 |
-
"learning_rate":
|
| 1211 |
-
"loss": 0.
|
| 1212 |
"step": 1750
|
| 1213 |
},
|
| 1214 |
{
|
| 1215 |
-
"epoch": 2.
|
| 1216 |
-
"learning_rate":
|
| 1217 |
-
"loss": 0.
|
| 1218 |
"step": 1760
|
| 1219 |
},
|
| 1220 |
{
|
| 1221 |
-
"epoch": 2.
|
| 1222 |
-
"learning_rate":
|
| 1223 |
-
"loss": 0.
|
| 1224 |
"step": 1770
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
-
"epoch": 2.
|
| 1228 |
-
"learning_rate":
|
| 1229 |
-
"loss": 0.
|
| 1230 |
"step": 1780
|
| 1231 |
},
|
| 1232 |
{
|
| 1233 |
-
"epoch": 2.
|
| 1234 |
-
"learning_rate":
|
| 1235 |
-
"loss": 0.
|
| 1236 |
"step": 1790
|
| 1237 |
},
|
| 1238 |
{
|
| 1239 |
-
"epoch": 2.
|
| 1240 |
-
"learning_rate":
|
| 1241 |
-
"loss": 0.
|
| 1242 |
"step": 1800
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
-
"epoch": 2.
|
| 1246 |
-
"eval_accuracy": 0.
|
| 1247 |
-
"eval_loss": 0.
|
| 1248 |
-
"eval_runtime":
|
| 1249 |
-
"eval_samples_per_second":
|
| 1250 |
-
"eval_steps_per_second":
|
| 1251 |
"step": 1800
|
| 1252 |
},
|
| 1253 |
{
|
| 1254 |
-
"epoch": 2.
|
| 1255 |
-
"learning_rate":
|
| 1256 |
-
"loss": 0.
|
| 1257 |
"step": 1810
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
-
"epoch": 2.
|
| 1261 |
-
"learning_rate":
|
| 1262 |
-
"loss": 0.
|
| 1263 |
"step": 1820
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
-
"epoch": 2.
|
| 1267 |
-
"learning_rate":
|
| 1268 |
-
"loss": 0.
|
| 1269 |
"step": 1830
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
-
"epoch": 2.
|
| 1273 |
-
"learning_rate":
|
| 1274 |
-
"loss": 0.
|
| 1275 |
"step": 1840
|
| 1276 |
},
|
| 1277 |
{
|
| 1278 |
-
"epoch": 2.
|
| 1279 |
-
"learning_rate":
|
| 1280 |
-
"loss": 0.
|
| 1281 |
"step": 1850
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
-
"epoch":
|
| 1285 |
-
"learning_rate":
|
| 1286 |
-
"loss": 0.
|
| 1287 |
"step": 1860
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
-
"epoch":
|
| 1291 |
-
"learning_rate":
|
| 1292 |
-
"loss": 0.
|
| 1293 |
"step": 1870
|
| 1294 |
},
|
| 1295 |
{
|
| 1296 |
-
"epoch":
|
| 1297 |
-
"learning_rate":
|
| 1298 |
-
"loss": 0.
|
| 1299 |
"step": 1880
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
-
"epoch":
|
| 1303 |
-
"learning_rate":
|
| 1304 |
-
"loss": 0.
|
| 1305 |
"step": 1890
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
-
"epoch":
|
| 1309 |
-
"learning_rate":
|
| 1310 |
-
"loss": 0.
|
| 1311 |
"step": 1900
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
-
"epoch":
|
| 1315 |
-
"eval_accuracy": 0.
|
| 1316 |
-
"eval_loss": 0.
|
| 1317 |
-
"eval_runtime":
|
| 1318 |
-
"eval_samples_per_second": 142.
|
| 1319 |
-
"eval_steps_per_second": 17.
|
| 1320 |
"step": 1900
|
| 1321 |
},
|
| 1322 |
{
|
| 1323 |
-
"epoch":
|
| 1324 |
-
"learning_rate":
|
| 1325 |
-
"loss": 0.
|
| 1326 |
"step": 1910
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
-
"epoch":
|
| 1330 |
-
"learning_rate":
|
| 1331 |
-
"loss": 0.
|
| 1332 |
"step": 1920
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
-
"epoch":
|
| 1336 |
-
"learning_rate":
|
| 1337 |
-
"loss": 0.
|
| 1338 |
"step": 1930
|
| 1339 |
},
|
| 1340 |
{
|
| 1341 |
-
"epoch":
|
| 1342 |
-
"learning_rate":
|
| 1343 |
-
"loss": 0.
|
| 1344 |
"step": 1940
|
| 1345 |
},
|
| 1346 |
{
|
| 1347 |
-
"epoch":
|
| 1348 |
-
"learning_rate":
|
| 1349 |
-
"loss": 0.
|
| 1350 |
"step": 1950
|
| 1351 |
},
|
| 1352 |
{
|
| 1353 |
-
"epoch":
|
| 1354 |
-
"learning_rate":
|
| 1355 |
-
"loss": 0.
|
| 1356 |
"step": 1960
|
| 1357 |
},
|
| 1358 |
{
|
| 1359 |
-
"epoch":
|
| 1360 |
-
"learning_rate":
|
| 1361 |
-
"loss": 0.
|
| 1362 |
"step": 1970
|
| 1363 |
},
|
| 1364 |
{
|
| 1365 |
-
"epoch":
|
| 1366 |
-
"learning_rate":
|
| 1367 |
-
"loss": 0.
|
| 1368 |
"step": 1980
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
-
"epoch":
|
| 1372 |
-
"learning_rate":
|
| 1373 |
-
"loss": 0.
|
| 1374 |
"step": 1990
|
| 1375 |
},
|
| 1376 |
{
|
| 1377 |
-
"epoch":
|
| 1378 |
-
"learning_rate":
|
| 1379 |
-
"loss": 0.
|
| 1380 |
"step": 2000
|
| 1381 |
},
|
| 1382 |
{
|
| 1383 |
-
"epoch":
|
| 1384 |
-
"eval_accuracy": 0.
|
| 1385 |
-
"eval_loss": 0.
|
| 1386 |
-
"eval_runtime":
|
| 1387 |
-
"eval_samples_per_second":
|
| 1388 |
-
"eval_steps_per_second": 16.
|
| 1389 |
"step": 2000
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
-
"epoch":
|
| 1393 |
-
"learning_rate":
|
| 1394 |
-
"loss": 0.
|
| 1395 |
"step": 2010
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
-
"epoch":
|
| 1399 |
-
"learning_rate":
|
| 1400 |
-
"loss": 0.
|
| 1401 |
"step": 2020
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
-
"epoch":
|
| 1405 |
-
"learning_rate":
|
| 1406 |
-
"loss": 0.
|
| 1407 |
"step": 2030
|
| 1408 |
},
|
| 1409 |
{
|
| 1410 |
-
"epoch":
|
| 1411 |
-
"learning_rate":
|
| 1412 |
-
"loss": 0.
|
| 1413 |
"step": 2040
|
| 1414 |
},
|
| 1415 |
{
|
| 1416 |
-
"epoch":
|
| 1417 |
-
"learning_rate":
|
| 1418 |
-
"loss": 0.
|
| 1419 |
"step": 2050
|
| 1420 |
},
|
| 1421 |
{
|
| 1422 |
-
"epoch":
|
| 1423 |
-
"learning_rate":
|
| 1424 |
-
"loss": 0.
|
| 1425 |
"step": 2060
|
| 1426 |
},
|
| 1427 |
{
|
| 1428 |
-
"epoch":
|
| 1429 |
-
"learning_rate":
|
| 1430 |
-
"loss": 0.
|
| 1431 |
"step": 2070
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
-
"epoch":
|
| 1435 |
-
"learning_rate":
|
| 1436 |
-
"loss": 0.
|
| 1437 |
"step": 2080
|
| 1438 |
},
|
| 1439 |
{
|
| 1440 |
-
"epoch":
|
| 1441 |
-
"learning_rate":
|
| 1442 |
-
"loss": 0.
|
| 1443 |
"step": 2090
|
| 1444 |
},
|
| 1445 |
{
|
| 1446 |
-
"epoch":
|
| 1447 |
-
"learning_rate":
|
| 1448 |
-
"loss": 0.
|
| 1449 |
"step": 2100
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
-
"epoch":
|
| 1453 |
-
"eval_accuracy": 0.
|
| 1454 |
-
"eval_loss": 0.
|
| 1455 |
-
"eval_runtime":
|
| 1456 |
-
"eval_samples_per_second":
|
| 1457 |
-
"eval_steps_per_second":
|
| 1458 |
"step": 2100
|
| 1459 |
},
|
| 1460 |
{
|
| 1461 |
-
"epoch":
|
| 1462 |
-
"learning_rate":
|
| 1463 |
-
"loss": 0.
|
| 1464 |
"step": 2110
|
| 1465 |
},
|
| 1466 |
{
|
| 1467 |
-
"epoch":
|
| 1468 |
-
"learning_rate":
|
| 1469 |
-
"loss": 0.
|
| 1470 |
"step": 2120
|
| 1471 |
},
|
| 1472 |
{
|
| 1473 |
-
"epoch":
|
| 1474 |
-
"learning_rate":
|
| 1475 |
-
"loss": 0.
|
| 1476 |
"step": 2130
|
| 1477 |
},
|
| 1478 |
{
|
| 1479 |
-
"epoch":
|
| 1480 |
-
"learning_rate":
|
| 1481 |
-
"loss": 0.
|
| 1482 |
"step": 2140
|
| 1483 |
},
|
| 1484 |
{
|
| 1485 |
-
"epoch":
|
| 1486 |
-
"learning_rate":
|
| 1487 |
-
"loss": 0.
|
| 1488 |
"step": 2150
|
| 1489 |
},
|
| 1490 |
{
|
| 1491 |
-
"epoch":
|
| 1492 |
-
"learning_rate":
|
| 1493 |
-
"loss": 0.
|
| 1494 |
"step": 2160
|
| 1495 |
},
|
| 1496 |
{
|
| 1497 |
-
"epoch":
|
| 1498 |
-
"learning_rate":
|
| 1499 |
-
"loss": 0.
|
| 1500 |
"step": 2170
|
| 1501 |
},
|
| 1502 |
{
|
| 1503 |
-
"epoch":
|
| 1504 |
-
"learning_rate":
|
| 1505 |
-
"loss": 0.
|
| 1506 |
"step": 2180
|
| 1507 |
},
|
| 1508 |
{
|
| 1509 |
-
"epoch":
|
| 1510 |
-
"learning_rate":
|
| 1511 |
-
"loss": 0.
|
| 1512 |
"step": 2190
|
| 1513 |
},
|
| 1514 |
{
|
| 1515 |
-
"epoch":
|
| 1516 |
-
"learning_rate":
|
| 1517 |
-
"loss": 0.
|
| 1518 |
"step": 2200
|
| 1519 |
},
|
| 1520 |
{
|
| 1521 |
-
"epoch":
|
| 1522 |
-
"eval_accuracy": 0.
|
| 1523 |
-
"eval_loss": 0.
|
| 1524 |
-
"eval_runtime":
|
| 1525 |
-
"eval_samples_per_second":
|
| 1526 |
-
"eval_steps_per_second":
|
| 1527 |
"step": 2200
|
| 1528 |
},
|
| 1529 |
{
|
| 1530 |
-
"epoch":
|
| 1531 |
-
"learning_rate":
|
| 1532 |
-
"loss": 0.
|
| 1533 |
"step": 2210
|
| 1534 |
},
|
| 1535 |
{
|
| 1536 |
-
"epoch":
|
| 1537 |
-
"learning_rate":
|
| 1538 |
-
"loss": 0.
|
| 1539 |
"step": 2220
|
| 1540 |
},
|
| 1541 |
{
|
| 1542 |
-
"epoch":
|
| 1543 |
-
"learning_rate":
|
| 1544 |
-
"loss": 0.
|
| 1545 |
"step": 2230
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
-
"epoch":
|
| 1549 |
-
"learning_rate":
|
| 1550 |
-
"loss": 0.
|
| 1551 |
"step": 2240
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
-
"epoch":
|
| 1555 |
-
"learning_rate":
|
| 1556 |
-
"loss": 0.
|
| 1557 |
"step": 2250
|
| 1558 |
},
|
| 1559 |
{
|
| 1560 |
-
"epoch":
|
| 1561 |
-
"learning_rate":
|
| 1562 |
-
"loss": 0.
|
| 1563 |
"step": 2260
|
| 1564 |
},
|
| 1565 |
{
|
| 1566 |
-
"epoch":
|
| 1567 |
-
"learning_rate":
|
| 1568 |
-
"loss": 0.
|
| 1569 |
"step": 2270
|
| 1570 |
},
|
| 1571 |
{
|
| 1572 |
-
"epoch":
|
| 1573 |
-
"learning_rate":
|
| 1574 |
-
"loss": 0.
|
| 1575 |
"step": 2280
|
| 1576 |
},
|
| 1577 |
{
|
| 1578 |
-
"epoch":
|
| 1579 |
-
"learning_rate":
|
| 1580 |
-
"loss": 0.
|
| 1581 |
"step": 2290
|
| 1582 |
},
|
| 1583 |
{
|
| 1584 |
-
"epoch":
|
| 1585 |
-
"learning_rate":
|
| 1586 |
-
"loss": 0.
|
| 1587 |
"step": 2300
|
| 1588 |
},
|
| 1589 |
{
|
| 1590 |
-
"epoch":
|
| 1591 |
-
"eval_accuracy": 0.
|
| 1592 |
-
"eval_loss": 0.
|
| 1593 |
-
"eval_runtime":
|
| 1594 |
-
"eval_samples_per_second":
|
| 1595 |
-
"eval_steps_per_second": 17.
|
| 1596 |
"step": 2300
|
| 1597 |
},
|
| 1598 |
{
|
| 1599 |
-
"epoch":
|
| 1600 |
-
"learning_rate":
|
| 1601 |
-
"loss": 0.
|
| 1602 |
"step": 2310
|
| 1603 |
},
|
| 1604 |
{
|
| 1605 |
-
"epoch":
|
| 1606 |
-
"learning_rate":
|
| 1607 |
-
"loss": 0.
|
| 1608 |
"step": 2320
|
| 1609 |
},
|
| 1610 |
{
|
| 1611 |
-
"epoch":
|
| 1612 |
-
"learning_rate":
|
| 1613 |
-
"loss": 0.
|
| 1614 |
"step": 2330
|
| 1615 |
},
|
| 1616 |
{
|
| 1617 |
-
"epoch":
|
| 1618 |
-
"learning_rate":
|
| 1619 |
-
"loss": 0.
|
| 1620 |
"step": 2340
|
| 1621 |
},
|
| 1622 |
{
|
| 1623 |
-
"epoch":
|
| 1624 |
-
"learning_rate":
|
| 1625 |
-
"loss": 0.
|
| 1626 |
"step": 2350
|
| 1627 |
},
|
| 1628 |
{
|
| 1629 |
-
"epoch":
|
| 1630 |
-
"learning_rate":
|
| 1631 |
-
"loss": 0.
|
| 1632 |
"step": 2360
|
| 1633 |
},
|
| 1634 |
{
|
| 1635 |
-
"epoch":
|
| 1636 |
-
"learning_rate":
|
| 1637 |
-
"loss": 0.
|
| 1638 |
"step": 2370
|
| 1639 |
},
|
| 1640 |
{
|
| 1641 |
-
"epoch":
|
| 1642 |
-
"learning_rate":
|
| 1643 |
-
"loss": 0.
|
| 1644 |
"step": 2380
|
| 1645 |
},
|
| 1646 |
{
|
| 1647 |
-
"epoch":
|
| 1648 |
-
"learning_rate": 6.
|
| 1649 |
-
"loss": 0.
|
| 1650 |
"step": 2390
|
| 1651 |
},
|
| 1652 |
{
|
| 1653 |
-
"epoch":
|
| 1654 |
-
"learning_rate":
|
| 1655 |
-
"loss": 0.
|
| 1656 |
"step": 2400
|
| 1657 |
},
|
| 1658 |
{
|
| 1659 |
-
"epoch":
|
| 1660 |
-
"eval_accuracy": 0.
|
| 1661 |
-
"eval_loss": 0.
|
| 1662 |
-
"eval_runtime":
|
| 1663 |
-
"eval_samples_per_second":
|
| 1664 |
-
"eval_steps_per_second":
|
| 1665 |
"step": 2400
|
| 1666 |
},
|
| 1667 |
{
|
| 1668 |
-
"epoch":
|
| 1669 |
-
"learning_rate":
|
| 1670 |
-
"loss": 0.
|
| 1671 |
"step": 2410
|
| 1672 |
},
|
| 1673 |
{
|
| 1674 |
-
"epoch":
|
| 1675 |
-
"learning_rate":
|
| 1676 |
-
"loss": 0.
|
| 1677 |
"step": 2420
|
| 1678 |
},
|
| 1679 |
{
|
| 1680 |
-
"epoch":
|
| 1681 |
-
"learning_rate":
|
| 1682 |
-
"loss": 0.
|
| 1683 |
"step": 2430
|
| 1684 |
},
|
| 1685 |
{
|
| 1686 |
-
"epoch":
|
| 1687 |
-
"learning_rate":
|
| 1688 |
-
"loss": 0.
|
| 1689 |
"step": 2440
|
| 1690 |
},
|
| 1691 |
{
|
| 1692 |
-
"epoch":
|
| 1693 |
-
"learning_rate":
|
| 1694 |
-
"loss": 0.
|
| 1695 |
"step": 2450
|
| 1696 |
},
|
| 1697 |
-
{
|
| 1698 |
-
"epoch": 2.81,
|
| 1699 |
-
"learning_rate": 5.9428571428571434e-05,
|
| 1700 |
-
"loss": 0.1636,
|
| 1701 |
-
"step": 2460
|
| 1702 |
-
},
|
| 1703 |
-
{
|
| 1704 |
-
"epoch": 2.82,
|
| 1705 |
-
"learning_rate": 5.885714285714285e-05,
|
| 1706 |
-
"loss": 0.1435,
|
| 1707 |
-
"step": 2470
|
| 1708 |
-
},
|
| 1709 |
-
{
|
| 1710 |
-
"epoch": 2.83,
|
| 1711 |
-
"learning_rate": 5.828571428571429e-05,
|
| 1712 |
-
"loss": 0.1587,
|
| 1713 |
-
"step": 2480
|
| 1714 |
-
},
|
| 1715 |
-
{
|
| 1716 |
-
"epoch": 2.85,
|
| 1717 |
-
"learning_rate": 5.771428571428572e-05,
|
| 1718 |
-
"loss": 0.1543,
|
| 1719 |
-
"step": 2490
|
| 1720 |
-
},
|
| 1721 |
-
{
|
| 1722 |
-
"epoch": 2.86,
|
| 1723 |
-
"learning_rate": 5.714285714285714e-05,
|
| 1724 |
-
"loss": 0.1669,
|
| 1725 |
-
"step": 2500
|
| 1726 |
-
},
|
| 1727 |
-
{
|
| 1728 |
-
"epoch": 2.86,
|
| 1729 |
-
"eval_accuracy": 0.8854371477257832,
|
| 1730 |
-
"eval_loss": 0.3132110834121704,
|
| 1731 |
-
"eval_runtime": 114.59,
|
| 1732 |
-
"eval_samples_per_second": 133.153,
|
| 1733 |
-
"eval_steps_per_second": 16.651,
|
| 1734 |
-
"step": 2500
|
| 1735 |
-
},
|
| 1736 |
-
{
|
| 1737 |
-
"epoch": 2.87,
|
| 1738 |
-
"learning_rate": 5.6571428571428574e-05,
|
| 1739 |
-
"loss": 0.1679,
|
| 1740 |
-
"step": 2510
|
| 1741 |
-
},
|
| 1742 |
-
{
|
| 1743 |
-
"epoch": 2.88,
|
| 1744 |
-
"learning_rate": 5.6000000000000006e-05,
|
| 1745 |
-
"loss": 0.1467,
|
| 1746 |
-
"step": 2520
|
| 1747 |
-
},
|
| 1748 |
-
{
|
| 1749 |
-
"epoch": 2.89,
|
| 1750 |
-
"learning_rate": 5.542857142857143e-05,
|
| 1751 |
-
"loss": 0.1495,
|
| 1752 |
-
"step": 2530
|
| 1753 |
-
},
|
| 1754 |
-
{
|
| 1755 |
-
"epoch": 2.9,
|
| 1756 |
-
"learning_rate": 5.485714285714286e-05,
|
| 1757 |
-
"loss": 0.184,
|
| 1758 |
-
"step": 2540
|
| 1759 |
-
},
|
| 1760 |
-
{
|
| 1761 |
-
"epoch": 2.91,
|
| 1762 |
-
"learning_rate": 5.428571428571428e-05,
|
| 1763 |
-
"loss": 0.1741,
|
| 1764 |
-
"step": 2550
|
| 1765 |
-
},
|
| 1766 |
-
{
|
| 1767 |
-
"epoch": 2.93,
|
| 1768 |
-
"learning_rate": 5.3714285714285714e-05,
|
| 1769 |
-
"loss": 0.1766,
|
| 1770 |
-
"step": 2560
|
| 1771 |
-
},
|
| 1772 |
-
{
|
| 1773 |
-
"epoch": 2.94,
|
| 1774 |
-
"learning_rate": 5.314285714285715e-05,
|
| 1775 |
-
"loss": 0.1745,
|
| 1776 |
-
"step": 2570
|
| 1777 |
-
},
|
| 1778 |
-
{
|
| 1779 |
-
"epoch": 2.95,
|
| 1780 |
-
"learning_rate": 5.257142857142857e-05,
|
| 1781 |
-
"loss": 0.156,
|
| 1782 |
-
"step": 2580
|
| 1783 |
-
},
|
| 1784 |
-
{
|
| 1785 |
-
"epoch": 2.96,
|
| 1786 |
-
"learning_rate": 5.2000000000000004e-05,
|
| 1787 |
-
"loss": 0.1478,
|
| 1788 |
-
"step": 2590
|
| 1789 |
-
},
|
| 1790 |
-
{
|
| 1791 |
-
"epoch": 2.97,
|
| 1792 |
-
"learning_rate": 5.142857142857143e-05,
|
| 1793 |
-
"loss": 0.1461,
|
| 1794 |
-
"step": 2600
|
| 1795 |
-
},
|
| 1796 |
-
{
|
| 1797 |
-
"epoch": 2.97,
|
| 1798 |
-
"eval_accuracy": 0.8883208808493905,
|
| 1799 |
-
"eval_loss": 0.3038625717163086,
|
| 1800 |
-
"eval_runtime": 109.6426,
|
| 1801 |
-
"eval_samples_per_second": 139.161,
|
| 1802 |
-
"eval_steps_per_second": 17.402,
|
| 1803 |
-
"step": 2600
|
| 1804 |
-
},
|
| 1805 |
-
{
|
| 1806 |
-
"epoch": 2.98,
|
| 1807 |
-
"learning_rate": 5.085714285714286e-05,
|
| 1808 |
-
"loss": 0.1587,
|
| 1809 |
-
"step": 2610
|
| 1810 |
-
},
|
| 1811 |
-
{
|
| 1812 |
-
"epoch": 2.99,
|
| 1813 |
-
"learning_rate": 5.028571428571429e-05,
|
| 1814 |
-
"loss": 0.1789,
|
| 1815 |
-
"step": 2620
|
| 1816 |
-
},
|
| 1817 |
-
{
|
| 1818 |
-
"epoch": 3.01,
|
| 1819 |
-
"learning_rate": 4.971428571428572e-05,
|
| 1820 |
-
"loss": 0.1253,
|
| 1821 |
-
"step": 2630
|
| 1822 |
-
},
|
| 1823 |
-
{
|
| 1824 |
-
"epoch": 3.02,
|
| 1825 |
-
"learning_rate": 4.9142857142857144e-05,
|
| 1826 |
-
"loss": 0.0804,
|
| 1827 |
-
"step": 2640
|
| 1828 |
-
},
|
| 1829 |
-
{
|
| 1830 |
-
"epoch": 3.03,
|
| 1831 |
-
"learning_rate": 4.8571428571428576e-05,
|
| 1832 |
-
"loss": 0.0721,
|
| 1833 |
-
"step": 2650
|
| 1834 |
-
},
|
| 1835 |
-
{
|
| 1836 |
-
"epoch": 3.04,
|
| 1837 |
-
"learning_rate": 4.8e-05,
|
| 1838 |
-
"loss": 0.0641,
|
| 1839 |
-
"step": 2660
|
| 1840 |
-
},
|
| 1841 |
-
{
|
| 1842 |
-
"epoch": 3.05,
|
| 1843 |
-
"learning_rate": 4.742857142857143e-05,
|
| 1844 |
-
"loss": 0.0762,
|
| 1845 |
-
"step": 2670
|
| 1846 |
-
},
|
| 1847 |
-
{
|
| 1848 |
-
"epoch": 3.06,
|
| 1849 |
-
"learning_rate": 4.685714285714286e-05,
|
| 1850 |
-
"loss": 0.0926,
|
| 1851 |
-
"step": 2680
|
| 1852 |
-
},
|
| 1853 |
-
{
|
| 1854 |
-
"epoch": 3.07,
|
| 1855 |
-
"learning_rate": 4.628571428571429e-05,
|
| 1856 |
-
"loss": 0.0677,
|
| 1857 |
-
"step": 2690
|
| 1858 |
-
},
|
| 1859 |
-
{
|
| 1860 |
-
"epoch": 3.09,
|
| 1861 |
-
"learning_rate": 4.5714285714285716e-05,
|
| 1862 |
-
"loss": 0.07,
|
| 1863 |
-
"step": 2700
|
| 1864 |
-
},
|
| 1865 |
-
{
|
| 1866 |
-
"epoch": 3.09,
|
| 1867 |
-
"eval_accuracy": 0.8921221654214183,
|
| 1868 |
-
"eval_loss": 0.3402355909347534,
|
| 1869 |
-
"eval_runtime": 178.6248,
|
| 1870 |
-
"eval_samples_per_second": 85.419,
|
| 1871 |
-
"eval_steps_per_second": 10.682,
|
| 1872 |
-
"step": 2700
|
| 1873 |
-
},
|
| 1874 |
-
{
|
| 1875 |
-
"epoch": 3.1,
|
| 1876 |
-
"learning_rate": 4.514285714285714e-05,
|
| 1877 |
-
"loss": 0.0897,
|
| 1878 |
-
"step": 2710
|
| 1879 |
-
},
|
| 1880 |
-
{
|
| 1881 |
-
"epoch": 3.11,
|
| 1882 |
-
"learning_rate": 4.4571428571428574e-05,
|
| 1883 |
-
"loss": 0.0644,
|
| 1884 |
-
"step": 2720
|
| 1885 |
-
},
|
| 1886 |
-
{
|
| 1887 |
-
"epoch": 3.12,
|
| 1888 |
-
"learning_rate": 4.4000000000000006e-05,
|
| 1889 |
-
"loss": 0.0591,
|
| 1890 |
-
"step": 2730
|
| 1891 |
-
},
|
| 1892 |
-
{
|
| 1893 |
-
"epoch": 3.13,
|
| 1894 |
-
"learning_rate": 4.342857142857143e-05,
|
| 1895 |
-
"loss": 0.067,
|
| 1896 |
-
"step": 2740
|
| 1897 |
-
},
|
| 1898 |
-
{
|
| 1899 |
-
"epoch": 3.14,
|
| 1900 |
-
"learning_rate": 4.2857142857142856e-05,
|
| 1901 |
-
"loss": 0.0915,
|
| 1902 |
-
"step": 2750
|
| 1903 |
-
},
|
| 1904 |
-
{
|
| 1905 |
-
"epoch": 3.15,
|
| 1906 |
-
"learning_rate": 4.228571428571429e-05,
|
| 1907 |
-
"loss": 0.0737,
|
| 1908 |
-
"step": 2760
|
| 1909 |
-
},
|
| 1910 |
-
{
|
| 1911 |
-
"epoch": 3.17,
|
| 1912 |
-
"learning_rate": 4.1714285714285714e-05,
|
| 1913 |
-
"loss": 0.0655,
|
| 1914 |
-
"step": 2770
|
| 1915 |
-
},
|
| 1916 |
-
{
|
| 1917 |
-
"epoch": 3.18,
|
| 1918 |
-
"learning_rate": 4.1142857142857146e-05,
|
| 1919 |
-
"loss": 0.0657,
|
| 1920 |
-
"step": 2780
|
| 1921 |
-
},
|
| 1922 |
-
{
|
| 1923 |
-
"epoch": 3.19,
|
| 1924 |
-
"learning_rate": 4.057142857142857e-05,
|
| 1925 |
-
"loss": 0.0651,
|
| 1926 |
-
"step": 2790
|
| 1927 |
-
},
|
| 1928 |
-
{
|
| 1929 |
-
"epoch": 3.2,
|
| 1930 |
-
"learning_rate": 4e-05,
|
| 1931 |
-
"loss": 0.0637,
|
| 1932 |
-
"step": 2800
|
| 1933 |
-
},
|
| 1934 |
-
{
|
| 1935 |
-
"epoch": 3.2,
|
| 1936 |
-
"eval_accuracy": 0.8944160440424695,
|
| 1937 |
-
"eval_loss": 0.34459471702575684,
|
| 1938 |
-
"eval_runtime": 102.3895,
|
| 1939 |
-
"eval_samples_per_second": 149.019,
|
| 1940 |
-
"eval_steps_per_second": 18.635,
|
| 1941 |
-
"step": 2800
|
| 1942 |
-
},
|
| 1943 |
-
{
|
| 1944 |
-
"epoch": 3.21,
|
| 1945 |
-
"learning_rate": 3.942857142857143e-05,
|
| 1946 |
-
"loss": 0.0629,
|
| 1947 |
-
"step": 2810
|
| 1948 |
-
},
|
| 1949 |
-
{
|
| 1950 |
-
"epoch": 3.22,
|
| 1951 |
-
"learning_rate": 3.885714285714286e-05,
|
| 1952 |
-
"loss": 0.0505,
|
| 1953 |
-
"step": 2820
|
| 1954 |
-
},
|
| 1955 |
-
{
|
| 1956 |
-
"epoch": 3.23,
|
| 1957 |
-
"learning_rate": 3.8285714285714286e-05,
|
| 1958 |
-
"loss": 0.0721,
|
| 1959 |
-
"step": 2830
|
| 1960 |
-
},
|
| 1961 |
-
{
|
| 1962 |
-
"epoch": 3.25,
|
| 1963 |
-
"learning_rate": 3.771428571428572e-05,
|
| 1964 |
-
"loss": 0.0686,
|
| 1965 |
-
"step": 2840
|
| 1966 |
-
},
|
| 1967 |
-
{
|
| 1968 |
-
"epoch": 3.26,
|
| 1969 |
-
"learning_rate": 3.7142857142857143e-05,
|
| 1970 |
-
"loss": 0.0703,
|
| 1971 |
-
"step": 2850
|
| 1972 |
-
},
|
| 1973 |
-
{
|
| 1974 |
-
"epoch": 3.27,
|
| 1975 |
-
"learning_rate": 3.6571428571428576e-05,
|
| 1976 |
-
"loss": 0.0768,
|
| 1977 |
-
"step": 2860
|
| 1978 |
-
},
|
| 1979 |
-
{
|
| 1980 |
-
"epoch": 3.28,
|
| 1981 |
-
"learning_rate": 3.6e-05,
|
| 1982 |
-
"loss": 0.0895,
|
| 1983 |
-
"step": 2870
|
| 1984 |
-
},
|
| 1985 |
-
{
|
| 1986 |
-
"epoch": 3.29,
|
| 1987 |
-
"learning_rate": 3.5428571428571426e-05,
|
| 1988 |
-
"loss": 0.0854,
|
| 1989 |
-
"step": 2880
|
| 1990 |
-
},
|
| 1991 |
-
{
|
| 1992 |
-
"epoch": 3.3,
|
| 1993 |
-
"learning_rate": 3.485714285714286e-05,
|
| 1994 |
-
"loss": 0.0843,
|
| 1995 |
-
"step": 2890
|
| 1996 |
-
},
|
| 1997 |
-
{
|
| 1998 |
-
"epoch": 3.31,
|
| 1999 |
-
"learning_rate": 3.428571428571429e-05,
|
| 2000 |
-
"loss": 0.0807,
|
| 2001 |
-
"step": 2900
|
| 2002 |
-
},
|
| 2003 |
-
{
|
| 2004 |
-
"epoch": 3.31,
|
| 2005 |
-
"eval_accuracy": 0.8946782015991611,
|
| 2006 |
-
"eval_loss": 0.3424582779407501,
|
| 2007 |
-
"eval_runtime": 112.9833,
|
| 2008 |
-
"eval_samples_per_second": 135.047,
|
| 2009 |
-
"eval_steps_per_second": 16.887,
|
| 2010 |
-
"step": 2900
|
| 2011 |
-
},
|
| 2012 |
-
{
|
| 2013 |
-
"epoch": 3.33,
|
| 2014 |
-
"learning_rate": 3.3714285714285716e-05,
|
| 2015 |
-
"loss": 0.0713,
|
| 2016 |
-
"step": 2910
|
| 2017 |
-
},
|
| 2018 |
-
{
|
| 2019 |
-
"epoch": 3.34,
|
| 2020 |
-
"learning_rate": 3.314285714285714e-05,
|
| 2021 |
-
"loss": 0.0551,
|
| 2022 |
-
"step": 2920
|
| 2023 |
-
},
|
| 2024 |
-
{
|
| 2025 |
-
"epoch": 3.35,
|
| 2026 |
-
"learning_rate": 3.257142857142857e-05,
|
| 2027 |
-
"loss": 0.0642,
|
| 2028 |
-
"step": 2930
|
| 2029 |
-
},
|
| 2030 |
-
{
|
| 2031 |
-
"epoch": 3.36,
|
| 2032 |
-
"learning_rate": 3.2000000000000005e-05,
|
| 2033 |
-
"loss": 0.0803,
|
| 2034 |
-
"step": 2940
|
| 2035 |
-
},
|
| 2036 |
-
{
|
| 2037 |
-
"epoch": 3.37,
|
| 2038 |
-
"learning_rate": 3.142857142857143e-05,
|
| 2039 |
-
"loss": 0.0849,
|
| 2040 |
-
"step": 2950
|
| 2041 |
-
},
|
| 2042 |
-
{
|
| 2043 |
-
"epoch": 3.38,
|
| 2044 |
-
"learning_rate": 3.0857142857142856e-05,
|
| 2045 |
-
"loss": 0.0545,
|
| 2046 |
-
"step": 2960
|
| 2047 |
-
},
|
| 2048 |
-
{
|
| 2049 |
-
"epoch": 3.39,
|
| 2050 |
-
"learning_rate": 3.0285714285714288e-05,
|
| 2051 |
-
"loss": 0.0546,
|
| 2052 |
-
"step": 2970
|
| 2053 |
-
},
|
| 2054 |
-
{
|
| 2055 |
-
"epoch": 3.41,
|
| 2056 |
-
"learning_rate": 2.9714285714285717e-05,
|
| 2057 |
-
"loss": 0.0732,
|
| 2058 |
-
"step": 2980
|
| 2059 |
-
},
|
| 2060 |
-
{
|
| 2061 |
-
"epoch": 3.42,
|
| 2062 |
-
"learning_rate": 2.9142857142857146e-05,
|
| 2063 |
-
"loss": 0.0676,
|
| 2064 |
-
"step": 2990
|
| 2065 |
-
},
|
| 2066 |
-
{
|
| 2067 |
-
"epoch": 3.43,
|
| 2068 |
-
"learning_rate": 2.857142857142857e-05,
|
| 2069 |
-
"loss": 0.0637,
|
| 2070 |
-
"step": 3000
|
| 2071 |
-
},
|
| 2072 |
-
{
|
| 2073 |
-
"epoch": 3.43,
|
| 2074 |
-
"eval_accuracy": 0.8964477651068292,
|
| 2075 |
-
"eval_loss": 0.33960404992103577,
|
| 2076 |
-
"eval_runtime": 107.6669,
|
| 2077 |
-
"eval_samples_per_second": 141.715,
|
| 2078 |
-
"eval_steps_per_second": 17.721,
|
| 2079 |
-
"step": 3000
|
| 2080 |
-
},
|
| 2081 |
-
{
|
| 2082 |
-
"epoch": 3.44,
|
| 2083 |
-
"learning_rate": 2.8000000000000003e-05,
|
| 2084 |
-
"loss": 0.0732,
|
| 2085 |
-
"step": 3010
|
| 2086 |
-
},
|
| 2087 |
-
{
|
| 2088 |
-
"epoch": 3.45,
|
| 2089 |
-
"learning_rate": 2.742857142857143e-05,
|
| 2090 |
-
"loss": 0.0704,
|
| 2091 |
-
"step": 3020
|
| 2092 |
-
},
|
| 2093 |
-
{
|
| 2094 |
-
"epoch": 3.46,
|
| 2095 |
-
"learning_rate": 2.6857142857142857e-05,
|
| 2096 |
-
"loss": 0.0801,
|
| 2097 |
-
"step": 3030
|
| 2098 |
-
},
|
| 2099 |
-
{
|
| 2100 |
-
"epoch": 3.47,
|
| 2101 |
-
"learning_rate": 2.6285714285714286e-05,
|
| 2102 |
-
"loss": 0.0606,
|
| 2103 |
-
"step": 3040
|
| 2104 |
-
},
|
| 2105 |
-
{
|
| 2106 |
-
"epoch": 3.49,
|
| 2107 |
-
"learning_rate": 2.5714285714285714e-05,
|
| 2108 |
-
"loss": 0.0553,
|
| 2109 |
-
"step": 3050
|
| 2110 |
-
},
|
| 2111 |
-
{
|
| 2112 |
-
"epoch": 3.5,
|
| 2113 |
-
"learning_rate": 2.5142857142857147e-05,
|
| 2114 |
-
"loss": 0.0411,
|
| 2115 |
-
"step": 3060
|
| 2116 |
-
},
|
| 2117 |
-
{
|
| 2118 |
-
"epoch": 3.51,
|
| 2119 |
-
"learning_rate": 2.4571428571428572e-05,
|
| 2120 |
-
"loss": 0.0535,
|
| 2121 |
-
"step": 3070
|
| 2122 |
-
},
|
| 2123 |
-
{
|
| 2124 |
-
"epoch": 3.52,
|
| 2125 |
-
"learning_rate": 2.4e-05,
|
| 2126 |
-
"loss": 0.0713,
|
| 2127 |
-
"step": 3080
|
| 2128 |
-
},
|
| 2129 |
-
{
|
| 2130 |
-
"epoch": 3.53,
|
| 2131 |
-
"learning_rate": 2.342857142857143e-05,
|
| 2132 |
-
"loss": 0.0562,
|
| 2133 |
-
"step": 3090
|
| 2134 |
-
},
|
| 2135 |
-
{
|
| 2136 |
-
"epoch": 3.54,
|
| 2137 |
-
"learning_rate": 2.2857142857142858e-05,
|
| 2138 |
-
"loss": 0.0535,
|
| 2139 |
-
"step": 3100
|
| 2140 |
-
},
|
| 2141 |
-
{
|
| 2142 |
-
"epoch": 3.54,
|
| 2143 |
-
"eval_accuracy": 0.8971031589985581,
|
| 2144 |
-
"eval_loss": 0.3407392203807831,
|
| 2145 |
-
"eval_runtime": 101.8347,
|
| 2146 |
-
"eval_samples_per_second": 149.831,
|
| 2147 |
-
"eval_steps_per_second": 18.736,
|
| 2148 |
-
"step": 3100
|
| 2149 |
-
},
|
| 2150 |
-
{
|
| 2151 |
-
"epoch": 3.55,
|
| 2152 |
-
"learning_rate": 2.2285714285714287e-05,
|
| 2153 |
-
"loss": 0.0685,
|
| 2154 |
-
"step": 3110
|
| 2155 |
-
},
|
| 2156 |
-
{
|
| 2157 |
-
"epoch": 3.57,
|
| 2158 |
-
"learning_rate": 2.1714285714285715e-05,
|
| 2159 |
-
"loss": 0.0666,
|
| 2160 |
-
"step": 3120
|
| 2161 |
-
},
|
| 2162 |
-
{
|
| 2163 |
-
"epoch": 3.58,
|
| 2164 |
-
"learning_rate": 2.1142857142857144e-05,
|
| 2165 |
-
"loss": 0.0745,
|
| 2166 |
-
"step": 3130
|
| 2167 |
-
},
|
| 2168 |
-
{
|
| 2169 |
-
"epoch": 3.59,
|
| 2170 |
-
"learning_rate": 2.0571428571428573e-05,
|
| 2171 |
-
"loss": 0.0507,
|
| 2172 |
-
"step": 3140
|
| 2173 |
-
},
|
| 2174 |
-
{
|
| 2175 |
-
"epoch": 3.6,
|
| 2176 |
-
"learning_rate": 2e-05,
|
| 2177 |
-
"loss": 0.0728,
|
| 2178 |
-
"step": 3150
|
| 2179 |
-
},
|
| 2180 |
-
{
|
| 2181 |
-
"epoch": 3.61,
|
| 2182 |
-
"learning_rate": 1.942857142857143e-05,
|
| 2183 |
-
"loss": 0.0476,
|
| 2184 |
-
"step": 3160
|
| 2185 |
-
},
|
| 2186 |
-
{
|
| 2187 |
-
"epoch": 3.62,
|
| 2188 |
-
"learning_rate": 1.885714285714286e-05,
|
| 2189 |
-
"loss": 0.0601,
|
| 2190 |
-
"step": 3170
|
| 2191 |
-
},
|
| 2192 |
-
{
|
| 2193 |
-
"epoch": 3.63,
|
| 2194 |
-
"learning_rate": 1.8285714285714288e-05,
|
| 2195 |
-
"loss": 0.0531,
|
| 2196 |
-
"step": 3180
|
| 2197 |
-
},
|
| 2198 |
-
{
|
| 2199 |
-
"epoch": 3.65,
|
| 2200 |
-
"learning_rate": 1.7714285714285713e-05,
|
| 2201 |
-
"loss": 0.0543,
|
| 2202 |
-
"step": 3190
|
| 2203 |
-
},
|
| 2204 |
-
{
|
| 2205 |
-
"epoch": 3.66,
|
| 2206 |
-
"learning_rate": 1.7142857142857145e-05,
|
| 2207 |
-
"loss": 0.064,
|
| 2208 |
-
"step": 3200
|
| 2209 |
-
},
|
| 2210 |
-
{
|
| 2211 |
-
"epoch": 3.66,
|
| 2212 |
-
"eval_accuracy": 0.9001835102896841,
|
| 2213 |
-
"eval_loss": 0.3420109748840332,
|
| 2214 |
-
"eval_runtime": 106.6302,
|
| 2215 |
-
"eval_samples_per_second": 143.093,
|
| 2216 |
-
"eval_steps_per_second": 17.894,
|
| 2217 |
-
"step": 3200
|
| 2218 |
-
},
|
| 2219 |
-
{
|
| 2220 |
-
"epoch": 3.67,
|
| 2221 |
-
"learning_rate": 1.657142857142857e-05,
|
| 2222 |
-
"loss": 0.0644,
|
| 2223 |
-
"step": 3210
|
| 2224 |
-
},
|
| 2225 |
-
{
|
| 2226 |
-
"epoch": 3.68,
|
| 2227 |
-
"learning_rate": 1.6000000000000003e-05,
|
| 2228 |
-
"loss": 0.0744,
|
| 2229 |
-
"step": 3220
|
| 2230 |
-
},
|
| 2231 |
-
{
|
| 2232 |
-
"epoch": 3.69,
|
| 2233 |
-
"learning_rate": 1.5428571428571428e-05,
|
| 2234 |
-
"loss": 0.0573,
|
| 2235 |
-
"step": 3230
|
| 2236 |
-
},
|
| 2237 |
-
{
|
| 2238 |
-
"epoch": 3.7,
|
| 2239 |
-
"learning_rate": 1.4857142857142858e-05,
|
| 2240 |
-
"loss": 0.0719,
|
| 2241 |
-
"step": 3240
|
| 2242 |
-
},
|
| 2243 |
-
{
|
| 2244 |
-
"epoch": 3.71,
|
| 2245 |
-
"learning_rate": 1.4285714285714285e-05,
|
| 2246 |
-
"loss": 0.0615,
|
| 2247 |
-
"step": 3250
|
| 2248 |
-
},
|
| 2249 |
-
{
|
| 2250 |
-
"epoch": 3.73,
|
| 2251 |
-
"learning_rate": 1.3714285714285716e-05,
|
| 2252 |
-
"loss": 0.0646,
|
| 2253 |
-
"step": 3260
|
| 2254 |
-
},
|
| 2255 |
-
{
|
| 2256 |
-
"epoch": 3.74,
|
| 2257 |
-
"learning_rate": 1.3142857142857143e-05,
|
| 2258 |
-
"loss": 0.0504,
|
| 2259 |
-
"step": 3270
|
| 2260 |
-
},
|
| 2261 |
-
{
|
| 2262 |
-
"epoch": 3.75,
|
| 2263 |
-
"learning_rate": 1.2571428571428573e-05,
|
| 2264 |
-
"loss": 0.0632,
|
| 2265 |
-
"step": 3280
|
| 2266 |
-
},
|
| 2267 |
-
{
|
| 2268 |
-
"epoch": 3.76,
|
| 2269 |
-
"learning_rate": 1.2e-05,
|
| 2270 |
-
"loss": 0.0698,
|
| 2271 |
-
"step": 3290
|
| 2272 |
-
},
|
| 2273 |
-
{
|
| 2274 |
-
"epoch": 3.77,
|
| 2275 |
-
"learning_rate": 1.1428571428571429e-05,
|
| 2276 |
-
"loss": 0.0707,
|
| 2277 |
-
"step": 3300
|
| 2278 |
-
},
|
| 2279 |
-
{
|
| 2280 |
-
"epoch": 3.77,
|
| 2281 |
-
"eval_accuracy": 0.8994625770087823,
|
| 2282 |
-
"eval_loss": 0.33135712146759033,
|
| 2283 |
-
"eval_runtime": 104.0422,
|
| 2284 |
-
"eval_samples_per_second": 146.652,
|
| 2285 |
-
"eval_steps_per_second": 18.339,
|
| 2286 |
-
"step": 3300
|
| 2287 |
-
},
|
| 2288 |
-
{
|
| 2289 |
-
"epoch": 3.78,
|
| 2290 |
-
"learning_rate": 1.0857142857142858e-05,
|
| 2291 |
-
"loss": 0.0645,
|
| 2292 |
-
"step": 3310
|
| 2293 |
-
},
|
| 2294 |
-
{
|
| 2295 |
-
"epoch": 3.79,
|
| 2296 |
-
"learning_rate": 1.0285714285714286e-05,
|
| 2297 |
-
"loss": 0.051,
|
| 2298 |
-
"step": 3320
|
| 2299 |
-
},
|
| 2300 |
-
{
|
| 2301 |
-
"epoch": 3.81,
|
| 2302 |
-
"learning_rate": 9.714285714285715e-06,
|
| 2303 |
-
"loss": 0.063,
|
| 2304 |
-
"step": 3330
|
| 2305 |
-
},
|
| 2306 |
-
{
|
| 2307 |
-
"epoch": 3.82,
|
| 2308 |
-
"learning_rate": 9.142857142857144e-06,
|
| 2309 |
-
"loss": 0.0667,
|
| 2310 |
-
"step": 3340
|
| 2311 |
-
},
|
| 2312 |
-
{
|
| 2313 |
-
"epoch": 3.83,
|
| 2314 |
-
"learning_rate": 8.571428571428573e-06,
|
| 2315 |
-
"loss": 0.0611,
|
| 2316 |
-
"step": 3350
|
| 2317 |
-
},
|
| 2318 |
-
{
|
| 2319 |
-
"epoch": 3.84,
|
| 2320 |
-
"learning_rate": 8.000000000000001e-06,
|
| 2321 |
-
"loss": 0.0633,
|
| 2322 |
-
"step": 3360
|
| 2323 |
-
},
|
| 2324 |
-
{
|
| 2325 |
-
"epoch": 3.85,
|
| 2326 |
-
"learning_rate": 7.428571428571429e-06,
|
| 2327 |
-
"loss": 0.054,
|
| 2328 |
-
"step": 3370
|
| 2329 |
-
},
|
| 2330 |
-
{
|
| 2331 |
-
"epoch": 3.86,
|
| 2332 |
-
"learning_rate": 6.857142857142858e-06,
|
| 2333 |
-
"loss": 0.0645,
|
| 2334 |
-
"step": 3380
|
| 2335 |
-
},
|
| 2336 |
-
{
|
| 2337 |
-
"epoch": 3.87,
|
| 2338 |
-
"learning_rate": 6.285714285714287e-06,
|
| 2339 |
-
"loss": 0.0632,
|
| 2340 |
-
"step": 3390
|
| 2341 |
-
},
|
| 2342 |
-
{
|
| 2343 |
-
"epoch": 3.89,
|
| 2344 |
-
"learning_rate": 5.7142857142857145e-06,
|
| 2345 |
-
"loss": 0.058,
|
| 2346 |
-
"step": 3400
|
| 2347 |
-
},
|
| 2348 |
-
{
|
| 2349 |
-
"epoch": 3.89,
|
| 2350 |
-
"eval_accuracy": 0.900249049678857,
|
| 2351 |
-
"eval_loss": 0.32864585518836975,
|
| 2352 |
-
"eval_runtime": 110.4035,
|
| 2353 |
-
"eval_samples_per_second": 138.202,
|
| 2354 |
-
"eval_steps_per_second": 17.282,
|
| 2355 |
-
"step": 3400
|
| 2356 |
-
},
|
| 2357 |
-
{
|
| 2358 |
-
"epoch": 3.9,
|
| 2359 |
-
"learning_rate": 5.142857142857143e-06,
|
| 2360 |
-
"loss": 0.0583,
|
| 2361 |
-
"step": 3410
|
| 2362 |
-
},
|
| 2363 |
-
{
|
| 2364 |
-
"epoch": 3.91,
|
| 2365 |
-
"learning_rate": 4.571428571428572e-06,
|
| 2366 |
-
"loss": 0.0833,
|
| 2367 |
-
"step": 3420
|
| 2368 |
-
},
|
| 2369 |
-
{
|
| 2370 |
-
"epoch": 3.92,
|
| 2371 |
-
"learning_rate": 4.000000000000001e-06,
|
| 2372 |
-
"loss": 0.0534,
|
| 2373 |
-
"step": 3430
|
| 2374 |
-
},
|
| 2375 |
-
{
|
| 2376 |
-
"epoch": 3.93,
|
| 2377 |
-
"learning_rate": 3.428571428571429e-06,
|
| 2378 |
-
"loss": 0.0789,
|
| 2379 |
-
"step": 3440
|
| 2380 |
-
},
|
| 2381 |
-
{
|
| 2382 |
-
"epoch": 3.94,
|
| 2383 |
-
"learning_rate": 2.8571428571428573e-06,
|
| 2384 |
-
"loss": 0.0608,
|
| 2385 |
-
"step": 3450
|
| 2386 |
-
},
|
| 2387 |
-
{
|
| 2388 |
-
"epoch": 3.95,
|
| 2389 |
-
"learning_rate": 2.285714285714286e-06,
|
| 2390 |
-
"loss": 0.0605,
|
| 2391 |
-
"step": 3460
|
| 2392 |
-
},
|
| 2393 |
-
{
|
| 2394 |
-
"epoch": 3.97,
|
| 2395 |
-
"learning_rate": 1.7142857142857145e-06,
|
| 2396 |
-
"loss": 0.0478,
|
| 2397 |
-
"step": 3470
|
| 2398 |
-
},
|
| 2399 |
{
|
| 2400 |
"epoch": 3.98,
|
| 2401 |
-
"learning_rate":
|
| 2402 |
-
"loss": 0.
|
| 2403 |
-
"step":
|
| 2404 |
-
},
|
| 2405 |
-
{
|
| 2406 |
-
"epoch": 3.99,
|
| 2407 |
-
"learning_rate": 5.714285714285715e-07,
|
| 2408 |
-
"loss": 0.0429,
|
| 2409 |
-
"step": 3490
|
| 2410 |
-
},
|
| 2411 |
-
{
|
| 2412 |
-
"epoch": 4.0,
|
| 2413 |
-
"learning_rate": 0.0,
|
| 2414 |
-
"loss": 0.048,
|
| 2415 |
-
"step": 3500
|
| 2416 |
},
|
| 2417 |
{
|
| 2418 |
"epoch": 4.0,
|
| 2419 |
-
"
|
| 2420 |
-
"
|
| 2421 |
-
"
|
| 2422 |
-
"eval_samples_per_second": 147.046,
|
| 2423 |
-
"eval_steps_per_second": 18.388,
|
| 2424 |
-
"step": 3500
|
| 2425 |
},
|
| 2426 |
{
|
| 2427 |
"epoch": 4.0,
|
| 2428 |
-
"step":
|
| 2429 |
-
"total_flos":
|
| 2430 |
-
"train_loss": 0.
|
| 2431 |
-
"train_runtime":
|
| 2432 |
-
"train_samples_per_second":
|
| 2433 |
-
"train_steps_per_second": 0.
|
| 2434 |
}
|
| 2435 |
],
|
| 2436 |
"logging_steps": 10,
|
| 2437 |
-
"max_steps":
|
| 2438 |
"num_train_epochs": 4,
|
| 2439 |
"save_steps": 100,
|
| 2440 |
-
"total_flos":
|
| 2441 |
"trial_name": null,
|
| 2442 |
"trial_params": null
|
| 2443 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.2805336117744446,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-tm/checkpoint-1800",
|
| 4 |
"epoch": 4.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 2472,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 0.00019919093851132686,
|
| 14 |
+
"loss": 1.2237,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.03,
|
| 19 |
+
"learning_rate": 0.00019838187702265374,
|
| 20 |
+
"loss": 1.1003,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.05,
|
| 25 |
+
"learning_rate": 0.0001975728155339806,
|
| 26 |
+
"loss": 1.0309,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.06,
|
| 31 |
+
"learning_rate": 0.00019676375404530745,
|
| 32 |
+
"loss": 0.9636,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.08,
|
| 37 |
+
"learning_rate": 0.00019595469255663433,
|
| 38 |
+
"loss": 0.8934,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
+
"epoch": 0.1,
|
| 43 |
+
"learning_rate": 0.00019514563106796118,
|
| 44 |
+
"loss": 0.8528,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 0.11,
|
| 49 |
+
"learning_rate": 0.00019433656957928804,
|
| 50 |
+
"loss": 0.8753,
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.13,
|
| 55 |
+
"learning_rate": 0.00019352750809061492,
|
| 56 |
+
"loss": 0.8577,
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 0.15,
|
| 61 |
+
"learning_rate": 0.00019271844660194177,
|
| 62 |
+
"loss": 0.821,
|
| 63 |
"step": 90
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 0.16,
|
| 67 |
+
"learning_rate": 0.00019190938511326862,
|
| 68 |
+
"loss": 0.7945,
|
| 69 |
"step": 100
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 0.16,
|
| 73 |
+
"eval_accuracy": 0.6915671572711284,
|
| 74 |
+
"eval_loss": 0.7725738286972046,
|
| 75 |
+
"eval_runtime": 149.5556,
|
| 76 |
+
"eval_samples_per_second": 144.468,
|
| 77 |
+
"eval_steps_per_second": 18.06,
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
+
"epoch": 0.18,
|
| 82 |
+
"learning_rate": 0.00019110032362459548,
|
| 83 |
+
"loss": 0.775,
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 0.19,
|
| 88 |
+
"learning_rate": 0.00019029126213592236,
|
| 89 |
+
"loss": 0.7727,
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 0.21,
|
| 94 |
+
"learning_rate": 0.0001894822006472492,
|
| 95 |
+
"loss": 0.7341,
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 0.23,
|
| 100 |
+
"learning_rate": 0.00018867313915857606,
|
| 101 |
+
"loss": 0.7121,
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 0.24,
|
| 106 |
+
"learning_rate": 0.00018786407766990291,
|
| 107 |
+
"loss": 0.7187,
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
+
"epoch": 0.26,
|
| 112 |
+
"learning_rate": 0.00018705501618122977,
|
| 113 |
+
"loss": 0.7346,
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.28,
|
| 118 |
+
"learning_rate": 0.00018624595469255665,
|
| 119 |
+
"loss": 0.6781,
|
| 120 |
"step": 170
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"epoch": 0.29,
|
| 124 |
+
"learning_rate": 0.0001854368932038835,
|
| 125 |
+
"loss": 0.7102,
|
| 126 |
"step": 180
|
| 127 |
},
|
| 128 |
{
|
| 129 |
+
"epoch": 0.31,
|
| 130 |
+
"learning_rate": 0.00018462783171521035,
|
| 131 |
+
"loss": 0.6582,
|
| 132 |
"step": 190
|
| 133 |
},
|
| 134 |
{
|
| 135 |
+
"epoch": 0.32,
|
| 136 |
+
"learning_rate": 0.0001838187702265372,
|
| 137 |
+
"loss": 0.6502,
|
| 138 |
"step": 200
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 0.32,
|
| 142 |
+
"eval_accuracy": 0.7326668518004258,
|
| 143 |
+
"eval_loss": 0.6581294536590576,
|
| 144 |
+
"eval_runtime": 156.2922,
|
| 145 |
+
"eval_samples_per_second": 138.241,
|
| 146 |
+
"eval_steps_per_second": 17.282,
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
+
"epoch": 0.34,
|
| 151 |
+
"learning_rate": 0.0001830097087378641,
|
| 152 |
+
"loss": 0.6807,
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 0.36,
|
| 157 |
+
"learning_rate": 0.00018220064724919094,
|
| 158 |
+
"loss": 0.683,
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 0.37,
|
| 163 |
+
"learning_rate": 0.0001813915857605178,
|
| 164 |
+
"loss": 0.6362,
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 0.39,
|
| 169 |
+
"learning_rate": 0.00018058252427184467,
|
| 170 |
+
"loss": 0.6389,
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 0.4,
|
| 175 |
+
"learning_rate": 0.00017977346278317153,
|
| 176 |
+
"loss": 0.6333,
|
| 177 |
"step": 250
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 0.42,
|
| 181 |
+
"learning_rate": 0.00017896440129449838,
|
| 182 |
+
"loss": 0.6116,
|
| 183 |
"step": 260
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"epoch": 0.44,
|
| 187 |
+
"learning_rate": 0.00017815533980582526,
|
| 188 |
+
"loss": 0.6021,
|
| 189 |
"step": 270
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 0.45,
|
| 193 |
+
"learning_rate": 0.0001773462783171521,
|
| 194 |
+
"loss": 0.5889,
|
| 195 |
"step": 280
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 0.47,
|
| 199 |
+
"learning_rate": 0.00017653721682847897,
|
| 200 |
+
"loss": 0.5977,
|
| 201 |
"step": 290
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 0.49,
|
| 205 |
+
"learning_rate": 0.00017572815533980585,
|
| 206 |
+
"loss": 0.5671,
|
| 207 |
"step": 300
|
| 208 |
},
|
| 209 |
{
|
| 210 |
+
"epoch": 0.49,
|
| 211 |
+
"eval_accuracy": 0.7589095621586597,
|
| 212 |
+
"eval_loss": 0.5891677737236023,
|
| 213 |
+
"eval_runtime": 152.5595,
|
| 214 |
+
"eval_samples_per_second": 141.623,
|
| 215 |
+
"eval_steps_per_second": 17.705,
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
+
"epoch": 0.5,
|
| 220 |
+
"learning_rate": 0.0001749190938511327,
|
| 221 |
+
"loss": 0.6115,
|
| 222 |
"step": 310
|
| 223 |
},
|
| 224 |
{
|
| 225 |
+
"epoch": 0.52,
|
| 226 |
+
"learning_rate": 0.00017411003236245955,
|
| 227 |
+
"loss": 0.5955,
|
| 228 |
"step": 320
|
| 229 |
},
|
| 230 |
{
|
| 231 |
+
"epoch": 0.53,
|
| 232 |
+
"learning_rate": 0.0001733009708737864,
|
| 233 |
+
"loss": 0.573,
|
| 234 |
"step": 330
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 0.55,
|
| 238 |
+
"learning_rate": 0.00017249190938511329,
|
| 239 |
+
"loss": 0.5681,
|
| 240 |
"step": 340
|
| 241 |
},
|
| 242 |
{
|
| 243 |
+
"epoch": 0.57,
|
| 244 |
+
"learning_rate": 0.00017168284789644014,
|
| 245 |
+
"loss": 0.5624,
|
| 246 |
"step": 350
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 0.58,
|
| 250 |
+
"learning_rate": 0.000170873786407767,
|
| 251 |
+
"loss": 0.5591,
|
| 252 |
"step": 360
|
| 253 |
},
|
| 254 |
{
|
| 255 |
+
"epoch": 0.6,
|
| 256 |
+
"learning_rate": 0.00017006472491909387,
|
| 257 |
+
"loss": 0.5836,
|
| 258 |
"step": 370
|
| 259 |
},
|
| 260 |
{
|
| 261 |
+
"epoch": 0.61,
|
| 262 |
+
"learning_rate": 0.00016925566343042073,
|
| 263 |
+
"loss": 0.5589,
|
| 264 |
"step": 380
|
| 265 |
},
|
| 266 |
{
|
| 267 |
+
"epoch": 0.63,
|
| 268 |
+
"learning_rate": 0.00016844660194174758,
|
| 269 |
+
"loss": 0.5433,
|
| 270 |
"step": 390
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"epoch": 0.65,
|
| 274 |
+
"learning_rate": 0.00016763754045307446,
|
| 275 |
+
"loss": 0.5625,
|
| 276 |
"step": 400
|
| 277 |
},
|
| 278 |
{
|
| 279 |
+
"epoch": 0.65,
|
| 280 |
+
"eval_accuracy": 0.7753864667222068,
|
| 281 |
+
"eval_loss": 0.5423863530158997,
|
| 282 |
+
"eval_runtime": 153.1943,
|
| 283 |
+
"eval_samples_per_second": 141.037,
|
| 284 |
+
"eval_steps_per_second": 17.631,
|
| 285 |
"step": 400
|
| 286 |
},
|
| 287 |
{
|
| 288 |
+
"epoch": 0.66,
|
| 289 |
+
"learning_rate": 0.0001668284789644013,
|
| 290 |
+
"loss": 0.5507,
|
| 291 |
"step": 410
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 0.68,
|
| 295 |
+
"learning_rate": 0.00016601941747572817,
|
| 296 |
+
"loss": 0.5439,
|
| 297 |
"step": 420
|
| 298 |
},
|
| 299 |
{
|
| 300 |
+
"epoch": 0.7,
|
| 301 |
+
"learning_rate": 0.00016521035598705505,
|
| 302 |
+
"loss": 0.5362,
|
| 303 |
"step": 430
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"epoch": 0.71,
|
| 307 |
+
"learning_rate": 0.0001644012944983819,
|
| 308 |
+
"loss": 0.5375,
|
| 309 |
"step": 440
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 0.73,
|
| 313 |
+
"learning_rate": 0.00016359223300970875,
|
| 314 |
+
"loss": 0.5379,
|
| 315 |
"step": 450
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 0.74,
|
| 319 |
+
"learning_rate": 0.0001627831715210356,
|
| 320 |
+
"loss": 0.5066,
|
| 321 |
"step": 460
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 0.76,
|
| 325 |
+
"learning_rate": 0.00016197411003236246,
|
| 326 |
+
"loss": 0.5419,
|
| 327 |
"step": 470
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 0.78,
|
| 331 |
+
"learning_rate": 0.0001611650485436893,
|
| 332 |
+
"loss": 0.476,
|
| 333 |
"step": 480
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 0.79,
|
| 337 |
+
"learning_rate": 0.0001603559870550162,
|
| 338 |
+
"loss": 0.522,
|
| 339 |
"step": 490
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 0.81,
|
| 343 |
+
"learning_rate": 0.00015954692556634304,
|
| 344 |
+
"loss": 0.5115,
|
| 345 |
"step": 500
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 0.81,
|
| 349 |
+
"eval_accuracy": 0.7930667407201704,
|
| 350 |
+
"eval_loss": 0.4989684820175171,
|
| 351 |
+
"eval_runtime": 148.2277,
|
| 352 |
+
"eval_samples_per_second": 145.762,
|
| 353 |
+
"eval_steps_per_second": 18.222,
|
| 354 |
"step": 500
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 0.83,
|
| 358 |
+
"learning_rate": 0.0001587378640776699,
|
| 359 |
+
"loss": 0.4866,
|
| 360 |
"step": 510
|
| 361 |
},
|
| 362 |
{
|
| 363 |
+
"epoch": 0.84,
|
| 364 |
+
"learning_rate": 0.00015792880258899675,
|
| 365 |
+
"loss": 0.495,
|
| 366 |
"step": 520
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"epoch": 0.86,
|
| 370 |
+
"learning_rate": 0.00015711974110032363,
|
| 371 |
+
"loss": 0.5037,
|
| 372 |
"step": 530
|
| 373 |
},
|
| 374 |
{
|
| 375 |
+
"epoch": 0.87,
|
| 376 |
+
"learning_rate": 0.00015631067961165048,
|
| 377 |
+
"loss": 0.4769,
|
| 378 |
"step": 540
|
| 379 |
},
|
| 380 |
{
|
| 381 |
+
"epoch": 0.89,
|
| 382 |
+
"learning_rate": 0.00015550161812297734,
|
| 383 |
+
"loss": 0.4818,
|
| 384 |
"step": 550
|
| 385 |
},
|
| 386 |
{
|
| 387 |
+
"epoch": 0.91,
|
| 388 |
+
"learning_rate": 0.00015469255663430422,
|
| 389 |
+
"loss": 0.4968,
|
| 390 |
"step": 560
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 0.92,
|
| 394 |
+
"learning_rate": 0.00015388349514563107,
|
| 395 |
+
"loss": 0.5027,
|
| 396 |
"step": 570
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"epoch": 0.94,
|
| 400 |
+
"learning_rate": 0.00015307443365695792,
|
| 401 |
+
"loss": 0.4727,
|
| 402 |
"step": 580
|
| 403 |
},
|
| 404 |
{
|
| 405 |
+
"epoch": 0.95,
|
| 406 |
+
"learning_rate": 0.0001522653721682848,
|
| 407 |
+
"loss": 0.4959,
|
| 408 |
"step": 590
|
| 409 |
},
|
| 410 |
{
|
| 411 |
+
"epoch": 0.97,
|
| 412 |
+
"learning_rate": 0.00015145631067961166,
|
| 413 |
+
"loss": 0.4643,
|
| 414 |
"step": 600
|
| 415 |
},
|
| 416 |
{
|
| 417 |
+
"epoch": 0.97,
|
| 418 |
+
"eval_accuracy": 0.8040359159492734,
|
| 419 |
+
"eval_loss": 0.4709508717060089,
|
| 420 |
+
"eval_runtime": 149.3766,
|
| 421 |
+
"eval_samples_per_second": 144.641,
|
| 422 |
+
"eval_steps_per_second": 18.082,
|
| 423 |
"step": 600
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"epoch": 0.99,
|
| 427 |
+
"learning_rate": 0.0001506472491909385,
|
| 428 |
+
"loss": 0.4721,
|
| 429 |
"step": 610
|
| 430 |
},
|
| 431 |
{
|
| 432 |
+
"epoch": 1.0,
|
| 433 |
+
"learning_rate": 0.0001498381877022654,
|
| 434 |
+
"loss": 0.4329,
|
| 435 |
"step": 620
|
| 436 |
},
|
| 437 |
{
|
| 438 |
+
"epoch": 1.02,
|
| 439 |
+
"learning_rate": 0.00014902912621359224,
|
| 440 |
+
"loss": 0.4058,
|
| 441 |
"step": 630
|
| 442 |
},
|
| 443 |
{
|
| 444 |
+
"epoch": 1.04,
|
| 445 |
+
"learning_rate": 0.0001482200647249191,
|
| 446 |
+
"loss": 0.3745,
|
| 447 |
"step": 640
|
| 448 |
},
|
| 449 |
{
|
| 450 |
+
"epoch": 1.05,
|
| 451 |
+
"learning_rate": 0.00014741100323624598,
|
| 452 |
+
"loss": 0.371,
|
| 453 |
"step": 650
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 1.07,
|
| 457 |
+
"learning_rate": 0.00014660194174757283,
|
| 458 |
+
"loss": 0.335,
|
| 459 |
"step": 660
|
| 460 |
},
|
| 461 |
{
|
| 462 |
+
"epoch": 1.08,
|
| 463 |
+
"learning_rate": 0.00014579288025889968,
|
| 464 |
+
"loss": 0.3992,
|
| 465 |
"step": 670
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 1.1,
|
| 469 |
+
"learning_rate": 0.00014498381877022656,
|
| 470 |
+
"loss": 0.3706,
|
| 471 |
"step": 680
|
| 472 |
},
|
| 473 |
{
|
| 474 |
+
"epoch": 1.12,
|
| 475 |
+
"learning_rate": 0.00014417475728155342,
|
| 476 |
+
"loss": 0.3747,
|
| 477 |
"step": 690
|
| 478 |
},
|
| 479 |
{
|
| 480 |
+
"epoch": 1.13,
|
| 481 |
+
"learning_rate": 0.00014336569579288027,
|
| 482 |
+
"loss": 0.3586,
|
| 483 |
"step": 700
|
| 484 |
},
|
| 485 |
{
|
| 486 |
+
"epoch": 1.13,
|
| 487 |
+
"eval_accuracy": 0.8220864574655188,
|
| 488 |
+
"eval_loss": 0.4336797595024109,
|
| 489 |
+
"eval_runtime": 152.9157,
|
| 490 |
+
"eval_samples_per_second": 141.293,
|
| 491 |
+
"eval_steps_per_second": 17.663,
|
| 492 |
"step": 700
|
| 493 |
},
|
| 494 |
{
|
| 495 |
+
"epoch": 1.15,
|
| 496 |
+
"learning_rate": 0.00014255663430420715,
|
| 497 |
+
"loss": 0.3842,
|
| 498 |
"step": 710
|
| 499 |
},
|
| 500 |
{
|
| 501 |
+
"epoch": 1.17,
|
| 502 |
+
"learning_rate": 0.000141747572815534,
|
| 503 |
+
"loss": 0.3386,
|
| 504 |
"step": 720
|
| 505 |
},
|
| 506 |
{
|
| 507 |
+
"epoch": 1.18,
|
| 508 |
+
"learning_rate": 0.00014093851132686086,
|
| 509 |
+
"loss": 0.3737,
|
| 510 |
"step": 730
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"epoch": 1.2,
|
| 514 |
+
"learning_rate": 0.0001401294498381877,
|
| 515 |
+
"loss": 0.3695,
|
| 516 |
"step": 740
|
| 517 |
},
|
| 518 |
{
|
| 519 |
+
"epoch": 1.21,
|
| 520 |
+
"learning_rate": 0.00013932038834951456,
|
| 521 |
+
"loss": 0.3766,
|
| 522 |
"step": 750
|
| 523 |
},
|
| 524 |
{
|
| 525 |
+
"epoch": 1.23,
|
| 526 |
+
"learning_rate": 0.00013851132686084141,
|
| 527 |
+
"loss": 0.3636,
|
| 528 |
"step": 760
|
| 529 |
},
|
| 530 |
{
|
| 531 |
+
"epoch": 1.25,
|
| 532 |
+
"learning_rate": 0.0001377022653721683,
|
| 533 |
+
"loss": 0.3419,
|
| 534 |
"step": 770
|
| 535 |
},
|
| 536 |
{
|
| 537 |
+
"epoch": 1.26,
|
| 538 |
+
"learning_rate": 0.00013689320388349515,
|
| 539 |
+
"loss": 0.382,
|
| 540 |
"step": 780
|
| 541 |
},
|
| 542 |
{
|
| 543 |
+
"epoch": 1.28,
|
| 544 |
+
"learning_rate": 0.000136084142394822,
|
| 545 |
+
"loss": 0.355,
|
| 546 |
"step": 790
|
| 547 |
},
|
| 548 |
{
|
| 549 |
+
"epoch": 1.29,
|
| 550 |
+
"learning_rate": 0.00013527508090614885,
|
| 551 |
+
"loss": 0.3421,
|
| 552 |
"step": 800
|
| 553 |
},
|
| 554 |
{
|
| 555 |
+
"epoch": 1.29,
|
| 556 |
+
"eval_accuracy": 0.8337498842914005,
|
| 557 |
+
"eval_loss": 0.4096640944480896,
|
| 558 |
+
"eval_runtime": 162.2619,
|
| 559 |
+
"eval_samples_per_second": 133.155,
|
| 560 |
+
"eval_steps_per_second": 16.646,
|
| 561 |
"step": 800
|
| 562 |
},
|
| 563 |
{
|
| 564 |
+
"epoch": 1.31,
|
| 565 |
+
"learning_rate": 0.00013446601941747573,
|
| 566 |
+
"loss": 0.3532,
|
| 567 |
"step": 810
|
| 568 |
},
|
| 569 |
{
|
| 570 |
+
"epoch": 1.33,
|
| 571 |
+
"learning_rate": 0.0001336569579288026,
|
| 572 |
+
"loss": 0.3251,
|
| 573 |
"step": 820
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"epoch": 1.34,
|
| 577 |
+
"learning_rate": 0.00013284789644012944,
|
| 578 |
+
"loss": 0.3406,
|
| 579 |
"step": 830
|
| 580 |
},
|
| 581 |
{
|
| 582 |
+
"epoch": 1.36,
|
| 583 |
+
"learning_rate": 0.00013203883495145632,
|
| 584 |
+
"loss": 0.356,
|
| 585 |
"step": 840
|
| 586 |
},
|
| 587 |
{
|
| 588 |
+
"epoch": 1.38,
|
| 589 |
+
"learning_rate": 0.00013122977346278317,
|
| 590 |
+
"loss": 0.3554,
|
| 591 |
"step": 850
|
| 592 |
},
|
| 593 |
{
|
| 594 |
+
"epoch": 1.39,
|
| 595 |
+
"learning_rate": 0.00013042071197411003,
|
| 596 |
+
"loss": 0.3502,
|
| 597 |
"step": 860
|
| 598 |
},
|
| 599 |
{
|
| 600 |
+
"epoch": 1.41,
|
| 601 |
+
"learning_rate": 0.0001296116504854369,
|
| 602 |
+
"loss": 0.3147,
|
| 603 |
"step": 870
|
| 604 |
},
|
| 605 |
{
|
| 606 |
+
"epoch": 1.42,
|
| 607 |
+
"learning_rate": 0.00012880258899676376,
|
| 608 |
+
"loss": 0.3399,
|
| 609 |
"step": 880
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 1.44,
|
| 613 |
+
"learning_rate": 0.00012799352750809061,
|
| 614 |
+
"loss": 0.3108,
|
| 615 |
"step": 890
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 1.46,
|
| 619 |
+
"learning_rate": 0.0001271844660194175,
|
| 620 |
+
"loss": 0.3478,
|
| 621 |
"step": 900
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 1.46,
|
| 625 |
+
"eval_accuracy": 0.844626492640933,
|
| 626 |
+
"eval_loss": 0.38169920444488525,
|
| 627 |
+
"eval_runtime": 156.4108,
|
| 628 |
+
"eval_samples_per_second": 138.136,
|
| 629 |
+
"eval_steps_per_second": 17.269,
|
| 630 |
"step": 900
|
| 631 |
},
|
| 632 |
{
|
| 633 |
+
"epoch": 1.47,
|
| 634 |
+
"learning_rate": 0.00012637540453074435,
|
| 635 |
+
"loss": 0.3344,
|
| 636 |
"step": 910
|
| 637 |
},
|
| 638 |
{
|
| 639 |
+
"epoch": 1.49,
|
| 640 |
+
"learning_rate": 0.0001255663430420712,
|
| 641 |
+
"loss": 0.323,
|
| 642 |
"step": 920
|
| 643 |
},
|
| 644 |
{
|
| 645 |
+
"epoch": 1.5,
|
| 646 |
+
"learning_rate": 0.00012475728155339805,
|
| 647 |
+
"loss": 0.3386,
|
| 648 |
"step": 930
|
| 649 |
},
|
| 650 |
{
|
| 651 |
+
"epoch": 1.52,
|
| 652 |
+
"learning_rate": 0.00012394822006472493,
|
| 653 |
+
"loss": 0.3236,
|
| 654 |
"step": 940
|
| 655 |
},
|
| 656 |
{
|
| 657 |
+
"epoch": 1.54,
|
| 658 |
+
"learning_rate": 0.00012313915857605179,
|
| 659 |
+
"loss": 0.332,
|
| 660 |
"step": 950
|
| 661 |
},
|
| 662 |
{
|
| 663 |
+
"epoch": 1.55,
|
| 664 |
+
"learning_rate": 0.00012233009708737864,
|
| 665 |
+
"loss": 0.3382,
|
| 666 |
"step": 960
|
| 667 |
},
|
| 668 |
{
|
| 669 |
+
"epoch": 1.57,
|
| 670 |
+
"learning_rate": 0.0001215210355987055,
|
| 671 |
+
"loss": 0.3033,
|
| 672 |
"step": 970
|
| 673 |
},
|
| 674 |
{
|
| 675 |
+
"epoch": 1.59,
|
| 676 |
+
"learning_rate": 0.00012071197411003237,
|
| 677 |
+
"loss": 0.3358,
|
| 678 |
"step": 980
|
| 679 |
},
|
| 680 |
{
|
| 681 |
+
"epoch": 1.6,
|
| 682 |
+
"learning_rate": 0.00011990291262135923,
|
| 683 |
+
"loss": 0.3098,
|
| 684 |
"step": 990
|
| 685 |
},
|
| 686 |
{
|
| 687 |
+
"epoch": 1.62,
|
| 688 |
+
"learning_rate": 0.00011909385113268609,
|
| 689 |
+
"loss": 0.2965,
|
| 690 |
"step": 1000
|
| 691 |
},
|
| 692 |
{
|
| 693 |
+
"epoch": 1.62,
|
| 694 |
+
"eval_accuracy": 0.8456910117559937,
|
| 695 |
+
"eval_loss": 0.37541016936302185,
|
| 696 |
+
"eval_runtime": 154.9131,
|
| 697 |
+
"eval_samples_per_second": 139.472,
|
| 698 |
+
"eval_steps_per_second": 17.436,
|
| 699 |
"step": 1000
|
| 700 |
},
|
| 701 |
{
|
| 702 |
+
"epoch": 1.63,
|
| 703 |
+
"learning_rate": 0.00011828478964401295,
|
| 704 |
+
"loss": 0.2885,
|
| 705 |
"step": 1010
|
| 706 |
},
|
| 707 |
{
|
| 708 |
+
"epoch": 1.65,
|
| 709 |
+
"learning_rate": 0.0001174757281553398,
|
| 710 |
+
"loss": 0.3348,
|
| 711 |
"step": 1020
|
| 712 |
},
|
| 713 |
{
|
| 714 |
+
"epoch": 1.67,
|
| 715 |
+
"learning_rate": 0.00011666666666666668,
|
| 716 |
+
"loss": 0.3234,
|
| 717 |
"step": 1030
|
| 718 |
},
|
| 719 |
{
|
| 720 |
+
"epoch": 1.68,
|
| 721 |
+
"learning_rate": 0.00011585760517799353,
|
| 722 |
+
"loss": 0.2942,
|
| 723 |
"step": 1040
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"epoch": 1.7,
|
| 727 |
+
"learning_rate": 0.00011504854368932039,
|
| 728 |
+
"loss": 0.3038,
|
| 729 |
"step": 1050
|
| 730 |
},
|
| 731 |
{
|
| 732 |
+
"epoch": 1.72,
|
| 733 |
+
"learning_rate": 0.00011423948220064727,
|
| 734 |
+
"loss": 0.2935,
|
| 735 |
"step": 1060
|
| 736 |
},
|
| 737 |
{
|
| 738 |
+
"epoch": 1.73,
|
| 739 |
+
"learning_rate": 0.00011343042071197412,
|
| 740 |
+
"loss": 0.3229,
|
| 741 |
"step": 1070
|
| 742 |
},
|
| 743 |
{
|
| 744 |
+
"epoch": 1.75,
|
| 745 |
+
"learning_rate": 0.00011262135922330097,
|
| 746 |
+
"loss": 0.2824,
|
| 747 |
"step": 1080
|
| 748 |
},
|
| 749 |
{
|
| 750 |
+
"epoch": 1.76,
|
| 751 |
+
"learning_rate": 0.00011181229773462785,
|
| 752 |
+
"loss": 0.3171,
|
| 753 |
"step": 1090
|
| 754 |
},
|
| 755 |
{
|
| 756 |
+
"epoch": 1.78,
|
| 757 |
+
"learning_rate": 0.0001110032362459547,
|
| 758 |
+
"loss": 0.2986,
|
| 759 |
"step": 1100
|
| 760 |
},
|
| 761 |
{
|
| 762 |
+
"epoch": 1.78,
|
| 763 |
+
"eval_accuracy": 0.8550402665926131,
|
| 764 |
+
"eval_loss": 0.3548347055912018,
|
| 765 |
+
"eval_runtime": 147.1647,
|
| 766 |
+
"eval_samples_per_second": 146.815,
|
| 767 |
+
"eval_steps_per_second": 18.354,
|
| 768 |
"step": 1100
|
| 769 |
},
|
| 770 |
{
|
| 771 |
+
"epoch": 1.8,
|
| 772 |
+
"learning_rate": 0.00011019417475728156,
|
| 773 |
+
"loss": 0.3117,
|
| 774 |
"step": 1110
|
| 775 |
},
|
| 776 |
{
|
| 777 |
+
"epoch": 1.81,
|
| 778 |
+
"learning_rate": 0.00010938511326860842,
|
| 779 |
+
"loss": 0.3094,
|
| 780 |
"step": 1120
|
| 781 |
},
|
| 782 |
{
|
| 783 |
+
"epoch": 1.83,
|
| 784 |
+
"learning_rate": 0.00010857605177993528,
|
| 785 |
+
"loss": 0.3089,
|
| 786 |
"step": 1130
|
| 787 |
},
|
| 788 |
{
|
| 789 |
+
"epoch": 1.84,
|
| 790 |
+
"learning_rate": 0.00010776699029126213,
|
| 791 |
+
"loss": 0.2854,
|
| 792 |
"step": 1140
|
| 793 |
},
|
| 794 |
{
|
| 795 |
+
"epoch": 1.86,
|
| 796 |
+
"learning_rate": 0.000106957928802589,
|
| 797 |
+
"loss": 0.2747,
|
| 798 |
"step": 1150
|
| 799 |
},
|
| 800 |
{
|
| 801 |
+
"epoch": 1.88,
|
| 802 |
+
"learning_rate": 0.00010614886731391586,
|
| 803 |
+
"loss": 0.2944,
|
| 804 |
"step": 1160
|
| 805 |
},
|
| 806 |
{
|
| 807 |
+
"epoch": 1.89,
|
| 808 |
+
"learning_rate": 0.00010533980582524272,
|
| 809 |
+
"loss": 0.307,
|
| 810 |
"step": 1170
|
| 811 |
},
|
| 812 |
{
|
| 813 |
+
"epoch": 1.91,
|
| 814 |
+
"learning_rate": 0.00010453074433656957,
|
| 815 |
+
"loss": 0.2829,
|
| 816 |
"step": 1180
|
| 817 |
},
|
| 818 |
{
|
| 819 |
+
"epoch": 1.93,
|
| 820 |
+
"learning_rate": 0.00010372168284789645,
|
| 821 |
+
"loss": 0.3038,
|
| 822 |
"step": 1190
|
| 823 |
},
|
| 824 |
{
|
| 825 |
+
"epoch": 1.94,
|
| 826 |
+
"learning_rate": 0.0001029126213592233,
|
| 827 |
+
"loss": 0.2932,
|
| 828 |
"step": 1200
|
| 829 |
},
|
| 830 |
{
|
| 831 |
+
"epoch": 1.94,
|
| 832 |
+
"eval_accuracy": 0.863325002314172,
|
| 833 |
+
"eval_loss": 0.3386863172054291,
|
| 834 |
+
"eval_runtime": 149.1575,
|
| 835 |
+
"eval_samples_per_second": 144.854,
|
| 836 |
+
"eval_steps_per_second": 18.108,
|
| 837 |
"step": 1200
|
| 838 |
},
|
| 839 |
{
|
| 840 |
+
"epoch": 1.96,
|
| 841 |
+
"learning_rate": 0.00010210355987055016,
|
| 842 |
+
"loss": 0.2873,
|
| 843 |
"step": 1210
|
| 844 |
},
|
| 845 |
{
|
| 846 |
+
"epoch": 1.97,
|
| 847 |
+
"learning_rate": 0.00010129449838187704,
|
| 848 |
+
"loss": 0.2857,
|
| 849 |
"step": 1220
|
| 850 |
},
|
| 851 |
{
|
| 852 |
+
"epoch": 1.99,
|
| 853 |
+
"learning_rate": 0.00010048543689320389,
|
| 854 |
+
"loss": 0.2621,
|
| 855 |
"step": 1230
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 2.01,
|
| 859 |
+
"learning_rate": 9.967637540453076e-05,
|
| 860 |
+
"loss": 0.2392,
|
| 861 |
"step": 1240
|
| 862 |
},
|
| 863 |
{
|
| 864 |
+
"epoch": 2.02,
|
| 865 |
+
"learning_rate": 9.886731391585761e-05,
|
| 866 |
+
"loss": 0.1795,
|
| 867 |
"step": 1250
|
| 868 |
},
|
| 869 |
{
|
| 870 |
+
"epoch": 2.04,
|
| 871 |
+
"learning_rate": 9.805825242718448e-05,
|
| 872 |
+
"loss": 0.1613,
|
| 873 |
"step": 1260
|
| 874 |
},
|
| 875 |
{
|
| 876 |
+
"epoch": 2.06,
|
| 877 |
+
"learning_rate": 9.724919093851133e-05,
|
| 878 |
+
"loss": 0.1699,
|
| 879 |
"step": 1270
|
| 880 |
},
|
| 881 |
{
|
| 882 |
+
"epoch": 2.07,
|
| 883 |
+
"learning_rate": 9.64401294498382e-05,
|
| 884 |
+
"loss": 0.1787,
|
| 885 |
"step": 1280
|
| 886 |
},
|
| 887 |
{
|
| 888 |
+
"epoch": 2.09,
|
| 889 |
+
"learning_rate": 9.563106796116505e-05,
|
| 890 |
+
"loss": 0.1731,
|
| 891 |
"step": 1290
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"epoch": 2.1,
|
| 895 |
+
"learning_rate": 9.48220064724919e-05,
|
| 896 |
+
"loss": 0.1701,
|
| 897 |
"step": 1300
|
| 898 |
},
|
| 899 |
{
|
| 900 |
+
"epoch": 2.1,
|
| 901 |
+
"eval_accuracy": 0.8677219290937702,
|
| 902 |
+
"eval_loss": 0.3415055274963379,
|
| 903 |
+
"eval_runtime": 160.755,
|
| 904 |
+
"eval_samples_per_second": 134.403,
|
| 905 |
+
"eval_steps_per_second": 16.802,
|
| 906 |
"step": 1300
|
| 907 |
},
|
| 908 |
{
|
| 909 |
+
"epoch": 2.12,
|
| 910 |
+
"learning_rate": 9.401294498381877e-05,
|
| 911 |
+
"loss": 0.1629,
|
| 912 |
"step": 1310
|
| 913 |
},
|
| 914 |
{
|
| 915 |
+
"epoch": 2.14,
|
| 916 |
+
"learning_rate": 9.320388349514564e-05,
|
| 917 |
+
"loss": 0.1694,
|
| 918 |
"step": 1320
|
| 919 |
},
|
| 920 |
{
|
| 921 |
+
"epoch": 2.15,
|
| 922 |
+
"learning_rate": 9.239482200647249e-05,
|
| 923 |
+
"loss": 0.1644,
|
| 924 |
"step": 1330
|
| 925 |
},
|
| 926 |
{
|
| 927 |
+
"epoch": 2.17,
|
| 928 |
+
"learning_rate": 9.158576051779936e-05,
|
| 929 |
+
"loss": 0.1631,
|
| 930 |
"step": 1340
|
| 931 |
},
|
| 932 |
{
|
| 933 |
+
"epoch": 2.18,
|
| 934 |
+
"learning_rate": 9.077669902912622e-05,
|
| 935 |
+
"loss": 0.1786,
|
| 936 |
"step": 1350
|
| 937 |
},
|
| 938 |
{
|
| 939 |
+
"epoch": 2.2,
|
| 940 |
+
"learning_rate": 8.996763754045308e-05,
|
| 941 |
+
"loss": 0.1792,
|
| 942 |
"step": 1360
|
| 943 |
},
|
| 944 |
{
|
| 945 |
+
"epoch": 2.22,
|
| 946 |
+
"learning_rate": 8.915857605177994e-05,
|
| 947 |
+
"loss": 0.1913,
|
| 948 |
"step": 1370
|
| 949 |
},
|
| 950 |
{
|
| 951 |
+
"epoch": 2.23,
|
| 952 |
+
"learning_rate": 8.834951456310681e-05,
|
| 953 |
+
"loss": 0.1751,
|
| 954 |
"step": 1380
|
| 955 |
},
|
| 956 |
{
|
| 957 |
+
"epoch": 2.25,
|
| 958 |
+
"learning_rate": 8.754045307443366e-05,
|
| 959 |
+
"loss": 0.165,
|
| 960 |
"step": 1390
|
| 961 |
},
|
| 962 |
{
|
| 963 |
+
"epoch": 2.27,
|
| 964 |
+
"learning_rate": 8.673139158576053e-05,
|
| 965 |
+
"loss": 0.1891,
|
| 966 |
"step": 1400
|
| 967 |
},
|
| 968 |
{
|
| 969 |
+
"epoch": 2.27,
|
| 970 |
+
"eval_accuracy": 0.876562066092752,
|
| 971 |
+
"eval_loss": 0.3259858787059784,
|
| 972 |
+
"eval_runtime": 179.0501,
|
| 973 |
+
"eval_samples_per_second": 120.67,
|
| 974 |
+
"eval_steps_per_second": 15.085,
|
| 975 |
"step": 1400
|
| 976 |
},
|
| 977 |
{
|
| 978 |
+
"epoch": 2.28,
|
| 979 |
+
"learning_rate": 8.592233009708738e-05,
|
| 980 |
+
"loss": 0.1645,
|
| 981 |
"step": 1410
|
| 982 |
},
|
| 983 |
{
|
| 984 |
+
"epoch": 2.3,
|
| 985 |
+
"learning_rate": 8.511326860841425e-05,
|
| 986 |
+
"loss": 0.1749,
|
| 987 |
"step": 1420
|
| 988 |
},
|
| 989 |
{
|
| 990 |
+
"epoch": 2.31,
|
| 991 |
+
"learning_rate": 8.43042071197411e-05,
|
| 992 |
+
"loss": 0.1635,
|
| 993 |
"step": 1430
|
| 994 |
},
|
| 995 |
{
|
| 996 |
+
"epoch": 2.33,
|
| 997 |
+
"learning_rate": 8.349514563106797e-05,
|
| 998 |
+
"loss": 0.1526,
|
| 999 |
"step": 1440
|
| 1000 |
},
|
| 1001 |
{
|
| 1002 |
+
"epoch": 2.35,
|
| 1003 |
+
"learning_rate": 8.268608414239482e-05,
|
| 1004 |
+
"loss": 0.1786,
|
| 1005 |
"step": 1450
|
| 1006 |
},
|
| 1007 |
{
|
| 1008 |
+
"epoch": 2.36,
|
| 1009 |
+
"learning_rate": 8.187702265372169e-05,
|
| 1010 |
+
"loss": 0.1726,
|
| 1011 |
"step": 1460
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
+
"epoch": 2.38,
|
| 1015 |
+
"learning_rate": 8.106796116504854e-05,
|
| 1016 |
+
"loss": 0.1625,
|
| 1017 |
"step": 1470
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
+
"epoch": 2.39,
|
| 1021 |
+
"learning_rate": 8.025889967637541e-05,
|
| 1022 |
+
"loss": 0.156,
|
| 1023 |
"step": 1480
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
+
"epoch": 2.41,
|
| 1027 |
+
"learning_rate": 7.944983818770227e-05,
|
| 1028 |
+
"loss": 0.1755,
|
| 1029 |
"step": 1490
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
+
"epoch": 2.43,
|
| 1033 |
+
"learning_rate": 7.864077669902913e-05,
|
| 1034 |
+
"loss": 0.1741,
|
| 1035 |
"step": 1500
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
+
"epoch": 2.43,
|
| 1039 |
+
"eval_accuracy": 0.8817920947884846,
|
| 1040 |
+
"eval_loss": 0.31030356884002686,
|
| 1041 |
+
"eval_runtime": 146.2047,
|
| 1042 |
+
"eval_samples_per_second": 147.779,
|
| 1043 |
+
"eval_steps_per_second": 18.474,
|
| 1044 |
"step": 1500
|
| 1045 |
},
|
| 1046 |
{
|
| 1047 |
+
"epoch": 2.44,
|
| 1048 |
+
"learning_rate": 7.7831715210356e-05,
|
| 1049 |
+
"loss": 0.1709,
|
| 1050 |
"step": 1510
|
| 1051 |
},
|
| 1052 |
{
|
| 1053 |
+
"epoch": 2.46,
|
| 1054 |
+
"learning_rate": 7.702265372168285e-05,
|
| 1055 |
+
"loss": 0.1689,
|
| 1056 |
"step": 1520
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
+
"epoch": 2.48,
|
| 1060 |
+
"learning_rate": 7.621359223300971e-05,
|
| 1061 |
+
"loss": 0.1458,
|
| 1062 |
"step": 1530
|
| 1063 |
},
|
| 1064 |
{
|
| 1065 |
+
"epoch": 2.49,
|
| 1066 |
+
"learning_rate": 7.540453074433658e-05,
|
| 1067 |
+
"loss": 0.1569,
|
| 1068 |
"step": 1540
|
| 1069 |
},
|
| 1070 |
{
|
| 1071 |
+
"epoch": 2.51,
|
| 1072 |
+
"learning_rate": 7.459546925566343e-05,
|
| 1073 |
+
"loss": 0.1689,
|
| 1074 |
"step": 1550
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
+
"epoch": 2.52,
|
| 1078 |
+
"learning_rate": 7.37864077669903e-05,
|
| 1079 |
+
"loss": 0.1571,
|
| 1080 |
"step": 1560
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
+
"epoch": 2.54,
|
| 1084 |
+
"learning_rate": 7.297734627831717e-05,
|
| 1085 |
+
"loss": 0.153,
|
| 1086 |
"step": 1570
|
| 1087 |
},
|
| 1088 |
{
|
| 1089 |
+
"epoch": 2.56,
|
| 1090 |
+
"learning_rate": 7.216828478964402e-05,
|
| 1091 |
+
"loss": 0.1585,
|
| 1092 |
"step": 1580
|
| 1093 |
},
|
| 1094 |
{
|
| 1095 |
+
"epoch": 2.57,
|
| 1096 |
+
"learning_rate": 7.135922330097087e-05,
|
| 1097 |
+
"loss": 0.1511,
|
| 1098 |
"step": 1590
|
| 1099 |
},
|
| 1100 |
{
|
| 1101 |
+
"epoch": 2.59,
|
| 1102 |
+
"learning_rate": 7.055016181229773e-05,
|
| 1103 |
+
"loss": 0.1542,
|
| 1104 |
"step": 1600
|
| 1105 |
},
|
| 1106 |
{
|
| 1107 |
+
"epoch": 2.59,
|
| 1108 |
+
"eval_accuracy": 0.8868832731648616,
|
| 1109 |
+
"eval_loss": 0.30611881613731384,
|
| 1110 |
+
"eval_runtime": 148.9187,
|
| 1111 |
+
"eval_samples_per_second": 145.086,
|
| 1112 |
+
"eval_steps_per_second": 18.137,
|
| 1113 |
"step": 1600
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
+
"epoch": 2.61,
|
| 1117 |
+
"learning_rate": 6.974110032362459e-05,
|
| 1118 |
+
"loss": 0.1684,
|
| 1119 |
"step": 1610
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
+
"epoch": 2.62,
|
| 1123 |
+
"learning_rate": 6.893203883495146e-05,
|
| 1124 |
+
"loss": 0.1593,
|
| 1125 |
"step": 1620
|
| 1126 |
},
|
| 1127 |
{
|
| 1128 |
+
"epoch": 2.64,
|
| 1129 |
+
"learning_rate": 6.812297734627831e-05,
|
| 1130 |
+
"loss": 0.1601,
|
| 1131 |
"step": 1630
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
+
"epoch": 2.65,
|
| 1135 |
+
"learning_rate": 6.731391585760518e-05,
|
| 1136 |
+
"loss": 0.1408,
|
| 1137 |
"step": 1640
|
| 1138 |
},
|
| 1139 |
{
|
| 1140 |
+
"epoch": 2.67,
|
| 1141 |
+
"learning_rate": 6.650485436893205e-05,
|
| 1142 |
+
"loss": 0.1542,
|
| 1143 |
"step": 1650
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
+
"epoch": 2.69,
|
| 1147 |
+
"learning_rate": 6.56957928802589e-05,
|
| 1148 |
+
"loss": 0.1609,
|
| 1149 |
"step": 1660
|
| 1150 |
},
|
| 1151 |
{
|
| 1152 |
+
"epoch": 2.7,
|
| 1153 |
+
"learning_rate": 6.488673139158577e-05,
|
| 1154 |
+
"loss": 0.1528,
|
| 1155 |
"step": 1670
|
| 1156 |
},
|
| 1157 |
{
|
| 1158 |
+
"epoch": 2.72,
|
| 1159 |
+
"learning_rate": 6.407766990291263e-05,
|
| 1160 |
+
"loss": 0.1447,
|
| 1161 |
"step": 1680
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
+
"epoch": 2.73,
|
| 1165 |
+
"learning_rate": 6.326860841423949e-05,
|
| 1166 |
+
"loss": 0.1522,
|
| 1167 |
"step": 1690
|
| 1168 |
},
|
| 1169 |
{
|
| 1170 |
+
"epoch": 2.75,
|
| 1171 |
+
"learning_rate": 6.245954692556635e-05,
|
| 1172 |
+
"loss": 0.172,
|
| 1173 |
"step": 1700
|
| 1174 |
},
|
| 1175 |
{
|
| 1176 |
+
"epoch": 2.75,
|
| 1177 |
+
"eval_accuracy": 0.8887808941960567,
|
| 1178 |
+
"eval_loss": 0.2924608886241913,
|
| 1179 |
+
"eval_runtime": 152.8021,
|
| 1180 |
+
"eval_samples_per_second": 141.399,
|
| 1181 |
+
"eval_steps_per_second": 17.676,
|
| 1182 |
"step": 1700
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
+
"epoch": 2.77,
|
| 1186 |
+
"learning_rate": 6.16504854368932e-05,
|
| 1187 |
+
"loss": 0.1467,
|
| 1188 |
"step": 1710
|
| 1189 |
},
|
| 1190 |
{
|
| 1191 |
+
"epoch": 2.78,
|
| 1192 |
+
"learning_rate": 6.0841423948220065e-05,
|
| 1193 |
+
"loss": 0.1678,
|
| 1194 |
"step": 1720
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
+
"epoch": 2.8,
|
| 1198 |
+
"learning_rate": 6.003236245954693e-05,
|
| 1199 |
+
"loss": 0.1538,
|
| 1200 |
"step": 1730
|
| 1201 |
},
|
| 1202 |
{
|
| 1203 |
+
"epoch": 2.82,
|
| 1204 |
+
"learning_rate": 5.9223300970873785e-05,
|
| 1205 |
+
"loss": 0.1609,
|
| 1206 |
"step": 1740
|
| 1207 |
},
|
| 1208 |
{
|
| 1209 |
+
"epoch": 2.83,
|
| 1210 |
+
"learning_rate": 5.841423948220065e-05,
|
| 1211 |
+
"loss": 0.1454,
|
| 1212 |
"step": 1750
|
| 1213 |
},
|
| 1214 |
{
|
| 1215 |
+
"epoch": 2.85,
|
| 1216 |
+
"learning_rate": 5.760517799352752e-05,
|
| 1217 |
+
"loss": 0.1505,
|
| 1218 |
"step": 1760
|
| 1219 |
},
|
| 1220 |
{
|
| 1221 |
+
"epoch": 2.86,
|
| 1222 |
+
"learning_rate": 5.679611650485437e-05,
|
| 1223 |
+
"loss": 0.1344,
|
| 1224 |
"step": 1770
|
| 1225 |
},
|
| 1226 |
{
|
| 1227 |
+
"epoch": 2.88,
|
| 1228 |
+
"learning_rate": 5.598705501618123e-05,
|
| 1229 |
+
"loss": 0.1425,
|
| 1230 |
"step": 1780
|
| 1231 |
},
|
| 1232 |
{
|
| 1233 |
+
"epoch": 2.9,
|
| 1234 |
+
"learning_rate": 5.51779935275081e-05,
|
| 1235 |
+
"loss": 0.143,
|
| 1236 |
"step": 1790
|
| 1237 |
},
|
| 1238 |
{
|
| 1239 |
+
"epoch": 2.91,
|
| 1240 |
+
"learning_rate": 5.436893203883495e-05,
|
| 1241 |
+
"loss": 0.1575,
|
| 1242 |
"step": 1800
|
| 1243 |
},
|
| 1244 |
{
|
| 1245 |
+
"epoch": 2.91,
|
| 1246 |
+
"eval_accuracy": 0.8962788114412663,
|
| 1247 |
+
"eval_loss": 0.2805336117744446,
|
| 1248 |
+
"eval_runtime": 148.8673,
|
| 1249 |
+
"eval_samples_per_second": 145.136,
|
| 1250 |
+
"eval_steps_per_second": 18.144,
|
| 1251 |
"step": 1800
|
| 1252 |
},
|
| 1253 |
{
|
| 1254 |
+
"epoch": 2.93,
|
| 1255 |
+
"learning_rate": 5.355987055016182e-05,
|
| 1256 |
+
"loss": 0.1469,
|
| 1257 |
"step": 1810
|
| 1258 |
},
|
| 1259 |
{
|
| 1260 |
+
"epoch": 2.94,
|
| 1261 |
+
"learning_rate": 5.275080906148867e-05,
|
| 1262 |
+
"loss": 0.1529,
|
| 1263 |
"step": 1820
|
| 1264 |
},
|
| 1265 |
{
|
| 1266 |
+
"epoch": 2.96,
|
| 1267 |
+
"learning_rate": 5.194174757281554e-05,
|
| 1268 |
+
"loss": 0.1346,
|
| 1269 |
"step": 1830
|
| 1270 |
},
|
| 1271 |
{
|
| 1272 |
+
"epoch": 2.98,
|
| 1273 |
+
"learning_rate": 5.11326860841424e-05,
|
| 1274 |
+
"loss": 0.1406,
|
| 1275 |
"step": 1840
|
| 1276 |
},
|
| 1277 |
{
|
| 1278 |
+
"epoch": 2.99,
|
| 1279 |
+
"learning_rate": 5.032362459546926e-05,
|
| 1280 |
+
"loss": 0.1352,
|
| 1281 |
"step": 1850
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
+
"epoch": 3.01,
|
| 1285 |
+
"learning_rate": 4.951456310679612e-05,
|
| 1286 |
+
"loss": 0.1039,
|
| 1287 |
"step": 1860
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
+
"epoch": 3.03,
|
| 1291 |
+
"learning_rate": 4.870550161812298e-05,
|
| 1292 |
+
"loss": 0.0652,
|
| 1293 |
"step": 1870
|
| 1294 |
},
|
| 1295 |
{
|
| 1296 |
+
"epoch": 3.04,
|
| 1297 |
+
"learning_rate": 4.789644012944984e-05,
|
| 1298 |
+
"loss": 0.0643,
|
| 1299 |
"step": 1880
|
| 1300 |
},
|
| 1301 |
{
|
| 1302 |
+
"epoch": 3.06,
|
| 1303 |
+
"learning_rate": 4.7087378640776703e-05,
|
| 1304 |
+
"loss": 0.0767,
|
| 1305 |
"step": 1890
|
| 1306 |
},
|
| 1307 |
{
|
| 1308 |
+
"epoch": 3.07,
|
| 1309 |
+
"learning_rate": 4.627831715210356e-05,
|
| 1310 |
+
"loss": 0.0698,
|
| 1311 |
"step": 1900
|
| 1312 |
},
|
| 1313 |
{
|
| 1314 |
+
"epoch": 3.07,
|
| 1315 |
+
"eval_accuracy": 0.8984078496713875,
|
| 1316 |
+
"eval_loss": 0.3030557930469513,
|
| 1317 |
+
"eval_runtime": 151.4685,
|
| 1318 |
+
"eval_samples_per_second": 142.643,
|
| 1319 |
+
"eval_steps_per_second": 17.832,
|
| 1320 |
"step": 1900
|
| 1321 |
},
|
| 1322 |
{
|
| 1323 |
+
"epoch": 3.09,
|
| 1324 |
+
"learning_rate": 4.546925566343042e-05,
|
| 1325 |
+
"loss": 0.0714,
|
| 1326 |
"step": 1910
|
| 1327 |
},
|
| 1328 |
{
|
| 1329 |
+
"epoch": 3.11,
|
| 1330 |
+
"learning_rate": 4.466019417475728e-05,
|
| 1331 |
+
"loss": 0.0615,
|
| 1332 |
"step": 1920
|
| 1333 |
},
|
| 1334 |
{
|
| 1335 |
+
"epoch": 3.12,
|
| 1336 |
+
"learning_rate": 4.385113268608414e-05,
|
| 1337 |
+
"loss": 0.0606,
|
| 1338 |
"step": 1930
|
| 1339 |
},
|
| 1340 |
{
|
| 1341 |
+
"epoch": 3.14,
|
| 1342 |
+
"learning_rate": 4.3042071197411e-05,
|
| 1343 |
+
"loss": 0.067,
|
| 1344 |
"step": 1940
|
| 1345 |
},
|
| 1346 |
{
|
| 1347 |
+
"epoch": 3.16,
|
| 1348 |
+
"learning_rate": 4.223300970873786e-05,
|
| 1349 |
+
"loss": 0.0671,
|
| 1350 |
"step": 1950
|
| 1351 |
},
|
| 1352 |
{
|
| 1353 |
+
"epoch": 3.17,
|
| 1354 |
+
"learning_rate": 4.142394822006473e-05,
|
| 1355 |
+
"loss": 0.059,
|
| 1356 |
"step": 1960
|
| 1357 |
},
|
| 1358 |
{
|
| 1359 |
+
"epoch": 3.19,
|
| 1360 |
+
"learning_rate": 4.061488673139159e-05,
|
| 1361 |
+
"loss": 0.0617,
|
| 1362 |
"step": 1970
|
| 1363 |
},
|
| 1364 |
{
|
| 1365 |
+
"epoch": 3.2,
|
| 1366 |
+
"learning_rate": 3.980582524271845e-05,
|
| 1367 |
+
"loss": 0.0702,
|
| 1368 |
"step": 1980
|
| 1369 |
},
|
| 1370 |
{
|
| 1371 |
+
"epoch": 3.22,
|
| 1372 |
+
"learning_rate": 3.899676375404531e-05,
|
| 1373 |
+
"loss": 0.0596,
|
| 1374 |
"step": 1990
|
| 1375 |
},
|
| 1376 |
{
|
| 1377 |
+
"epoch": 3.24,
|
| 1378 |
+
"learning_rate": 3.818770226537217e-05,
|
| 1379 |
+
"loss": 0.0671,
|
| 1380 |
"step": 2000
|
| 1381 |
},
|
| 1382 |
{
|
| 1383 |
+
"epoch": 3.24,
|
| 1384 |
+
"eval_accuracy": 0.9008608719800055,
|
| 1385 |
+
"eval_loss": 0.3075368106365204,
|
| 1386 |
+
"eval_runtime": 167.8743,
|
| 1387 |
+
"eval_samples_per_second": 128.703,
|
| 1388 |
+
"eval_steps_per_second": 16.089,
|
| 1389 |
"step": 2000
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
+
"epoch": 3.25,
|
| 1393 |
+
"learning_rate": 3.737864077669903e-05,
|
| 1394 |
+
"loss": 0.0588,
|
| 1395 |
"step": 2010
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
+
"epoch": 3.27,
|
| 1399 |
+
"learning_rate": 3.656957928802589e-05,
|
| 1400 |
+
"loss": 0.0551,
|
| 1401 |
"step": 2020
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
+
"epoch": 3.28,
|
| 1405 |
+
"learning_rate": 3.5760517799352755e-05,
|
| 1406 |
+
"loss": 0.0497,
|
| 1407 |
"step": 2030
|
| 1408 |
},
|
| 1409 |
{
|
| 1410 |
+
"epoch": 3.3,
|
| 1411 |
+
"learning_rate": 3.4951456310679615e-05,
|
| 1412 |
+
"loss": 0.0644,
|
| 1413 |
"step": 2040
|
| 1414 |
},
|
| 1415 |
{
|
| 1416 |
+
"epoch": 3.32,
|
| 1417 |
+
"learning_rate": 3.4142394822006475e-05,
|
| 1418 |
+
"loss": 0.0622,
|
| 1419 |
"step": 2050
|
| 1420 |
},
|
| 1421 |
{
|
| 1422 |
+
"epoch": 3.33,
|
| 1423 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 1424 |
+
"loss": 0.0637,
|
| 1425 |
"step": 2060
|
| 1426 |
},
|
| 1427 |
{
|
| 1428 |
+
"epoch": 3.35,
|
| 1429 |
+
"learning_rate": 3.2524271844660195e-05,
|
| 1430 |
+
"loss": 0.0633,
|
| 1431 |
"step": 2070
|
| 1432 |
},
|
| 1433 |
{
|
| 1434 |
+
"epoch": 3.37,
|
| 1435 |
+
"learning_rate": 3.1715210355987055e-05,
|
| 1436 |
+
"loss": 0.0633,
|
| 1437 |
"step": 2080
|
| 1438 |
},
|
| 1439 |
{
|
| 1440 |
+
"epoch": 3.38,
|
| 1441 |
+
"learning_rate": 3.0906148867313915e-05,
|
| 1442 |
+
"loss": 0.0576,
|
| 1443 |
"step": 2090
|
| 1444 |
},
|
| 1445 |
{
|
| 1446 |
+
"epoch": 3.4,
|
| 1447 |
+
"learning_rate": 3.0097087378640774e-05,
|
| 1448 |
+
"loss": 0.0576,
|
| 1449 |
"step": 2100
|
| 1450 |
},
|
| 1451 |
{
|
| 1452 |
+
"epoch": 3.4,
|
| 1453 |
+
"eval_accuracy": 0.9028510598907711,
|
| 1454 |
+
"eval_loss": 0.30513760447502136,
|
| 1455 |
+
"eval_runtime": 172.0167,
|
| 1456 |
+
"eval_samples_per_second": 125.604,
|
| 1457 |
+
"eval_steps_per_second": 15.702,
|
| 1458 |
"step": 2100
|
| 1459 |
},
|
| 1460 |
{
|
| 1461 |
+
"epoch": 3.41,
|
| 1462 |
+
"learning_rate": 2.928802588996764e-05,
|
| 1463 |
+
"loss": 0.0624,
|
| 1464 |
"step": 2110
|
| 1465 |
},
|
| 1466 |
{
|
| 1467 |
+
"epoch": 3.43,
|
| 1468 |
+
"learning_rate": 2.84789644012945e-05,
|
| 1469 |
+
"loss": 0.0518,
|
| 1470 |
"step": 2120
|
| 1471 |
},
|
| 1472 |
{
|
| 1473 |
+
"epoch": 3.45,
|
| 1474 |
+
"learning_rate": 2.766990291262136e-05,
|
| 1475 |
+
"loss": 0.0581,
|
| 1476 |
"step": 2130
|
| 1477 |
},
|
| 1478 |
{
|
| 1479 |
+
"epoch": 3.46,
|
| 1480 |
+
"learning_rate": 2.6860841423948217e-05,
|
| 1481 |
+
"loss": 0.0516,
|
| 1482 |
"step": 2140
|
| 1483 |
},
|
| 1484 |
{
|
| 1485 |
+
"epoch": 3.48,
|
| 1486 |
+
"learning_rate": 2.6051779935275084e-05,
|
| 1487 |
+
"loss": 0.0595,
|
| 1488 |
"step": 2150
|
| 1489 |
},
|
| 1490 |
{
|
| 1491 |
+
"epoch": 3.5,
|
| 1492 |
+
"learning_rate": 2.5242718446601944e-05,
|
| 1493 |
+
"loss": 0.0541,
|
| 1494 |
"step": 2160
|
| 1495 |
},
|
| 1496 |
{
|
| 1497 |
+
"epoch": 3.51,
|
| 1498 |
+
"learning_rate": 2.4433656957928804e-05,
|
| 1499 |
+
"loss": 0.0447,
|
| 1500 |
"step": 2170
|
| 1501 |
},
|
| 1502 |
{
|
| 1503 |
+
"epoch": 3.53,
|
| 1504 |
+
"learning_rate": 2.3624595469255664e-05,
|
| 1505 |
+
"loss": 0.0621,
|
| 1506 |
"step": 2180
|
| 1507 |
},
|
| 1508 |
{
|
| 1509 |
+
"epoch": 3.54,
|
| 1510 |
+
"learning_rate": 2.2815533980582527e-05,
|
| 1511 |
+
"loss": 0.0577,
|
| 1512 |
"step": 2190
|
| 1513 |
},
|
| 1514 |
{
|
| 1515 |
+
"epoch": 3.56,
|
| 1516 |
+
"learning_rate": 2.2006472491909387e-05,
|
| 1517 |
+
"loss": 0.0519,
|
| 1518 |
"step": 2200
|
| 1519 |
},
|
| 1520 |
{
|
| 1521 |
+
"epoch": 3.56,
|
| 1522 |
+
"eval_accuracy": 0.9066000185133759,
|
| 1523 |
+
"eval_loss": 0.298173189163208,
|
| 1524 |
+
"eval_runtime": 170.0796,
|
| 1525 |
+
"eval_samples_per_second": 127.035,
|
| 1526 |
+
"eval_steps_per_second": 15.881,
|
| 1527 |
"step": 2200
|
| 1528 |
},
|
| 1529 |
{
|
| 1530 |
+
"epoch": 3.58,
|
| 1531 |
+
"learning_rate": 2.1197411003236247e-05,
|
| 1532 |
+
"loss": 0.057,
|
| 1533 |
"step": 2210
|
| 1534 |
},
|
| 1535 |
{
|
| 1536 |
+
"epoch": 3.59,
|
| 1537 |
+
"learning_rate": 2.0388349514563107e-05,
|
| 1538 |
+
"loss": 0.0521,
|
| 1539 |
"step": 2220
|
| 1540 |
},
|
| 1541 |
{
|
| 1542 |
+
"epoch": 3.61,
|
| 1543 |
+
"learning_rate": 1.957928802588997e-05,
|
| 1544 |
+
"loss": 0.0525,
|
| 1545 |
"step": 2230
|
| 1546 |
},
|
| 1547 |
{
|
| 1548 |
+
"epoch": 3.62,
|
| 1549 |
+
"learning_rate": 1.877022653721683e-05,
|
| 1550 |
+
"loss": 0.0499,
|
| 1551 |
"step": 2240
|
| 1552 |
},
|
| 1553 |
{
|
| 1554 |
+
"epoch": 3.64,
|
| 1555 |
+
"learning_rate": 1.796116504854369e-05,
|
| 1556 |
+
"loss": 0.0566,
|
| 1557 |
"step": 2250
|
| 1558 |
},
|
| 1559 |
{
|
| 1560 |
+
"epoch": 3.66,
|
| 1561 |
+
"learning_rate": 1.715210355987055e-05,
|
| 1562 |
+
"loss": 0.0569,
|
| 1563 |
"step": 2260
|
| 1564 |
},
|
| 1565 |
{
|
| 1566 |
+
"epoch": 3.67,
|
| 1567 |
+
"learning_rate": 1.6343042071197413e-05,
|
| 1568 |
+
"loss": 0.0453,
|
| 1569 |
"step": 2270
|
| 1570 |
},
|
| 1571 |
{
|
| 1572 |
+
"epoch": 3.69,
|
| 1573 |
+
"learning_rate": 1.5533980582524273e-05,
|
| 1574 |
+
"loss": 0.0509,
|
| 1575 |
"step": 2280
|
| 1576 |
},
|
| 1577 |
{
|
| 1578 |
+
"epoch": 3.71,
|
| 1579 |
+
"learning_rate": 1.4724919093851133e-05,
|
| 1580 |
+
"loss": 0.0482,
|
| 1581 |
"step": 2290
|
| 1582 |
},
|
| 1583 |
{
|
| 1584 |
+
"epoch": 3.72,
|
| 1585 |
+
"learning_rate": 1.3915857605177996e-05,
|
| 1586 |
+
"loss": 0.0527,
|
| 1587 |
"step": 2300
|
| 1588 |
},
|
| 1589 |
{
|
| 1590 |
+
"epoch": 3.72,
|
| 1591 |
+
"eval_accuracy": 0.9072479866703693,
|
| 1592 |
+
"eval_loss": 0.29742419719696045,
|
| 1593 |
+
"eval_runtime": 156.9396,
|
| 1594 |
+
"eval_samples_per_second": 137.671,
|
| 1595 |
+
"eval_steps_per_second": 17.21,
|
| 1596 |
"step": 2300
|
| 1597 |
},
|
| 1598 |
{
|
| 1599 |
+
"epoch": 3.74,
|
| 1600 |
+
"learning_rate": 1.3106796116504854e-05,
|
| 1601 |
+
"loss": 0.0513,
|
| 1602 |
"step": 2310
|
| 1603 |
},
|
| 1604 |
{
|
| 1605 |
+
"epoch": 3.75,
|
| 1606 |
+
"learning_rate": 1.2297734627831716e-05,
|
| 1607 |
+
"loss": 0.0619,
|
| 1608 |
"step": 2320
|
| 1609 |
},
|
| 1610 |
{
|
| 1611 |
+
"epoch": 3.77,
|
| 1612 |
+
"learning_rate": 1.1488673139158575e-05,
|
| 1613 |
+
"loss": 0.0509,
|
| 1614 |
"step": 2330
|
| 1615 |
},
|
| 1616 |
{
|
| 1617 |
+
"epoch": 3.79,
|
| 1618 |
+
"learning_rate": 1.0679611650485437e-05,
|
| 1619 |
+
"loss": 0.0491,
|
| 1620 |
"step": 2340
|
| 1621 |
},
|
| 1622 |
{
|
| 1623 |
+
"epoch": 3.8,
|
| 1624 |
+
"learning_rate": 9.870550161812297e-06,
|
| 1625 |
+
"loss": 0.0511,
|
| 1626 |
"step": 2350
|
| 1627 |
},
|
| 1628 |
{
|
| 1629 |
+
"epoch": 3.82,
|
| 1630 |
+
"learning_rate": 9.06148867313916e-06,
|
| 1631 |
+
"loss": 0.0475,
|
| 1632 |
"step": 2360
|
| 1633 |
},
|
| 1634 |
{
|
| 1635 |
+
"epoch": 3.83,
|
| 1636 |
+
"learning_rate": 8.25242718446602e-06,
|
| 1637 |
+
"loss": 0.0542,
|
| 1638 |
"step": 2370
|
| 1639 |
},
|
| 1640 |
{
|
| 1641 |
+
"epoch": 3.85,
|
| 1642 |
+
"learning_rate": 7.443365695792881e-06,
|
| 1643 |
+
"loss": 0.047,
|
| 1644 |
"step": 2380
|
| 1645 |
},
|
| 1646 |
{
|
| 1647 |
+
"epoch": 3.87,
|
| 1648 |
+
"learning_rate": 6.6343042071197415e-06,
|
| 1649 |
+
"loss": 0.0461,
|
| 1650 |
"step": 2390
|
| 1651 |
},
|
| 1652 |
{
|
| 1653 |
+
"epoch": 3.88,
|
| 1654 |
+
"learning_rate": 5.825242718446602e-06,
|
| 1655 |
+
"loss": 0.0561,
|
| 1656 |
"step": 2400
|
| 1657 |
},
|
| 1658 |
{
|
| 1659 |
+
"epoch": 3.88,
|
| 1660 |
+
"eval_accuracy": 0.9091456077015644,
|
| 1661 |
+
"eval_loss": 0.2911698520183563,
|
| 1662 |
+
"eval_runtime": 166.2697,
|
| 1663 |
+
"eval_samples_per_second": 129.945,
|
| 1664 |
+
"eval_steps_per_second": 16.245,
|
| 1665 |
"step": 2400
|
| 1666 |
},
|
| 1667 |
{
|
| 1668 |
+
"epoch": 3.9,
|
| 1669 |
+
"learning_rate": 5.016181229773463e-06,
|
| 1670 |
+
"loss": 0.0572,
|
| 1671 |
"step": 2410
|
| 1672 |
},
|
| 1673 |
{
|
| 1674 |
+
"epoch": 3.92,
|
| 1675 |
+
"learning_rate": 4.207119741100324e-06,
|
| 1676 |
+
"loss": 0.0374,
|
| 1677 |
"step": 2420
|
| 1678 |
},
|
| 1679 |
{
|
| 1680 |
+
"epoch": 3.93,
|
| 1681 |
+
"learning_rate": 3.3980582524271844e-06,
|
| 1682 |
+
"loss": 0.0472,
|
| 1683 |
"step": 2430
|
| 1684 |
},
|
| 1685 |
{
|
| 1686 |
+
"epoch": 3.95,
|
| 1687 |
+
"learning_rate": 2.5889967637540456e-06,
|
| 1688 |
+
"loss": 0.0384,
|
| 1689 |
"step": 2440
|
| 1690 |
},
|
| 1691 |
{
|
| 1692 |
+
"epoch": 3.96,
|
| 1693 |
+
"learning_rate": 1.7799352750809063e-06,
|
| 1694 |
+
"loss": 0.0439,
|
| 1695 |
"step": 2450
|
| 1696 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1697 |
{
|
| 1698 |
"epoch": 3.98,
|
| 1699 |
+
"learning_rate": 9.70873786407767e-07,
|
| 1700 |
+
"loss": 0.0432,
|
| 1701 |
+
"step": 2460
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1702 |
},
|
| 1703 |
{
|
| 1704 |
"epoch": 4.0,
|
| 1705 |
+
"learning_rate": 1.6181229773462785e-07,
|
| 1706 |
+
"loss": 0.0533,
|
| 1707 |
+
"step": 2470
|
|
|
|
|
|
|
|
|
|
| 1708 |
},
|
| 1709 |
{
|
| 1710 |
"epoch": 4.0,
|
| 1711 |
+
"step": 2472,
|
| 1712 |
+
"total_flos": 4.902492055502632e+19,
|
| 1713 |
+
"train_loss": 0.2963483939957006,
|
| 1714 |
+
"train_runtime": 9311.4957,
|
| 1715 |
+
"train_samples_per_second": 67.941,
|
| 1716 |
+
"train_steps_per_second": 0.265
|
| 1717 |
}
|
| 1718 |
],
|
| 1719 |
"logging_steps": 10,
|
| 1720 |
+
"max_steps": 2472,
|
| 1721 |
"num_train_epochs": 4,
|
| 1722 |
"save_steps": 100,
|
| 1723 |
+
"total_flos": 4.902492055502632e+19,
|
| 1724 |
"trial_name": null,
|
| 1725 |
"trial_params": null
|
| 1726 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4027
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95191dbaeaed65d25b211b1e6f98c929203ec44d65736eda326a894d9c7bfbe5
|
| 3 |
size 4027
|