End of training
Browse files- README.md +38 -58
- pytorch_model.bin +1 -1
- trainer_state.json +1095 -1001
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -13,11 +13,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# text-normalization-ru-new
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [cointegrated/rut5-small](https://huggingface.co/cointegrated/rut5-small) on
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
-
- Loss: 0.
|
| 19 |
- Mean Distance: 0
|
| 20 |
-
- Max Distance:
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
@@ -37,68 +37,48 @@ More information needed
|
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
- learning_rate: 0.001
|
| 40 |
-
- train_batch_size:
|
| 41 |
-
- eval_batch_size:
|
| 42 |
- seed: 42
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
- lr_scheduler_warmup_ratio: 0.1
|
| 46 |
-
- num_epochs:
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
-
| Training Loss | Epoch | Step
|
| 51 |
-
|
| 52 |
-
| 0.
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 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 |
-
| 0.0026 | 31.0 | 358081 | 0.0596 | 0 | 15 |
|
| 83 |
-
| 0.0026 | 32.0 | 369632 | 0.0617 | 0 | 30 |
|
| 84 |
-
| 0.0024 | 33.0 | 381183 | 0.0572 | 0 | 20 |
|
| 85 |
-
| 0.0022 | 34.0 | 392734 | 0.0557 | 0 | 19 |
|
| 86 |
-
| 0.002 | 35.0 | 404285 | 0.0612 | 0 | 17 |
|
| 87 |
-
| 0.0017 | 36.0 | 415836 | 0.0576 | 0 | 14 |
|
| 88 |
-
| 0.0016 | 37.0 | 427387 | 0.0606 | 0 | 16 |
|
| 89 |
-
| 0.0015 | 38.0 | 438938 | 0.0570 | 0 | 24 |
|
| 90 |
-
| 0.0013 | 39.0 | 450489 | 0.0594 | 0 | 24 |
|
| 91 |
-
| 0.0012 | 40.0 | 462040 | 0.0603 | 0 | 24 |
|
| 92 |
-
| 0.001 | 41.0 | 473591 | 0.0642 | 0 | 20 |
|
| 93 |
-
| 0.0009 | 42.0 | 485142 | 0.0638 | 0 | 29 |
|
| 94 |
-
| 0.0008 | 43.0 | 496693 | 0.0659 | 0 | 29 |
|
| 95 |
-
| 0.0007 | 44.0 | 508244 | 0.0672 | 0 | 29 |
|
| 96 |
-
| 0.0007 | 45.0 | 519795 | 0.0662 | 0 | 19 |
|
| 97 |
-
| 0.0006 | 46.0 | 531346 | 0.0666 | 0 | 24 |
|
| 98 |
-
| 0.0006 | 47.0 | 542897 | 0.0690 | 0 | 22 |
|
| 99 |
-
| 0.0005 | 48.0 | 554448 | 0.0670 | 0 | 15 |
|
| 100 |
-
| 0.0005 | 49.0 | 565999 | 0.0662 | 0 | 18 |
|
| 101 |
-
| 0.0004 | 50.0 | 577550 | 0.0664 | 0 | 18 |
|
| 102 |
|
| 103 |
|
| 104 |
### Framework versions
|
|
|
|
| 13 |
|
| 14 |
# text-normalization-ru-new
|
| 15 |
|
| 16 |
+
This model is a fine-tuned version of [cointegrated/rut5-small](https://huggingface.co/cointegrated/rut5-small) on an unknown dataset.
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
+
- Loss: 0.0985
|
| 19 |
- Mean Distance: 0
|
| 20 |
+
- Max Distance: 9
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
- learning_rate: 0.001
|
| 40 |
+
- train_batch_size: 30
|
| 41 |
+
- eval_batch_size: 30
|
| 42 |
- seed: 42
|
| 43 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
- lr_scheduler_type: linear
|
| 45 |
- lr_scheduler_warmup_ratio: 0.1
|
| 46 |
+
- num_epochs: 30
|
| 47 |
|
| 48 |
### Training results
|
| 49 |
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss | Mean Distance | Max Distance |
|
| 51 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------------:|:------------:|
|
| 52 |
+
| 0.181 | 1.0 | 2598 | 0.2140 | 4 | 36 |
|
| 53 |
+
| 0.1067 | 2.0 | 5196 | 0.1416 | 2 | 29 |
|
| 54 |
+
| 0.0801 | 3.0 | 7794 | 0.1098 | 2 | 22 |
|
| 55 |
+
| 0.0575 | 4.0 | 10392 | 0.1081 | 2 | 18 |
|
| 56 |
+
| 0.0452 | 5.0 | 12990 | 0.0897 | 1 | 14 |
|
| 57 |
+
| 0.0372 | 6.0 | 15588 | 0.0720 | 1 | 15 |
|
| 58 |
+
| 0.0323 | 7.0 | 18186 | 0.0840 | 1 | 12 |
|
| 59 |
+
| 0.0267 | 8.0 | 20784 | 0.0768 | 1 | 16 |
|
| 60 |
+
| 0.0231 | 9.0 | 23382 | 0.0697 | 1 | 10 |
|
| 61 |
+
| 0.0199 | 10.0 | 25980 | 0.0717 | 1 | 9 |
|
| 62 |
+
| 0.0168 | 11.0 | 28578 | 0.0812 | 1 | 16 |
|
| 63 |
+
| 0.0148 | 12.0 | 31176 | 0.0961 | 1 | 12 |
|
| 64 |
+
| 0.0128 | 13.0 | 33774 | 0.0823 | 1 | 9 |
|
| 65 |
+
| 0.0112 | 14.0 | 36372 | 0.0766 | 1 | 12 |
|
| 66 |
+
| 0.0093 | 15.0 | 38970 | 0.0713 | 1 | 9 |
|
| 67 |
+
| 0.0083 | 16.0 | 41568 | 0.0847 | 1 | 14 |
|
| 68 |
+
| 0.0076 | 17.0 | 44166 | 0.0863 | 1 | 11 |
|
| 69 |
+
| 0.0064 | 18.0 | 46764 | 0.0830 | 1 | 14 |
|
| 70 |
+
| 0.0054 | 19.0 | 49362 | 0.0884 | 1 | 11 |
|
| 71 |
+
| 0.0052 | 20.0 | 51960 | 0.0821 | 1 | 10 |
|
| 72 |
+
| 0.0045 | 21.0 | 54558 | 0.0915 | 1 | 14 |
|
| 73 |
+
| 0.0037 | 22.0 | 57156 | 0.0931 | 1 | 14 |
|
| 74 |
+
| 0.0036 | 23.0 | 59754 | 0.0941 | 1 | 9 |
|
| 75 |
+
| 0.0028 | 24.0 | 62352 | 0.0861 | 1 | 13 |
|
| 76 |
+
| 0.0026 | 25.0 | 64950 | 0.0912 | 1 | 12 |
|
| 77 |
+
| 0.0024 | 26.0 | 67548 | 0.0916 | 0 | 9 |
|
| 78 |
+
| 0.002 | 27.0 | 70146 | 0.0888 | 0 | 9 |
|
| 79 |
+
| 0.0017 | 28.0 | 72744 | 0.0888 | 0 | 9 |
|
| 80 |
+
| 0.0017 | 29.0 | 75342 | 0.0952 | 0 | 9 |
|
| 81 |
+
| 0.0014 | 30.0 | 77940 | 0.0985 | 0 | 9 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 82 |
|
| 83 |
|
| 84 |
### Framework versions
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 258643461
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:18151e9614a6f5186bbe2400883a83025ffebc2e1a09f7899f724917c104504e
|
| 3 |
size 258643461
|
trainer_state.json
CHANGED
|
@@ -1,1434 +1,1528 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 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": 0.0,
|
| 13 |
-
"learning_rate": 1.
|
| 14 |
-
"loss": 13.
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 5.
|
| 20 |
-
"loss":
|
| 21 |
-
"step":
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch": 0.
|
| 25 |
-
"learning_rate": 0.
|
| 26 |
-
"loss": 0.
|
| 27 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.75,
|
| 31 |
-
"learning_rate": 0.
|
| 32 |
-
"loss": 0.
|
| 33 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 1.0,
|
| 37 |
-
"eval_loss": 0.
|
| 38 |
-
"eval_max_distance":
|
| 39 |
-
"eval_mean_distance":
|
| 40 |
-
"eval_runtime":
|
| 41 |
-
"eval_samples_per_second":
|
| 42 |
-
"eval_steps_per_second":
|
| 43 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 44 |
},
|
| 45 |
{
|
| 46 |
-
"epoch": 1.
|
| 47 |
-
"learning_rate": 0.
|
| 48 |
-
"loss":
|
| 49 |
-
"step":
|
| 50 |
},
|
| 51 |
{
|
| 52 |
-
"epoch": 1.
|
| 53 |
-
"learning_rate": 0.
|
| 54 |
-
"loss": 0.
|
| 55 |
-
"step":
|
| 56 |
},
|
| 57 |
{
|
| 58 |
-
"epoch": 1.
|
| 59 |
-
"learning_rate": 0.
|
| 60 |
-
"loss": 0.
|
| 61 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
},
|
| 63 |
{
|
| 64 |
"epoch": 2.0,
|
| 65 |
-
"eval_loss": 0.
|
| 66 |
-
"eval_max_distance":
|
| 67 |
-
"eval_mean_distance":
|
| 68 |
-
"eval_runtime":
|
| 69 |
-
"eval_samples_per_second":
|
| 70 |
-
"eval_steps_per_second":
|
| 71 |
-
"step":
|
| 72 |
},
|
| 73 |
{
|
| 74 |
-
"epoch": 2.
|
| 75 |
-
"learning_rate": 0.
|
| 76 |
-
"loss": 0.
|
| 77 |
-
"step":
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"epoch": 2.
|
| 81 |
-
"learning_rate": 0.
|
| 82 |
-
"loss": 0.
|
| 83 |
-
"step":
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch": 2.
|
| 87 |
-
"learning_rate": 0.
|
| 88 |
-
"loss": 0.
|
| 89 |
-
"step":
|
| 90 |
},
|
| 91 |
{
|
| 92 |
-
"epoch":
|
| 93 |
-
"
|
| 94 |
-
"
|
| 95 |
-
"
|
| 96 |
-
"eval_runtime": 8.5861,
|
| 97 |
-
"eval_samples_per_second": 58.233,
|
| 98 |
-
"eval_steps_per_second": 2.912,
|
| 99 |
-
"step": 34653
|
| 100 |
},
|
| 101 |
{
|
| 102 |
-
"epoch":
|
| 103 |
-
"learning_rate": 0.
|
| 104 |
-
"loss": 0.
|
| 105 |
-
"step":
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"epoch":
|
| 109 |
-
"learning_rate": 0.
|
| 110 |
-
"loss": 0.
|
| 111 |
-
"step":
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch": 3.
|
| 115 |
-
"
|
| 116 |
-
"
|
| 117 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch":
|
| 121 |
-
"
|
| 122 |
-
"
|
| 123 |
-
"
|
| 124 |
-
"eval_runtime": 10.1617,
|
| 125 |
-
"eval_samples_per_second": 49.204,
|
| 126 |
-
"eval_steps_per_second": 2.46,
|
| 127 |
-
"step": 46204
|
| 128 |
},
|
| 129 |
{
|
| 130 |
-
"epoch":
|
| 131 |
-
"learning_rate": 0.
|
| 132 |
-
"loss": 0.
|
| 133 |
-
"step":
|
| 134 |
},
|
| 135 |
{
|
| 136 |
-
"epoch":
|
| 137 |
-
"learning_rate": 0.
|
| 138 |
-
"loss": 0.
|
| 139 |
-
"step":
|
| 140 |
},
|
| 141 |
{
|
| 142 |
-
"epoch":
|
| 143 |
-
"learning_rate": 0.
|
| 144 |
-
"loss": 0.
|
| 145 |
-
"step":
|
| 146 |
},
|
| 147 |
{
|
| 148 |
-
"epoch":
|
| 149 |
-
"
|
| 150 |
-
"
|
| 151 |
-
"
|
| 152 |
-
"eval_runtime": 8.5226,
|
| 153 |
-
"eval_samples_per_second": 58.667,
|
| 154 |
-
"eval_steps_per_second": 2.933,
|
| 155 |
-
"step": 57755
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch":
|
| 159 |
-
"learning_rate": 0.
|
| 160 |
-
"loss": 0.
|
| 161 |
-
"step":
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch":
|
| 165 |
-
"learning_rate": 0.
|
| 166 |
-
"loss": 0.
|
| 167 |
-
"step":
|
| 168 |
},
|
| 169 |
{
|
| 170 |
-
"epoch":
|
| 171 |
-
"
|
| 172 |
-
"
|
| 173 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch":
|
| 177 |
-
"
|
| 178 |
-
"
|
| 179 |
-
"
|
| 180 |
-
"eval_runtime": 8.627,
|
| 181 |
-
"eval_samples_per_second": 57.958,
|
| 182 |
-
"eval_steps_per_second": 2.898,
|
| 183 |
-
"step": 69306
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"learning_rate": 0.
|
| 188 |
-
"loss": 0.
|
| 189 |
-
"step":
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"learning_rate": 0.
|
| 194 |
-
"loss": 0.
|
| 195 |
-
"step":
|
| 196 |
},
|
| 197 |
{
|
| 198 |
-
"epoch":
|
| 199 |
-
"learning_rate": 0.
|
| 200 |
-
"loss": 0.
|
| 201 |
-
"step":
|
| 202 |
},
|
| 203 |
{
|
| 204 |
-
"epoch":
|
| 205 |
-
"
|
| 206 |
-
"
|
| 207 |
-
"
|
| 208 |
-
"eval_runtime": 8.3481,
|
| 209 |
-
"eval_samples_per_second": 59.894,
|
| 210 |
-
"eval_steps_per_second": 2.995,
|
| 211 |
-
"step": 80857
|
| 212 |
},
|
| 213 |
{
|
| 214 |
-
"epoch":
|
| 215 |
-
"learning_rate": 0.
|
| 216 |
-
"loss": 0.
|
| 217 |
-
"step":
|
| 218 |
},
|
| 219 |
{
|
| 220 |
-
"epoch":
|
| 221 |
-
"learning_rate": 0.
|
| 222 |
-
"loss": 0.
|
| 223 |
-
"step":
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"epoch":
|
| 227 |
-
"
|
| 228 |
-
"
|
| 229 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 230 |
},
|
| 231 |
{
|
| 232 |
-
"epoch":
|
| 233 |
-
"
|
| 234 |
-
"
|
| 235 |
-
"
|
| 236 |
-
"eval_runtime": 8.2389,
|
| 237 |
-
"eval_samples_per_second": 60.688,
|
| 238 |
-
"eval_steps_per_second": 3.034,
|
| 239 |
-
"step": 92408
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch":
|
| 243 |
-
"learning_rate": 0.
|
| 244 |
"loss": 0.0397,
|
| 245 |
-
"step":
|
| 246 |
},
|
| 247 |
{
|
| 248 |
-
"epoch":
|
| 249 |
-
"learning_rate": 0.
|
| 250 |
-
"loss": 0.
|
| 251 |
-
"step":
|
| 252 |
},
|
| 253 |
{
|
| 254 |
-
"epoch":
|
| 255 |
-
"learning_rate": 0.
|
| 256 |
-
"loss": 0.
|
| 257 |
-
"step":
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"epoch":
|
| 261 |
-
"
|
| 262 |
-
"
|
| 263 |
-
"
|
| 264 |
-
"eval_runtime": 7.3095,
|
| 265 |
-
"eval_samples_per_second": 68.405,
|
| 266 |
-
"eval_steps_per_second": 3.42,
|
| 267 |
-
"step": 103959
|
| 268 |
},
|
| 269 |
{
|
| 270 |
-
"epoch":
|
| 271 |
-
"learning_rate": 0.
|
| 272 |
-
"loss": 0.
|
| 273 |
-
"step":
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"epoch":
|
| 277 |
-
"
|
| 278 |
-
"
|
| 279 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 280 |
},
|
| 281 |
{
|
| 282 |
-
"epoch":
|
| 283 |
-
"learning_rate": 0.
|
| 284 |
-
"loss": 0.
|
| 285 |
-
"step":
|
| 286 |
},
|
| 287 |
{
|
| 288 |
-
"epoch":
|
| 289 |
-
"
|
| 290 |
-
"
|
| 291 |
-
"
|
| 292 |
-
"eval_runtime": 7.6397,
|
| 293 |
-
"eval_samples_per_second": 65.447,
|
| 294 |
-
"eval_steps_per_second": 3.272,
|
| 295 |
-
"step": 115510
|
| 296 |
},
|
| 297 |
{
|
| 298 |
-
"epoch":
|
| 299 |
-
"learning_rate": 0.
|
| 300 |
-
"loss": 0.
|
| 301 |
-
"step":
|
| 302 |
},
|
| 303 |
{
|
| 304 |
-
"epoch":
|
| 305 |
-
"learning_rate": 0.
|
| 306 |
-
"loss": 0.
|
| 307 |
-
"step":
|
| 308 |
},
|
| 309 |
{
|
| 310 |
-
"epoch":
|
| 311 |
-
"learning_rate": 0.
|
| 312 |
-
"loss": 0.
|
| 313 |
-
"step":
|
| 314 |
},
|
| 315 |
{
|
| 316 |
-
"epoch":
|
| 317 |
-
"
|
| 318 |
-
"
|
| 319 |
-
"
|
| 320 |
-
"eval_runtime": 7.8584,
|
| 321 |
-
"eval_samples_per_second": 63.626,
|
| 322 |
-
"eval_steps_per_second": 3.181,
|
| 323 |
-
"step": 127061
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch":
|
| 327 |
-
"learning_rate": 0.
|
| 328 |
-
"loss": 0.
|
| 329 |
-
"step":
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch":
|
| 333 |
-
"
|
| 334 |
-
"
|
| 335 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch":
|
| 339 |
-
"learning_rate": 0.
|
| 340 |
-
"loss": 0.
|
| 341 |
-
"step":
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch":
|
| 345 |
-
"
|
| 346 |
-
"
|
| 347 |
-
"
|
| 348 |
-
"eval_runtime": 7.7916,
|
| 349 |
-
"eval_samples_per_second": 64.172,
|
| 350 |
-
"eval_steps_per_second": 3.209,
|
| 351 |
-
"step": 138612
|
| 352 |
},
|
| 353 |
{
|
| 354 |
-
"epoch":
|
| 355 |
-
"learning_rate": 0.
|
| 356 |
-
"loss": 0.
|
| 357 |
-
"step":
|
| 358 |
},
|
| 359 |
{
|
| 360 |
-
"epoch":
|
| 361 |
-
"learning_rate": 0.
|
| 362 |
-
"loss": 0.
|
| 363 |
-
"step":
|
| 364 |
},
|
| 365 |
{
|
| 366 |
-
"epoch":
|
| 367 |
-
"learning_rate": 0.
|
| 368 |
-
"loss": 0.
|
| 369 |
-
"step":
|
| 370 |
},
|
| 371 |
{
|
| 372 |
-
"epoch":
|
| 373 |
-
"
|
| 374 |
-
"
|
| 375 |
-
"
|
| 376 |
-
"eval_runtime": 7.0966,
|
| 377 |
-
"eval_samples_per_second": 70.457,
|
| 378 |
-
"eval_steps_per_second": 3.523,
|
| 379 |
-
"step": 150163
|
| 380 |
},
|
| 381 |
{
|
| 382 |
-
"epoch":
|
| 383 |
-
"learning_rate": 0.
|
| 384 |
-
"loss": 0.
|
| 385 |
-
"step":
|
| 386 |
},
|
| 387 |
{
|
| 388 |
-
"epoch":
|
| 389 |
-
"
|
| 390 |
-
"
|
| 391 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 392 |
},
|
| 393 |
{
|
| 394 |
-
"epoch":
|
| 395 |
-
"learning_rate": 0.
|
| 396 |
-
"loss": 0.
|
| 397 |
-
"step":
|
| 398 |
},
|
| 399 |
{
|
| 400 |
-
"epoch":
|
| 401 |
-
"
|
| 402 |
-
"
|
| 403 |
-
"
|
| 404 |
-
"eval_runtime": 7.3512,
|
| 405 |
-
"eval_samples_per_second": 68.016,
|
| 406 |
-
"eval_steps_per_second": 3.401,
|
| 407 |
-
"step": 161714
|
| 408 |
},
|
| 409 |
{
|
| 410 |
-
"epoch":
|
| 411 |
-
"learning_rate": 0.
|
| 412 |
-
"loss": 0.
|
| 413 |
-
"step":
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch":
|
| 417 |
-
"learning_rate": 0.
|
| 418 |
-
"loss": 0.
|
| 419 |
-
"step":
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"learning_rate": 0.
|
| 424 |
-
"loss": 0.
|
| 425 |
-
"step":
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch":
|
| 429 |
-
"
|
| 430 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 431 |
"eval_mean_distance": 1,
|
| 432 |
-
"eval_runtime":
|
| 433 |
-
"eval_samples_per_second":
|
| 434 |
-
"eval_steps_per_second":
|
| 435 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 436 |
},
|
| 437 |
{
|
| 438 |
-
"epoch":
|
| 439 |
-
"learning_rate": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 440 |
"loss": 0.019,
|
| 441 |
-
"step":
|
| 442 |
},
|
| 443 |
{
|
| 444 |
-
"epoch":
|
| 445 |
-
"learning_rate": 0.
|
| 446 |
-
"loss": 0.
|
| 447 |
-
"step":
|
| 448 |
},
|
| 449 |
{
|
| 450 |
-
"epoch":
|
| 451 |
-
"learning_rate": 0.
|
| 452 |
-
"loss": 0.
|
| 453 |
-
"step":
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"epoch":
|
| 457 |
-
"eval_loss": 0.
|
| 458 |
-
"eval_max_distance":
|
| 459 |
"eval_mean_distance": 1,
|
| 460 |
-
"eval_runtime":
|
| 461 |
-
"eval_samples_per_second":
|
| 462 |
-
"eval_steps_per_second":
|
| 463 |
-
"step":
|
| 464 |
},
|
| 465 |
{
|
| 466 |
-
"epoch":
|
| 467 |
-
"learning_rate": 0.
|
| 468 |
-
"loss": 0.
|
| 469 |
-
"step":
|
| 470 |
},
|
| 471 |
{
|
| 472 |
-
"epoch":
|
| 473 |
-
"learning_rate": 0.
|
| 474 |
-
"loss": 0.
|
| 475 |
-
"step":
|
| 476 |
},
|
| 477 |
{
|
| 478 |
-
"epoch":
|
| 479 |
-
"learning_rate": 0.
|
| 480 |
-
"loss": 0.
|
| 481 |
-
"step":
|
| 482 |
},
|
| 483 |
{
|
| 484 |
-
"epoch":
|
| 485 |
-
"
|
| 486 |
-
"
|
| 487 |
-
"
|
| 488 |
-
"eval_runtime": 7.1541,
|
| 489 |
-
"eval_samples_per_second": 69.89,
|
| 490 |
-
"eval_steps_per_second": 3.494,
|
| 491 |
-
"step": 196367
|
| 492 |
},
|
| 493 |
{
|
| 494 |
-
"epoch":
|
| 495 |
-
"learning_rate": 0.
|
| 496 |
-
"loss": 0.
|
| 497 |
-
"step":
|
| 498 |
},
|
| 499 |
{
|
| 500 |
-
"epoch":
|
| 501 |
-
"learning_rate": 0.
|
| 502 |
-
"loss": 0.
|
| 503 |
-
"step":
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"epoch":
|
| 507 |
-
"learning_rate": 0.
|
| 508 |
-
"loss": 0.
|
| 509 |
-
"step":
|
| 510 |
},
|
| 511 |
{
|
| 512 |
-
"epoch":
|
| 513 |
-
"eval_loss": 0.
|
| 514 |
-
"eval_max_distance":
|
| 515 |
-
"eval_mean_distance":
|
| 516 |
-
"eval_runtime":
|
| 517 |
-
"eval_samples_per_second":
|
| 518 |
-
"eval_steps_per_second":
|
| 519 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 520 |
},
|
| 521 |
{
|
| 522 |
-
"epoch":
|
| 523 |
-
"learning_rate": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 524 |
"loss": 0.0144,
|
| 525 |
-
"step":
|
| 526 |
},
|
| 527 |
{
|
| 528 |
-
"epoch":
|
| 529 |
-
"learning_rate": 0.
|
| 530 |
-
"loss": 0.
|
| 531 |
-
"step":
|
| 532 |
},
|
| 533 |
{
|
| 534 |
-
"epoch":
|
| 535 |
-
"learning_rate": 0.
|
| 536 |
-
"loss": 0.
|
| 537 |
-
"step":
|
| 538 |
},
|
| 539 |
{
|
| 540 |
-
"epoch":
|
| 541 |
-
"eval_loss": 0.
|
| 542 |
-
"eval_max_distance":
|
| 543 |
-
"eval_mean_distance":
|
| 544 |
-
"eval_runtime":
|
| 545 |
-
"eval_samples_per_second":
|
| 546 |
-
"eval_steps_per_second":
|
| 547 |
-
"step":
|
| 548 |
},
|
| 549 |
{
|
| 550 |
-
"epoch":
|
| 551 |
-
"learning_rate": 0.
|
| 552 |
-
"loss": 0.
|
| 553 |
-
"step":
|
| 554 |
},
|
| 555 |
{
|
| 556 |
-
"epoch":
|
| 557 |
-
"learning_rate": 0.
|
| 558 |
-
"loss": 0.
|
| 559 |
-
"step":
|
| 560 |
},
|
| 561 |
{
|
| 562 |
-
"epoch":
|
| 563 |
-
"learning_rate": 0.
|
| 564 |
-
"loss": 0.
|
| 565 |
-
"step":
|
| 566 |
},
|
| 567 |
{
|
| 568 |
-
"epoch":
|
| 569 |
-
"
|
| 570 |
-
"
|
| 571 |
-
"
|
| 572 |
-
"eval_runtime": 8.9369,
|
| 573 |
-
"eval_samples_per_second": 55.948,
|
| 574 |
-
"eval_steps_per_second": 2.797,
|
| 575 |
-
"step": 231020
|
| 576 |
},
|
| 577 |
{
|
| 578 |
-
"epoch":
|
| 579 |
-
"learning_rate": 0.
|
| 580 |
-
"loss": 0.
|
| 581 |
-
"step":
|
| 582 |
},
|
| 583 |
{
|
| 584 |
-
"epoch":
|
| 585 |
-
"learning_rate": 0.
|
| 586 |
-
"loss": 0.
|
| 587 |
-
"step":
|
| 588 |
},
|
| 589 |
{
|
| 590 |
-
"epoch":
|
| 591 |
-
"learning_rate": 0.
|
| 592 |
-
"loss": 0.
|
| 593 |
-
"step":
|
| 594 |
},
|
| 595 |
{
|
| 596 |
-
"epoch":
|
| 597 |
-
"eval_loss": 0.
|
| 598 |
-
"eval_max_distance":
|
| 599 |
-
"eval_mean_distance":
|
| 600 |
-
"eval_runtime":
|
| 601 |
-
"eval_samples_per_second":
|
| 602 |
-
"eval_steps_per_second":
|
| 603 |
-
"step":
|
| 604 |
},
|
| 605 |
{
|
| 606 |
-
"epoch":
|
| 607 |
-
"learning_rate": 0.
|
| 608 |
"loss": 0.0116,
|
| 609 |
-
"step":
|
| 610 |
},
|
| 611 |
{
|
| 612 |
-
"epoch": 21
|
| 613 |
-
"learning_rate": 0.
|
| 614 |
-
"loss": 0.
|
| 615 |
-
"step":
|
| 616 |
},
|
| 617 |
{
|
| 618 |
-
"epoch":
|
| 619 |
-
"learning_rate": 0.
|
| 620 |
-
"loss": 0.
|
| 621 |
-
"step":
|
| 622 |
},
|
| 623 |
{
|
| 624 |
-
"epoch":
|
| 625 |
-
"
|
| 626 |
-
"
|
| 627 |
-
"
|
| 628 |
-
"eval_runtime": 6.4223,
|
| 629 |
-
"eval_samples_per_second": 77.853,
|
| 630 |
-
"eval_steps_per_second": 3.893,
|
| 631 |
-
"step": 254122
|
| 632 |
},
|
| 633 |
{
|
| 634 |
-
"epoch":
|
| 635 |
-
"learning_rate": 0.
|
| 636 |
-
"loss": 0.
|
| 637 |
-
"step":
|
| 638 |
},
|
| 639 |
{
|
| 640 |
-
"epoch":
|
| 641 |
-
"learning_rate": 0.
|
| 642 |
-
"loss": 0.
|
| 643 |
-
"step":
|
| 644 |
},
|
| 645 |
{
|
| 646 |
-
"epoch":
|
| 647 |
-
"learning_rate": 0.
|
| 648 |
-
"loss": 0.
|
| 649 |
-
"step":
|
| 650 |
},
|
| 651 |
{
|
| 652 |
-
"epoch":
|
| 653 |
-
"eval_loss": 0.
|
| 654 |
-
"eval_max_distance":
|
| 655 |
-
"eval_mean_distance":
|
| 656 |
-
"eval_runtime":
|
| 657 |
-
"eval_samples_per_second":
|
| 658 |
-
"eval_steps_per_second":
|
| 659 |
-
"step":
|
| 660 |
},
|
| 661 |
{
|
| 662 |
-
"epoch":
|
| 663 |
-
"learning_rate": 0.
|
| 664 |
-
"loss": 0.
|
| 665 |
-
"step":
|
| 666 |
},
|
| 667 |
{
|
| 668 |
-
"epoch":
|
| 669 |
-
"learning_rate": 0.
|
| 670 |
-
"loss": 0.
|
| 671 |
-
"step":
|
| 672 |
},
|
| 673 |
{
|
| 674 |
-
"epoch":
|
| 675 |
-
"learning_rate": 0.
|
| 676 |
-
"loss": 0.
|
| 677 |
-
"step":
|
| 678 |
},
|
| 679 |
{
|
| 680 |
-
"epoch":
|
| 681 |
-
"
|
| 682 |
-
"
|
| 683 |
-
"
|
| 684 |
-
"eval_runtime": 6.5025,
|
| 685 |
-
"eval_samples_per_second": 76.893,
|
| 686 |
-
"eval_steps_per_second": 3.845,
|
| 687 |
-
"step": 277224
|
| 688 |
},
|
| 689 |
{
|
| 690 |
-
"epoch":
|
| 691 |
-
"learning_rate": 0.
|
| 692 |
-
"loss": 0.
|
| 693 |
-
"step":
|
| 694 |
},
|
| 695 |
{
|
| 696 |
-
"epoch":
|
| 697 |
-
"learning_rate": 0.
|
| 698 |
-
"loss": 0.
|
| 699 |
-
"step":
|
| 700 |
},
|
| 701 |
{
|
| 702 |
-
"epoch":
|
| 703 |
-
"
|
| 704 |
-
"
|
| 705 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 706 |
},
|
| 707 |
{
|
| 708 |
-
"epoch":
|
| 709 |
-
"
|
| 710 |
-
"
|
| 711 |
-
"
|
| 712 |
-
"eval_runtime": 6.3704,
|
| 713 |
-
"eval_samples_per_second": 78.489,
|
| 714 |
-
"eval_steps_per_second": 3.924,
|
| 715 |
-
"step": 288775
|
| 716 |
},
|
| 717 |
{
|
| 718 |
-
"epoch":
|
| 719 |
-
"learning_rate": 0.
|
| 720 |
-
"loss": 0.
|
| 721 |
-
"step":
|
| 722 |
},
|
| 723 |
{
|
| 724 |
-
"epoch":
|
| 725 |
-
"learning_rate": 0.
|
| 726 |
-
"loss": 0.
|
| 727 |
-
"step":
|
| 728 |
},
|
| 729 |
{
|
| 730 |
-
"epoch":
|
| 731 |
-
"learning_rate": 0.
|
| 732 |
-
"loss": 0.
|
| 733 |
-
"step":
|
| 734 |
},
|
| 735 |
{
|
| 736 |
-
"epoch":
|
| 737 |
-
"
|
| 738 |
-
"
|
| 739 |
-
"
|
| 740 |
-
"eval_runtime": 6.5774,
|
| 741 |
-
"eval_samples_per_second": 76.017,
|
| 742 |
-
"eval_steps_per_second": 3.801,
|
| 743 |
-
"step": 300326
|
| 744 |
},
|
| 745 |
{
|
| 746 |
-
"epoch":
|
| 747 |
-
"learning_rate": 0.
|
| 748 |
-
"loss": 0.
|
| 749 |
-
"step":
|
| 750 |
},
|
| 751 |
{
|
| 752 |
-
"epoch":
|
| 753 |
-
"learning_rate": 0.
|
| 754 |
-
"loss": 0.
|
| 755 |
-
"step":
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch":
|
| 759 |
-
"
|
| 760 |
-
"
|
| 761 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch":
|
| 765 |
-
"
|
| 766 |
-
"
|
| 767 |
-
"
|
| 768 |
-
|
| 769 |
-
|
| 770 |
-
"
|
| 771 |
-
"
|
|
|
|
|
|
|
| 772 |
},
|
| 773 |
{
|
| 774 |
-
"epoch":
|
| 775 |
-
"learning_rate": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 776 |
"loss": 0.0073,
|
| 777 |
-
"step":
|
| 778 |
},
|
| 779 |
{
|
| 780 |
-
"epoch":
|
| 781 |
-
"learning_rate": 0.
|
| 782 |
-
"loss": 0.
|
| 783 |
-
"step":
|
| 784 |
},
|
| 785 |
{
|
| 786 |
-
"epoch":
|
| 787 |
-
"learning_rate": 0.
|
| 788 |
-
"loss": 0.
|
| 789 |
-
"step":
|
| 790 |
},
|
| 791 |
{
|
| 792 |
-
"epoch":
|
| 793 |
-
"eval_loss": 0.
|
| 794 |
-
"eval_max_distance":
|
| 795 |
-
"eval_mean_distance":
|
| 796 |
-
"eval_runtime":
|
| 797 |
-
"eval_samples_per_second":
|
| 798 |
-
"eval_steps_per_second":
|
| 799 |
-
"step":
|
| 800 |
},
|
| 801 |
{
|
| 802 |
-
"epoch":
|
| 803 |
-
"learning_rate": 0.
|
| 804 |
-
"loss": 0.
|
| 805 |
-
"step":
|
| 806 |
},
|
| 807 |
{
|
| 808 |
-
"epoch":
|
| 809 |
-
"learning_rate": 0.
|
| 810 |
-
"loss": 0.
|
| 811 |
-
"step":
|
| 812 |
},
|
| 813 |
{
|
| 814 |
-
"epoch":
|
| 815 |
-
"learning_rate": 0.
|
| 816 |
-
"loss": 0.
|
| 817 |
-
"step":
|
| 818 |
},
|
| 819 |
{
|
| 820 |
-
"epoch":
|
| 821 |
-
"
|
| 822 |
-
"
|
| 823 |
-
"
|
| 824 |
-
"eval_runtime": 6.3104,
|
| 825 |
-
"eval_samples_per_second": 79.234,
|
| 826 |
-
"eval_steps_per_second": 3.962,
|
| 827 |
-
"step": 334979
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"epoch":
|
| 831 |
-
"learning_rate": 0.
|
| 832 |
-
"loss": 0.
|
| 833 |
-
"step":
|
| 834 |
},
|
| 835 |
{
|
| 836 |
-
"epoch":
|
| 837 |
-
"learning_rate": 0.
|
| 838 |
-
"loss": 0.
|
| 839 |
-
"step":
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"
|
| 844 |
-
"
|
| 845 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 846 |
},
|
| 847 |
{
|
| 848 |
-
"epoch":
|
| 849 |
-
"
|
| 850 |
-
"
|
| 851 |
-
"
|
| 852 |
-
"eval_runtime": 6.2956,
|
| 853 |
-
"eval_samples_per_second": 79.42,
|
| 854 |
-
"eval_steps_per_second": 3.971,
|
| 855 |
-
"step": 346530
|
| 856 |
},
|
| 857 |
{
|
| 858 |
-
"epoch":
|
| 859 |
-
"learning_rate": 0.
|
| 860 |
-
"loss": 0.
|
| 861 |
-
"step":
|
| 862 |
},
|
| 863 |
{
|
| 864 |
-
"epoch":
|
| 865 |
-
"learning_rate": 0.
|
| 866 |
-
"loss": 0.
|
| 867 |
-
"step":
|
| 868 |
},
|
| 869 |
{
|
| 870 |
-
"epoch":
|
| 871 |
-
"learning_rate": 0.
|
| 872 |
-
"loss": 0.
|
| 873 |
-
"step":
|
| 874 |
},
|
| 875 |
{
|
| 876 |
-
"epoch":
|
| 877 |
-
"
|
| 878 |
-
"
|
| 879 |
-
"
|
| 880 |
-
"eval_runtime": 6.4083,
|
| 881 |
-
"eval_samples_per_second": 78.023,
|
| 882 |
-
"eval_steps_per_second": 3.901,
|
| 883 |
-
"step": 358081
|
| 884 |
},
|
| 885 |
{
|
| 886 |
-
"epoch":
|
| 887 |
-
"learning_rate": 0.
|
| 888 |
-
"loss": 0.
|
| 889 |
-
"step":
|
| 890 |
},
|
| 891 |
{
|
| 892 |
-
"epoch":
|
| 893 |
-
"learning_rate": 0.
|
| 894 |
-
"loss": 0.
|
| 895 |
-
"step":
|
| 896 |
},
|
| 897 |
{
|
| 898 |
-
"epoch":
|
| 899 |
-
"
|
| 900 |
-
"
|
| 901 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 902 |
},
|
| 903 |
{
|
| 904 |
-
"epoch":
|
| 905 |
-
"
|
| 906 |
-
"
|
| 907 |
-
"
|
| 908 |
-
"eval_runtime": 6.2839,
|
| 909 |
-
"eval_samples_per_second": 79.568,
|
| 910 |
-
"eval_steps_per_second": 3.978,
|
| 911 |
-
"step": 369632
|
| 912 |
},
|
| 913 |
{
|
| 914 |
-
"epoch":
|
| 915 |
-
"learning_rate": 0.
|
| 916 |
"loss": 0.0047,
|
| 917 |
-
"step":
|
| 918 |
},
|
| 919 |
{
|
| 920 |
-
"epoch":
|
| 921 |
-
"learning_rate": 0.
|
| 922 |
-
"loss": 0.
|
| 923 |
-
"step":
|
| 924 |
},
|
| 925 |
{
|
| 926 |
-
"epoch":
|
| 927 |
-
"learning_rate": 0.
|
| 928 |
-
"loss": 0.
|
| 929 |
-
"step":
|
| 930 |
},
|
| 931 |
{
|
| 932 |
-
"epoch":
|
| 933 |
-
"
|
| 934 |
-
"
|
| 935 |
-
"
|
| 936 |
-
"eval_runtime": 6.221,
|
| 937 |
-
"eval_samples_per_second": 80.373,
|
| 938 |
-
"eval_steps_per_second": 4.019,
|
| 939 |
-
"step": 381183
|
| 940 |
},
|
| 941 |
{
|
| 942 |
-
"epoch":
|
| 943 |
-
"learning_rate": 0.
|
| 944 |
-
"loss": 0.
|
| 945 |
-
"step":
|
| 946 |
},
|
| 947 |
{
|
| 948 |
-
"epoch":
|
| 949 |
-
"learning_rate": 0.
|
| 950 |
-
"loss": 0.
|
| 951 |
-
"step":
|
| 952 |
},
|
| 953 |
{
|
| 954 |
-
"epoch":
|
| 955 |
-
"
|
| 956 |
-
"
|
| 957 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 958 |
},
|
| 959 |
{
|
| 960 |
-
"epoch":
|
| 961 |
-
"
|
| 962 |
-
"
|
| 963 |
-
"
|
| 964 |
-
"eval_runtime": 6.2592,
|
| 965 |
-
"eval_samples_per_second": 79.883,
|
| 966 |
-
"eval_steps_per_second": 3.994,
|
| 967 |
-
"step": 392734
|
| 968 |
},
|
| 969 |
{
|
| 970 |
-
"epoch":
|
| 971 |
-
"learning_rate": 0.
|
| 972 |
-
"loss": 0.
|
| 973 |
-
"step":
|
| 974 |
},
|
| 975 |
{
|
| 976 |
-
"epoch":
|
| 977 |
-
"learning_rate": 0.
|
| 978 |
-
"loss": 0.
|
| 979 |
-
"step":
|
| 980 |
},
|
| 981 |
{
|
| 982 |
-
"epoch":
|
| 983 |
-
"learning_rate": 0.
|
| 984 |
-
"loss": 0.
|
| 985 |
-
"step":
|
| 986 |
},
|
| 987 |
{
|
| 988 |
-
"epoch":
|
| 989 |
-
"
|
| 990 |
-
"
|
| 991 |
-
"
|
| 992 |
-
|
| 993 |
-
|
| 994 |
-
"
|
| 995 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 996 |
},
|
| 997 |
{
|
| 998 |
-
"epoch":
|
| 999 |
-
"learning_rate": 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1000 |
"loss": 0.0035,
|
| 1001 |
-
"step":
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
-
"epoch":
|
| 1005 |
-
"learning_rate": 0.
|
| 1006 |
-
"loss": 0.
|
| 1007 |
-
"step":
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
-
"epoch":
|
| 1011 |
-
"learning_rate": 0.
|
| 1012 |
-
"loss": 0.
|
| 1013 |
-
"step":
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
-
"epoch":
|
| 1017 |
-
"
|
| 1018 |
-
"
|
| 1019 |
-
"
|
| 1020 |
-
"eval_runtime": 6.2578,
|
| 1021 |
-
"eval_samples_per_second": 79.9,
|
| 1022 |
-
"eval_steps_per_second": 3.995,
|
| 1023 |
-
"step": 415836
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"epoch":
|
| 1027 |
-
"learning_rate": 0.
|
| 1028 |
-
"loss": 0.
|
| 1029 |
-
"step":
|
| 1030 |
},
|
| 1031 |
{
|
| 1032 |
-
"epoch":
|
| 1033 |
-
"learning_rate": 0.
|
| 1034 |
-
"loss": 0.
|
| 1035 |
-
"step":
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
-
"epoch":
|
| 1039 |
-
"
|
| 1040 |
-
"
|
| 1041 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1042 |
},
|
| 1043 |
{
|
| 1044 |
-
"epoch":
|
| 1045 |
-
"
|
| 1046 |
-
"
|
| 1047 |
-
"
|
| 1048 |
-
"eval_runtime": 6.2656,
|
| 1049 |
-
"eval_samples_per_second": 79.801,
|
| 1050 |
-
"eval_steps_per_second": 3.99,
|
| 1051 |
-
"step": 427387
|
| 1052 |
},
|
| 1053 |
{
|
| 1054 |
-
"epoch":
|
| 1055 |
-
"learning_rate": 0.
|
| 1056 |
-
"loss": 0.
|
| 1057 |
-
"step":
|
| 1058 |
},
|
| 1059 |
{
|
| 1060 |
-
"epoch": 37
|
| 1061 |
-
"learning_rate": 0.
|
| 1062 |
-
"loss": 0.
|
| 1063 |
-
"step":
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
-
"epoch":
|
| 1067 |
-
"learning_rate": 0.
|
| 1068 |
-
"loss": 0.
|
| 1069 |
-
"step":
|
| 1070 |
},
|
| 1071 |
{
|
| 1072 |
-
"epoch":
|
| 1073 |
-
"
|
| 1074 |
-
"
|
| 1075 |
-
"
|
| 1076 |
-
"eval_runtime": 6.266,
|
| 1077 |
-
"eval_samples_per_second": 79.796,
|
| 1078 |
-
"eval_steps_per_second": 3.99,
|
| 1079 |
-
"step": 438938
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
-
"epoch":
|
| 1083 |
-
"learning_rate": 0.
|
| 1084 |
-
"loss": 0.
|
| 1085 |
-
"step":
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
-
"epoch":
|
| 1089 |
-
"learning_rate": 0.
|
| 1090 |
-
"loss": 0.
|
| 1091 |
-
"step":
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
-
"epoch":
|
| 1095 |
-
"
|
| 1096 |
-
"
|
| 1097 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1098 |
},
|
| 1099 |
{
|
| 1100 |
-
"epoch":
|
| 1101 |
-
"
|
| 1102 |
-
"
|
| 1103 |
-
"
|
| 1104 |
-
"eval_runtime": 6.2947,
|
| 1105 |
-
"eval_samples_per_second": 79.432,
|
| 1106 |
-
"eval_steps_per_second": 3.972,
|
| 1107 |
-
"step": 450489
|
| 1108 |
},
|
| 1109 |
{
|
| 1110 |
-
"epoch":
|
| 1111 |
-
"learning_rate": 0.
|
| 1112 |
-
"loss": 0.
|
| 1113 |
-
"step":
|
| 1114 |
},
|
| 1115 |
{
|
| 1116 |
-
"epoch":
|
| 1117 |
-
"learning_rate": 0.
|
| 1118 |
-
"loss": 0.
|
| 1119 |
-
"step":
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
-
"epoch":
|
| 1123 |
-
"learning_rate": 0.
|
| 1124 |
-
"loss": 0.
|
| 1125 |
-
"step":
|
| 1126 |
},
|
| 1127 |
{
|
| 1128 |
-
"epoch":
|
| 1129 |
-
"
|
| 1130 |
-
"
|
| 1131 |
-
"
|
| 1132 |
-
"eval_runtime": 6.2086,
|
| 1133 |
-
"eval_samples_per_second": 80.533,
|
| 1134 |
-
"eval_steps_per_second": 4.027,
|
| 1135 |
-
"step": 462040
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
-
"epoch":
|
| 1139 |
-
"learning_rate": 0.
|
| 1140 |
-
"loss": 0.
|
| 1141 |
-
"step":
|
| 1142 |
},
|
| 1143 |
{
|
| 1144 |
-
"epoch":
|
| 1145 |
-
"
|
| 1146 |
-
"
|
| 1147 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1148 |
},
|
| 1149 |
{
|
| 1150 |
-
"epoch":
|
| 1151 |
-
"learning_rate": 0.
|
| 1152 |
-
"loss": 0.
|
| 1153 |
-
"step":
|
| 1154 |
},
|
| 1155 |
{
|
| 1156 |
-
"epoch":
|
| 1157 |
-
"
|
| 1158 |
-
"
|
| 1159 |
-
"
|
| 1160 |
-
"eval_runtime": 6.2439,
|
| 1161 |
-
"eval_samples_per_second": 80.078,
|
| 1162 |
-
"eval_steps_per_second": 4.004,
|
| 1163 |
-
"step": 473591
|
| 1164 |
},
|
| 1165 |
{
|
| 1166 |
-
"epoch":
|
| 1167 |
-
"learning_rate": 0.
|
| 1168 |
-
"loss": 0.
|
| 1169 |
-
"step":
|
| 1170 |
},
|
| 1171 |
{
|
| 1172 |
-
"epoch":
|
| 1173 |
-
"learning_rate": 0.
|
| 1174 |
-
"loss": 0.
|
| 1175 |
-
"step":
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
-
"epoch":
|
| 1179 |
-
"learning_rate": 0.
|
| 1180 |
-
"loss": 0.
|
| 1181 |
-
"step":
|
| 1182 |
},
|
| 1183 |
{
|
| 1184 |
-
"epoch":
|
| 1185 |
-
"
|
| 1186 |
-
"
|
| 1187 |
-
"
|
| 1188 |
-
"eval_runtime": 6.2294,
|
| 1189 |
-
"eval_samples_per_second": 80.264,
|
| 1190 |
-
"eval_steps_per_second": 4.013,
|
| 1191 |
-
"step": 485142
|
| 1192 |
},
|
| 1193 |
{
|
| 1194 |
-
"epoch":
|
| 1195 |
-
"learning_rate": 0.
|
| 1196 |
-
"loss": 0.
|
| 1197 |
-
"step":
|
| 1198 |
},
|
| 1199 |
{
|
| 1200 |
-
"epoch":
|
| 1201 |
-
"
|
| 1202 |
-
"
|
| 1203 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1204 |
},
|
| 1205 |
{
|
| 1206 |
-
"epoch":
|
| 1207 |
-
"learning_rate": 0.
|
| 1208 |
-
"loss": 0.
|
| 1209 |
-
"step":
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
-
"epoch":
|
| 1213 |
-
"
|
| 1214 |
-
"
|
| 1215 |
-
"
|
| 1216 |
-
"eval_runtime": 6.2148,
|
| 1217 |
-
"eval_samples_per_second": 80.454,
|
| 1218 |
-
"eval_steps_per_second": 4.023,
|
| 1219 |
-
"step": 496693
|
| 1220 |
},
|
| 1221 |
{
|
| 1222 |
-
"epoch":
|
| 1223 |
-
"learning_rate": 0.
|
| 1224 |
-
"loss": 0.
|
| 1225 |
-
"step":
|
| 1226 |
},
|
| 1227 |
{
|
| 1228 |
-
"epoch":
|
| 1229 |
-
"learning_rate": 0.
|
| 1230 |
-
"loss": 0.
|
| 1231 |
-
"step":
|
| 1232 |
},
|
| 1233 |
{
|
| 1234 |
-
"epoch":
|
| 1235 |
-
"learning_rate": 0.
|
| 1236 |
-
"loss": 0.
|
| 1237 |
-
"step":
|
| 1238 |
},
|
| 1239 |
{
|
| 1240 |
-
"epoch":
|
| 1241 |
-
"
|
| 1242 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1243 |
"eval_mean_distance": 0,
|
| 1244 |
-
"eval_runtime":
|
| 1245 |
-
"eval_samples_per_second":
|
| 1246 |
-
"eval_steps_per_second":
|
| 1247 |
-
"step":
|
| 1248 |
},
|
| 1249 |
{
|
| 1250 |
-
"epoch":
|
| 1251 |
-
"learning_rate": 0.
|
| 1252 |
-
"loss": 0.
|
| 1253 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1254 |
},
|
| 1255 |
{
|
| 1256 |
-
"epoch":
|
| 1257 |
-
"learning_rate": 0.
|
| 1258 |
-
"loss": 0.
|
| 1259 |
-
"step":
|
| 1260 |
},
|
| 1261 |
{
|
| 1262 |
-
"epoch":
|
| 1263 |
-
"learning_rate": 0.
|
| 1264 |
-
"loss": 0.
|
| 1265 |
-
"step":
|
| 1266 |
},
|
| 1267 |
{
|
| 1268 |
-
"epoch":
|
| 1269 |
-
"eval_loss": 0.
|
| 1270 |
-
"eval_max_distance":
|
| 1271 |
"eval_mean_distance": 0,
|
| 1272 |
-
"eval_runtime":
|
| 1273 |
-
"eval_samples_per_second":
|
| 1274 |
-
"eval_steps_per_second":
|
| 1275 |
-
"step":
|
| 1276 |
},
|
| 1277 |
{
|
| 1278 |
-
"epoch":
|
| 1279 |
-
"learning_rate": 0.
|
| 1280 |
-
"loss": 0.
|
| 1281 |
-
"step":
|
| 1282 |
},
|
| 1283 |
{
|
| 1284 |
-
"epoch":
|
| 1285 |
-
"learning_rate": 0.
|
| 1286 |
-
"loss": 0.
|
| 1287 |
-
"step":
|
| 1288 |
},
|
| 1289 |
{
|
| 1290 |
-
"epoch":
|
| 1291 |
-
"learning_rate": 9.
|
| 1292 |
-
"loss": 0.
|
| 1293 |
-
"step":
|
| 1294 |
},
|
| 1295 |
{
|
| 1296 |
-
"epoch":
|
| 1297 |
-
"
|
| 1298 |
-
"
|
| 1299 |
-
"
|
| 1300 |
-
"eval_runtime": 6.231,
|
| 1301 |
-
"eval_samples_per_second": 80.244,
|
| 1302 |
-
"eval_steps_per_second": 4.012,
|
| 1303 |
-
"step": 531346
|
| 1304 |
},
|
| 1305 |
{
|
| 1306 |
-
"epoch":
|
| 1307 |
-
"learning_rate": 8.
|
| 1308 |
-
"loss": 0.
|
| 1309 |
-
"step":
|
| 1310 |
},
|
| 1311 |
{
|
| 1312 |
-
"epoch":
|
| 1313 |
-
"learning_rate": 8.
|
| 1314 |
-
"loss": 0.
|
| 1315 |
-
"step":
|
| 1316 |
},
|
| 1317 |
{
|
| 1318 |
-
"epoch":
|
| 1319 |
-
"learning_rate": 7.
|
| 1320 |
-
"loss": 0.
|
| 1321 |
-
"step":
|
| 1322 |
},
|
| 1323 |
{
|
| 1324 |
-
"epoch":
|
| 1325 |
-
"eval_loss": 0.
|
| 1326 |
-
"eval_max_distance":
|
| 1327 |
"eval_mean_distance": 0,
|
| 1328 |
-
"eval_runtime":
|
| 1329 |
-
"eval_samples_per_second":
|
| 1330 |
-
"eval_steps_per_second":
|
| 1331 |
-
"step":
|
| 1332 |
},
|
| 1333 |
{
|
| 1334 |
-
"epoch":
|
| 1335 |
-
"learning_rate":
|
| 1336 |
-
"loss": 0.
|
| 1337 |
-
"step":
|
| 1338 |
},
|
| 1339 |
{
|
| 1340 |
-
"epoch":
|
| 1341 |
-
"learning_rate":
|
| 1342 |
-
"loss": 0.
|
| 1343 |
-
"step":
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
-
"epoch":
|
| 1347 |
-
"learning_rate":
|
| 1348 |
-
"loss": 0.
|
| 1349 |
-
"step":
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
-
"epoch":
|
| 1353 |
-
"
|
| 1354 |
-
"
|
| 1355 |
-
"
|
| 1356 |
-
"eval_runtime": 6.2182,
|
| 1357 |
-
"eval_samples_per_second": 80.408,
|
| 1358 |
-
"eval_steps_per_second": 4.02,
|
| 1359 |
-
"step": 554448
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
-
"epoch":
|
| 1363 |
-
"learning_rate": 4.
|
| 1364 |
-
"loss": 0.
|
| 1365 |
-
"step":
|
| 1366 |
},
|
| 1367 |
{
|
| 1368 |
-
"epoch":
|
| 1369 |
-
"learning_rate":
|
| 1370 |
-
"loss": 0.
|
| 1371 |
-
"step":
|
| 1372 |
},
|
| 1373 |
{
|
| 1374 |
-
"epoch":
|
| 1375 |
-
"learning_rate":
|
| 1376 |
-
"loss": 0.
|
| 1377 |
-
"step":
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
-
"epoch":
|
| 1381 |
-
"eval_loss": 0.
|
| 1382 |
-
"eval_max_distance":
|
| 1383 |
"eval_mean_distance": 0,
|
| 1384 |
-
"eval_runtime":
|
| 1385 |
-
"eval_samples_per_second":
|
| 1386 |
-
"eval_steps_per_second":
|
| 1387 |
-
"step":
|
| 1388 |
},
|
| 1389 |
{
|
| 1390 |
-
"epoch":
|
| 1391 |
-
"learning_rate":
|
| 1392 |
-
"loss": 0.
|
| 1393 |
-
"step":
|
| 1394 |
},
|
| 1395 |
{
|
| 1396 |
-
"epoch":
|
| 1397 |
-
"learning_rate":
|
| 1398 |
-
"loss": 0.
|
| 1399 |
-
"step":
|
| 1400 |
},
|
| 1401 |
{
|
| 1402 |
-
"epoch":
|
| 1403 |
-
"learning_rate":
|
| 1404 |
-
"loss": 0.
|
| 1405 |
-
"step":
|
| 1406 |
},
|
| 1407 |
{
|
| 1408 |
-
"epoch":
|
| 1409 |
-
"
|
| 1410 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1411 |
"eval_mean_distance": 0,
|
| 1412 |
-
"eval_runtime":
|
| 1413 |
-
"eval_samples_per_second":
|
| 1414 |
-
"eval_steps_per_second":
|
| 1415 |
-
"step":
|
| 1416 |
-
},
|
| 1417 |
-
{
|
| 1418 |
-
"epoch":
|
| 1419 |
-
"step":
|
| 1420 |
-
"total_flos":
|
| 1421 |
-
"train_loss": 0.
|
| 1422 |
-
"train_runtime":
|
| 1423 |
-
"train_samples_per_second":
|
| 1424 |
-
"train_steps_per_second":
|
| 1425 |
}
|
| 1426 |
],
|
| 1427 |
-
"logging_steps":
|
| 1428 |
-
"max_steps":
|
| 1429 |
-
"num_train_epochs":
|
| 1430 |
-
"save_steps":
|
| 1431 |
-
"total_flos":
|
| 1432 |
"trial_name": null,
|
| 1433 |
"trial_params": null
|
| 1434 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 77940,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0,
|
| 13 |
+
"learning_rate": 1.2830382345393893e-07,
|
| 14 |
+
"loss": 13.6904,
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.15,
|
| 19 |
+
"learning_rate": 5.003849114703618e-05,
|
| 20 |
+
"loss": 5.273,
|
| 21 |
+
"step": 390
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.3,
|
| 25 |
+
"learning_rate": 0.00010007698229407236,
|
| 26 |
+
"loss": 0.4624,
|
| 27 |
+
"step": 780
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.45,
|
| 31 |
+
"learning_rate": 0.00015011547344110854,
|
| 32 |
+
"loss": 0.3258,
|
| 33 |
+
"step": 1170
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.6,
|
| 37 |
+
"learning_rate": 0.00020015396458814472,
|
| 38 |
+
"loss": 0.2553,
|
| 39 |
+
"step": 1560
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.75,
|
| 43 |
+
"learning_rate": 0.0002501924557351809,
|
| 44 |
+
"loss": 0.2143,
|
| 45 |
+
"step": 1950
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.9,
|
| 49 |
+
"learning_rate": 0.0003002309468822171,
|
| 50 |
+
"loss": 0.181,
|
| 51 |
+
"step": 2340
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 1.0,
|
| 55 |
+
"eval_loss": 0.21401263773441315,
|
| 56 |
+
"eval_max_distance": 36,
|
| 57 |
+
"eval_mean_distance": 4,
|
| 58 |
+
"eval_runtime": 0.4225,
|
| 59 |
+
"eval_samples_per_second": 118.338,
|
| 60 |
+
"eval_steps_per_second": 4.734,
|
| 61 |
+
"step": 2598
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 1.05,
|
| 65 |
+
"learning_rate": 0.0003502694380292533,
|
| 66 |
+
"loss": 0.1557,
|
| 67 |
+
"step": 2730
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.2,
|
| 71 |
+
"learning_rate": 0.00040030792917628943,
|
| 72 |
+
"loss": 0.1402,
|
| 73 |
+
"step": 3120
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"epoch": 1.35,
|
| 77 |
+
"learning_rate": 0.00045034642032332564,
|
| 78 |
+
"loss": 0.1283,
|
| 79 |
+
"step": 3510
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 1.5,
|
| 83 |
+
"learning_rate": 0.0005003849114703618,
|
| 84 |
+
"loss": 0.1194,
|
| 85 |
+
"step": 3900
|
| 86 |
},
|
| 87 |
{
|
| 88 |
+
"epoch": 1.65,
|
| 89 |
+
"learning_rate": 0.0005504234026173979,
|
| 90 |
+
"loss": 0.1158,
|
| 91 |
+
"step": 4290
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.8,
|
| 95 |
+
"learning_rate": 0.0006004618937644341,
|
| 96 |
+
"loss": 0.108,
|
| 97 |
+
"step": 4680
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 1.95,
|
| 101 |
+
"learning_rate": 0.0006505003849114704,
|
| 102 |
+
"loss": 0.1067,
|
| 103 |
+
"step": 5070
|
| 104 |
},
|
| 105 |
{
|
| 106 |
"epoch": 2.0,
|
| 107 |
+
"eval_loss": 0.14157189428806305,
|
| 108 |
+
"eval_max_distance": 29,
|
| 109 |
+
"eval_mean_distance": 2,
|
| 110 |
+
"eval_runtime": 0.4066,
|
| 111 |
+
"eval_samples_per_second": 122.96,
|
| 112 |
+
"eval_steps_per_second": 4.918,
|
| 113 |
+
"step": 5196
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 2.1,
|
| 117 |
+
"learning_rate": 0.0007005388760585066,
|
| 118 |
+
"loss": 0.0924,
|
| 119 |
+
"step": 5460
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 2.25,
|
| 123 |
+
"learning_rate": 0.0007505773672055427,
|
| 124 |
+
"loss": 0.0927,
|
| 125 |
+
"step": 5850
|
| 126 |
},
|
| 127 |
{
|
| 128 |
+
"epoch": 2.4,
|
| 129 |
+
"learning_rate": 0.0008006158583525789,
|
| 130 |
+
"loss": 0.0872,
|
| 131 |
+
"step": 6240
|
| 132 |
},
|
| 133 |
{
|
| 134 |
+
"epoch": 2.55,
|
| 135 |
+
"learning_rate": 0.0008506543494996151,
|
| 136 |
+
"loss": 0.0841,
|
| 137 |
+
"step": 6630
|
|
|
|
|
|
|
|
|
|
|
|
|
| 138 |
},
|
| 139 |
{
|
| 140 |
+
"epoch": 2.7,
|
| 141 |
+
"learning_rate": 0.0009006928406466513,
|
| 142 |
+
"loss": 0.0808,
|
| 143 |
+
"step": 7020
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 2.85,
|
| 147 |
+
"learning_rate": 0.0009507313317936874,
|
| 148 |
+
"loss": 0.0801,
|
| 149 |
+
"step": 7410
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 3.0,
|
| 153 |
+
"eval_loss": 0.10980188101530075,
|
| 154 |
+
"eval_max_distance": 22,
|
| 155 |
+
"eval_mean_distance": 2,
|
| 156 |
+
"eval_runtime": 0.4052,
|
| 157 |
+
"eval_samples_per_second": 123.401,
|
| 158 |
+
"eval_steps_per_second": 4.936,
|
| 159 |
+
"step": 7794
|
| 160 |
},
|
| 161 |
{
|
| 162 |
+
"epoch": 3.0,
|
| 163 |
+
"learning_rate": 0.0009999144641176974,
|
| 164 |
+
"loss": 0.0778,
|
| 165 |
+
"step": 7800
|
|
|
|
|
|
|
|
|
|
|
|
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"epoch": 3.15,
|
| 169 |
+
"learning_rate": 0.0009943546317680268,
|
| 170 |
+
"loss": 0.0684,
|
| 171 |
+
"step": 8190
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 3.3,
|
| 175 |
+
"learning_rate": 0.000988794799418356,
|
| 176 |
+
"loss": 0.0692,
|
| 177 |
+
"step": 8580
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"epoch": 3.45,
|
| 181 |
+
"learning_rate": 0.0009832349670686853,
|
| 182 |
+
"loss": 0.0629,
|
| 183 |
+
"step": 8970
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"epoch": 3.6,
|
| 187 |
+
"learning_rate": 0.0009776751347190145,
|
| 188 |
+
"loss": 0.0653,
|
| 189 |
+
"step": 9360
|
|
|
|
|
|
|
|
|
|
|
|
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"epoch": 3.75,
|
| 193 |
+
"learning_rate": 0.0009721153023693439,
|
| 194 |
+
"loss": 0.0595,
|
| 195 |
+
"step": 9750
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"epoch": 3.9,
|
| 199 |
+
"learning_rate": 0.0009665554700196733,
|
| 200 |
+
"loss": 0.0575,
|
| 201 |
+
"step": 10140
|
| 202 |
},
|
| 203 |
{
|
| 204 |
+
"epoch": 4.0,
|
| 205 |
+
"eval_loss": 0.10807737708091736,
|
| 206 |
+
"eval_max_distance": 18,
|
| 207 |
+
"eval_mean_distance": 2,
|
| 208 |
+
"eval_runtime": 0.3927,
|
| 209 |
+
"eval_samples_per_second": 127.308,
|
| 210 |
+
"eval_steps_per_second": 5.092,
|
| 211 |
+
"step": 10392
|
| 212 |
},
|
| 213 |
{
|
| 214 |
+
"epoch": 4.05,
|
| 215 |
+
"learning_rate": 0.0009609956376700025,
|
| 216 |
+
"loss": 0.1012,
|
| 217 |
+
"step": 10530
|
|
|
|
|
|
|
|
|
|
|
|
|
| 218 |
},
|
| 219 |
{
|
| 220 |
+
"epoch": 4.2,
|
| 221 |
+
"learning_rate": 0.0009554358053203319,
|
| 222 |
+
"loss": 0.0526,
|
| 223 |
+
"step": 10920
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"epoch": 4.35,
|
| 227 |
+
"learning_rate": 0.0009498759729706612,
|
| 228 |
+
"loss": 0.0496,
|
| 229 |
+
"step": 11310
|
| 230 |
},
|
| 231 |
{
|
| 232 |
+
"epoch": 4.5,
|
| 233 |
+
"learning_rate": 0.0009443161406209905,
|
| 234 |
+
"loss": 0.0492,
|
| 235 |
+
"step": 11700
|
| 236 |
},
|
| 237 |
{
|
| 238 |
+
"epoch": 4.65,
|
| 239 |
+
"learning_rate": 0.0009387563082713198,
|
| 240 |
+
"loss": 0.0483,
|
| 241 |
+
"step": 12090
|
|
|
|
|
|
|
|
|
|
|
|
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 4.8,
|
| 245 |
+
"learning_rate": 0.0009331964759216492,
|
| 246 |
+
"loss": 0.0469,
|
| 247 |
+
"step": 12480
|
| 248 |
},
|
| 249 |
{
|
| 250 |
+
"epoch": 4.95,
|
| 251 |
+
"learning_rate": 0.0009276366435719784,
|
| 252 |
+
"loss": 0.0452,
|
| 253 |
+
"step": 12870
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"epoch": 5.0,
|
| 257 |
+
"eval_loss": 0.08966636657714844,
|
| 258 |
+
"eval_max_distance": 14,
|
| 259 |
+
"eval_mean_distance": 1,
|
| 260 |
+
"eval_runtime": 0.3879,
|
| 261 |
+
"eval_samples_per_second": 128.906,
|
| 262 |
+
"eval_steps_per_second": 5.156,
|
| 263 |
+
"step": 12990
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 5.1,
|
| 267 |
+
"learning_rate": 0.0009220768112223078,
|
| 268 |
+
"loss": 0.0392,
|
| 269 |
+
"step": 13260
|
|
|
|
|
|
|
|
|
|
|
|
|
| 270 |
},
|
| 271 |
{
|
| 272 |
+
"epoch": 5.25,
|
| 273 |
+
"learning_rate": 0.0009165169788726371,
|
| 274 |
"loss": 0.0397,
|
| 275 |
+
"step": 13650
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 5.4,
|
| 279 |
+
"learning_rate": 0.0009109571465229664,
|
| 280 |
+
"loss": 0.0393,
|
| 281 |
+
"step": 14040
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 5.55,
|
| 285 |
+
"learning_rate": 0.0009053973141732957,
|
| 286 |
+
"loss": 0.0399,
|
| 287 |
+
"step": 14430
|
| 288 |
},
|
| 289 |
{
|
| 290 |
+
"epoch": 5.7,
|
| 291 |
+
"learning_rate": 0.0008998374818236251,
|
| 292 |
+
"loss": 0.039,
|
| 293 |
+
"step": 14820
|
|
|
|
|
|
|
|
|
|
|
|
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"epoch": 5.85,
|
| 297 |
+
"learning_rate": 0.0008942776494739543,
|
| 298 |
+
"loss": 0.0372,
|
| 299 |
+
"step": 15210
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"epoch": 6.0,
|
| 303 |
+
"eval_loss": 0.07197271287441254,
|
| 304 |
+
"eval_max_distance": 15,
|
| 305 |
+
"eval_mean_distance": 1,
|
| 306 |
+
"eval_runtime": 0.3764,
|
| 307 |
+
"eval_samples_per_second": 132.832,
|
| 308 |
+
"eval_steps_per_second": 5.313,
|
| 309 |
+
"step": 15588
|
| 310 |
},
|
| 311 |
{
|
| 312 |
+
"epoch": 6.0,
|
| 313 |
+
"learning_rate": 0.0008887178171242837,
|
| 314 |
+
"loss": 0.039,
|
| 315 |
+
"step": 15600
|
| 316 |
},
|
| 317 |
{
|
| 318 |
+
"epoch": 6.15,
|
| 319 |
+
"learning_rate": 0.000883157984774613,
|
| 320 |
+
"loss": 0.0312,
|
| 321 |
+
"step": 15990
|
|
|
|
|
|
|
|
|
|
|
|
|
| 322 |
},
|
| 323 |
{
|
| 324 |
+
"epoch": 6.3,
|
| 325 |
+
"learning_rate": 0.0008775981524249422,
|
| 326 |
+
"loss": 0.0337,
|
| 327 |
+
"step": 16380
|
| 328 |
},
|
| 329 |
{
|
| 330 |
+
"epoch": 6.45,
|
| 331 |
+
"learning_rate": 0.0008720383200752716,
|
| 332 |
+
"loss": 0.0328,
|
| 333 |
+
"step": 16770
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 6.61,
|
| 337 |
+
"learning_rate": 0.0008664784877256009,
|
| 338 |
+
"loss": 0.0327,
|
| 339 |
+
"step": 17160
|
| 340 |
},
|
| 341 |
{
|
| 342 |
+
"epoch": 6.76,
|
| 343 |
+
"learning_rate": 0.0008609186553759302,
|
| 344 |
+
"loss": 0.0305,
|
| 345 |
+
"step": 17550
|
|
|
|
|
|
|
|
|
|
|
|
|
| 346 |
},
|
| 347 |
{
|
| 348 |
+
"epoch": 6.91,
|
| 349 |
+
"learning_rate": 0.0008553588230262595,
|
| 350 |
+
"loss": 0.0323,
|
| 351 |
+
"step": 17940
|
| 352 |
},
|
| 353 |
{
|
| 354 |
+
"epoch": 7.0,
|
| 355 |
+
"eval_loss": 0.08398188650608063,
|
| 356 |
+
"eval_max_distance": 12,
|
| 357 |
+
"eval_mean_distance": 1,
|
| 358 |
+
"eval_runtime": 0.3765,
|
| 359 |
+
"eval_samples_per_second": 132.8,
|
| 360 |
+
"eval_steps_per_second": 5.312,
|
| 361 |
+
"step": 18186
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 7.06,
|
| 365 |
+
"learning_rate": 0.0008497989906765889,
|
| 366 |
+
"loss": 0.0286,
|
| 367 |
+
"step": 18330
|
| 368 |
},
|
| 369 |
{
|
| 370 |
+
"epoch": 7.21,
|
| 371 |
+
"learning_rate": 0.0008442391583269181,
|
| 372 |
+
"loss": 0.0263,
|
| 373 |
+
"step": 18720
|
|
|
|
|
|
|
|
|
|
|
|
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"epoch": 7.36,
|
| 377 |
+
"learning_rate": 0.0008386793259772475,
|
| 378 |
+
"loss": 0.0269,
|
| 379 |
+
"step": 19110
|
| 380 |
},
|
| 381 |
{
|
| 382 |
+
"epoch": 7.51,
|
| 383 |
+
"learning_rate": 0.0008331194936275768,
|
| 384 |
+
"loss": 0.0268,
|
| 385 |
+
"step": 19500
|
| 386 |
},
|
| 387 |
{
|
| 388 |
+
"epoch": 7.66,
|
| 389 |
+
"learning_rate": 0.0008275596612779061,
|
| 390 |
+
"loss": 0.0283,
|
| 391 |
+
"step": 19890
|
| 392 |
},
|
| 393 |
{
|
| 394 |
+
"epoch": 7.81,
|
| 395 |
+
"learning_rate": 0.0008219998289282354,
|
| 396 |
+
"loss": 0.0274,
|
| 397 |
+
"step": 20280
|
|
|
|
|
|
|
|
|
|
|
|
|
| 398 |
},
|
| 399 |
{
|
| 400 |
+
"epoch": 7.96,
|
| 401 |
+
"learning_rate": 0.0008164399965785648,
|
| 402 |
+
"loss": 0.0267,
|
| 403 |
+
"step": 20670
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"epoch": 8.0,
|
| 407 |
+
"eval_loss": 0.07682657241821289,
|
| 408 |
+
"eval_max_distance": 16,
|
| 409 |
+
"eval_mean_distance": 1,
|
| 410 |
+
"eval_runtime": 0.3687,
|
| 411 |
+
"eval_samples_per_second": 135.62,
|
| 412 |
+
"eval_steps_per_second": 5.425,
|
| 413 |
+
"step": 20784
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"epoch": 8.11,
|
| 417 |
+
"learning_rate": 0.000810880164228894,
|
| 418 |
+
"loss": 0.0235,
|
| 419 |
+
"step": 21060
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 8.26,
|
| 423 |
+
"learning_rate": 0.0008053203318792234,
|
| 424 |
+
"loss": 0.0221,
|
| 425 |
+
"step": 21450
|
|
|
|
|
|
|
|
|
|
|
|
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 8.41,
|
| 429 |
+
"learning_rate": 0.0007997604995295527,
|
| 430 |
+
"loss": 0.0221,
|
| 431 |
+
"step": 21840
|
| 432 |
},
|
| 433 |
{
|
| 434 |
+
"epoch": 8.56,
|
| 435 |
+
"learning_rate": 0.0007942006671798819,
|
| 436 |
+
"loss": 0.0223,
|
| 437 |
+
"step": 22230
|
| 438 |
},
|
| 439 |
{
|
| 440 |
+
"epoch": 8.71,
|
| 441 |
+
"learning_rate": 0.0007886408348302113,
|
| 442 |
+
"loss": 0.0233,
|
| 443 |
+
"step": 22620
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"epoch": 8.86,
|
| 447 |
+
"learning_rate": 0.0007830810024805405,
|
| 448 |
+
"loss": 0.0231,
|
| 449 |
+
"step": 23010
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 9.0,
|
| 453 |
+
"eval_loss": 0.06973634660243988,
|
| 454 |
+
"eval_max_distance": 10,
|
| 455 |
"eval_mean_distance": 1,
|
| 456 |
+
"eval_runtime": 0.3759,
|
| 457 |
+
"eval_samples_per_second": 133.026,
|
| 458 |
+
"eval_steps_per_second": 5.321,
|
| 459 |
+
"step": 23382
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 9.01,
|
| 463 |
+
"learning_rate": 0.0007775211701308699,
|
| 464 |
+
"loss": 0.0227,
|
| 465 |
+
"step": 23400
|
| 466 |
},
|
| 467 |
{
|
| 468 |
+
"epoch": 9.16,
|
| 469 |
+
"learning_rate": 0.0007719613377811992,
|
| 470 |
+
"loss": 0.0185,
|
| 471 |
+
"step": 23790
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 9.31,
|
| 475 |
+
"learning_rate": 0.0007664015054315285,
|
| 476 |
+
"loss": 0.0183,
|
| 477 |
+
"step": 24180
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 9.46,
|
| 481 |
+
"learning_rate": 0.0007608416730818578,
|
| 482 |
+
"loss": 0.0191,
|
| 483 |
+
"step": 24570
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 9.61,
|
| 487 |
+
"learning_rate": 0.0007552818407321872,
|
| 488 |
"loss": 0.019,
|
| 489 |
+
"step": 24960
|
| 490 |
},
|
| 491 |
{
|
| 492 |
+
"epoch": 9.76,
|
| 493 |
+
"learning_rate": 0.0007497220083825164,
|
| 494 |
+
"loss": 0.0193,
|
| 495 |
+
"step": 25350
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 9.91,
|
| 499 |
+
"learning_rate": 0.0007441621760328458,
|
| 500 |
+
"loss": 0.0199,
|
| 501 |
+
"step": 25740
|
| 502 |
},
|
| 503 |
{
|
| 504 |
+
"epoch": 10.0,
|
| 505 |
+
"eval_loss": 0.07169829308986664,
|
| 506 |
+
"eval_max_distance": 9,
|
| 507 |
"eval_mean_distance": 1,
|
| 508 |
+
"eval_runtime": 0.3704,
|
| 509 |
+
"eval_samples_per_second": 134.993,
|
| 510 |
+
"eval_steps_per_second": 5.4,
|
| 511 |
+
"step": 25980
|
| 512 |
},
|
| 513 |
{
|
| 514 |
+
"epoch": 10.06,
|
| 515 |
+
"learning_rate": 0.0007386023436831751,
|
| 516 |
+
"loss": 0.0184,
|
| 517 |
+
"step": 26130
|
| 518 |
},
|
| 519 |
{
|
| 520 |
+
"epoch": 10.21,
|
| 521 |
+
"learning_rate": 0.0007330425113335044,
|
| 522 |
+
"loss": 0.016,
|
| 523 |
+
"step": 26520
|
| 524 |
},
|
| 525 |
{
|
| 526 |
+
"epoch": 10.36,
|
| 527 |
+
"learning_rate": 0.0007274826789838337,
|
| 528 |
+
"loss": 0.0164,
|
| 529 |
+
"step": 26910
|
| 530 |
},
|
| 531 |
{
|
| 532 |
+
"epoch": 10.51,
|
| 533 |
+
"learning_rate": 0.0007219228466341631,
|
| 534 |
+
"loss": 0.016,
|
| 535 |
+
"step": 27300
|
|
|
|
|
|
|
|
|
|
|
|
|
| 536 |
},
|
| 537 |
{
|
| 538 |
+
"epoch": 10.66,
|
| 539 |
+
"learning_rate": 0.0007163630142844923,
|
| 540 |
+
"loss": 0.0169,
|
| 541 |
+
"step": 27690
|
| 542 |
},
|
| 543 |
{
|
| 544 |
+
"epoch": 10.81,
|
| 545 |
+
"learning_rate": 0.0007108031819348217,
|
| 546 |
+
"loss": 0.0165,
|
| 547 |
+
"step": 28080
|
| 548 |
},
|
| 549 |
{
|
| 550 |
+
"epoch": 10.96,
|
| 551 |
+
"learning_rate": 0.000705243349585151,
|
| 552 |
+
"loss": 0.0168,
|
| 553 |
+
"step": 28470
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"epoch": 11.0,
|
| 557 |
+
"eval_loss": 0.08123478293418884,
|
| 558 |
+
"eval_max_distance": 16,
|
| 559 |
+
"eval_mean_distance": 1,
|
| 560 |
+
"eval_runtime": 0.3865,
|
| 561 |
+
"eval_samples_per_second": 129.356,
|
| 562 |
+
"eval_steps_per_second": 5.174,
|
| 563 |
+
"step": 28578
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 11.11,
|
| 567 |
+
"learning_rate": 0.0006996835172354803,
|
| 568 |
+
"loss": 0.015,
|
| 569 |
+
"step": 28860
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 11.26,
|
| 573 |
+
"learning_rate": 0.0006941236848858096,
|
| 574 |
+
"loss": 0.0137,
|
| 575 |
+
"step": 29250
|
| 576 |
},
|
| 577 |
{
|
| 578 |
+
"epoch": 11.41,
|
| 579 |
+
"learning_rate": 0.0006885638525361389,
|
| 580 |
+
"loss": 0.0151,
|
| 581 |
+
"step": 29640
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 11.56,
|
| 585 |
+
"learning_rate": 0.0006830040201864682,
|
| 586 |
"loss": 0.0144,
|
| 587 |
+
"step": 30030
|
| 588 |
},
|
| 589 |
{
|
| 590 |
+
"epoch": 11.71,
|
| 591 |
+
"learning_rate": 0.0006774441878367975,
|
| 592 |
+
"loss": 0.0147,
|
| 593 |
+
"step": 30420
|
| 594 |
},
|
| 595 |
{
|
| 596 |
+
"epoch": 11.86,
|
| 597 |
+
"learning_rate": 0.0006718843554871269,
|
| 598 |
+
"loss": 0.0148,
|
| 599 |
+
"step": 30810
|
| 600 |
},
|
| 601 |
{
|
| 602 |
+
"epoch": 12.0,
|
| 603 |
+
"eval_loss": 0.09610763192176819,
|
| 604 |
+
"eval_max_distance": 12,
|
| 605 |
+
"eval_mean_distance": 1,
|
| 606 |
+
"eval_runtime": 0.3633,
|
| 607 |
+
"eval_samples_per_second": 137.639,
|
| 608 |
+
"eval_steps_per_second": 5.506,
|
| 609 |
+
"step": 31176
|
| 610 |
},
|
| 611 |
{
|
| 612 |
+
"epoch": 12.01,
|
| 613 |
+
"learning_rate": 0.0006663245231374561,
|
| 614 |
+
"loss": 0.0145,
|
| 615 |
+
"step": 31200
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 12.16,
|
| 619 |
+
"learning_rate": 0.0006607646907877855,
|
| 620 |
+
"loss": 0.0124,
|
| 621 |
+
"step": 31590
|
| 622 |
},
|
| 623 |
{
|
| 624 |
+
"epoch": 12.31,
|
| 625 |
+
"learning_rate": 0.0006552048584381148,
|
| 626 |
+
"loss": 0.0117,
|
| 627 |
+
"step": 31980
|
| 628 |
},
|
| 629 |
{
|
| 630 |
+
"epoch": 12.46,
|
| 631 |
+
"learning_rate": 0.0006496450260884441,
|
| 632 |
+
"loss": 0.0121,
|
| 633 |
+
"step": 32370
|
|
|
|
|
|
|
|
|
|
|
|
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"epoch": 12.61,
|
| 637 |
+
"learning_rate": 0.0006440851937387734,
|
| 638 |
+
"loss": 0.0124,
|
| 639 |
+
"step": 32760
|
| 640 |
},
|
| 641 |
{
|
| 642 |
+
"epoch": 12.76,
|
| 643 |
+
"learning_rate": 0.0006385253613891028,
|
| 644 |
+
"loss": 0.0125,
|
| 645 |
+
"step": 33150
|
| 646 |
},
|
| 647 |
{
|
| 648 |
+
"epoch": 12.91,
|
| 649 |
+
"learning_rate": 0.000632965529039432,
|
| 650 |
+
"loss": 0.0128,
|
| 651 |
+
"step": 33540
|
| 652 |
},
|
| 653 |
{
|
| 654 |
+
"epoch": 13.0,
|
| 655 |
+
"eval_loss": 0.08225859701633453,
|
| 656 |
+
"eval_max_distance": 9,
|
| 657 |
+
"eval_mean_distance": 1,
|
| 658 |
+
"eval_runtime": 0.3712,
|
| 659 |
+
"eval_samples_per_second": 134.695,
|
| 660 |
+
"eval_steps_per_second": 5.388,
|
| 661 |
+
"step": 33774
|
| 662 |
},
|
| 663 |
{
|
| 664 |
+
"epoch": 13.06,
|
| 665 |
+
"learning_rate": 0.0006274056966897614,
|
| 666 |
"loss": 0.0116,
|
| 667 |
+
"step": 33930
|
| 668 |
},
|
| 669 |
{
|
| 670 |
+
"epoch": 13.21,
|
| 671 |
+
"learning_rate": 0.0006218458643400907,
|
| 672 |
+
"loss": 0.0106,
|
| 673 |
+
"step": 34320
|
| 674 |
},
|
| 675 |
{
|
| 676 |
+
"epoch": 13.36,
|
| 677 |
+
"learning_rate": 0.00061628603199042,
|
| 678 |
+
"loss": 0.0104,
|
| 679 |
+
"step": 34710
|
| 680 |
},
|
| 681 |
{
|
| 682 |
+
"epoch": 13.51,
|
| 683 |
+
"learning_rate": 0.0006107261996407493,
|
| 684 |
+
"loss": 0.011,
|
| 685 |
+
"step": 35100
|
|
|
|
|
|
|
|
|
|
|
|
|
| 686 |
},
|
| 687 |
{
|
| 688 |
+
"epoch": 13.66,
|
| 689 |
+
"learning_rate": 0.0006051663672910787,
|
| 690 |
+
"loss": 0.0108,
|
| 691 |
+
"step": 35490
|
| 692 |
},
|
| 693 |
{
|
| 694 |
+
"epoch": 13.81,
|
| 695 |
+
"learning_rate": 0.0005996065349414079,
|
| 696 |
+
"loss": 0.0111,
|
| 697 |
+
"step": 35880
|
| 698 |
},
|
| 699 |
{
|
| 700 |
+
"epoch": 13.96,
|
| 701 |
+
"learning_rate": 0.0005940467025917372,
|
| 702 |
+
"loss": 0.0112,
|
| 703 |
+
"step": 36270
|
| 704 |
},
|
| 705 |
{
|
| 706 |
+
"epoch": 14.0,
|
| 707 |
+
"eval_loss": 0.07655028253793716,
|
| 708 |
+
"eval_max_distance": 12,
|
| 709 |
+
"eval_mean_distance": 1,
|
| 710 |
+
"eval_runtime": 0.361,
|
| 711 |
+
"eval_samples_per_second": 138.506,
|
| 712 |
+
"eval_steps_per_second": 5.54,
|
| 713 |
+
"step": 36372
|
| 714 |
},
|
| 715 |
{
|
| 716 |
+
"epoch": 14.11,
|
| 717 |
+
"learning_rate": 0.0005884868702420666,
|
| 718 |
+
"loss": 0.0098,
|
| 719 |
+
"step": 36660
|
| 720 |
},
|
| 721 |
{
|
| 722 |
+
"epoch": 14.26,
|
| 723 |
+
"learning_rate": 0.0005829270378923958,
|
| 724 |
+
"loss": 0.009,
|
| 725 |
+
"step": 37050
|
| 726 |
},
|
| 727 |
{
|
| 728 |
+
"epoch": 14.41,
|
| 729 |
+
"learning_rate": 0.0005773672055427252,
|
| 730 |
+
"loss": 0.0093,
|
| 731 |
+
"step": 37440
|
| 732 |
},
|
| 733 |
{
|
| 734 |
+
"epoch": 14.56,
|
| 735 |
+
"learning_rate": 0.0005718073731930545,
|
| 736 |
+
"loss": 0.0095,
|
| 737 |
+
"step": 37830
|
|
|
|
|
|
|
|
|
|
|
|
|
| 738 |
},
|
| 739 |
{
|
| 740 |
+
"epoch": 14.71,
|
| 741 |
+
"learning_rate": 0.0005662475408433838,
|
| 742 |
+
"loss": 0.0093,
|
| 743 |
+
"step": 38220
|
| 744 |
},
|
| 745 |
{
|
| 746 |
+
"epoch": 14.86,
|
| 747 |
+
"learning_rate": 0.0005606877084937131,
|
| 748 |
+
"loss": 0.0093,
|
| 749 |
+
"step": 38610
|
| 750 |
},
|
| 751 |
{
|
| 752 |
+
"epoch": 15.0,
|
| 753 |
+
"eval_loss": 0.07127052545547485,
|
| 754 |
+
"eval_max_distance": 9,
|
| 755 |
+
"eval_mean_distance": 1,
|
| 756 |
+
"eval_runtime": 0.3671,
|
| 757 |
+
"eval_samples_per_second": 136.219,
|
| 758 |
+
"eval_steps_per_second": 5.449,
|
| 759 |
+
"step": 38970
|
| 760 |
},
|
| 761 |
{
|
| 762 |
+
"epoch": 15.01,
|
| 763 |
+
"learning_rate": 0.0005551278761440425,
|
| 764 |
+
"loss": 0.0101,
|
| 765 |
+
"step": 39000
|
|
|
|
|
|
|
|
|
|
|
|
|
| 766 |
},
|
| 767 |
{
|
| 768 |
+
"epoch": 15.16,
|
| 769 |
+
"learning_rate": 0.0005495680437943717,
|
| 770 |
+
"loss": 0.0078,
|
| 771 |
+
"step": 39390
|
| 772 |
},
|
| 773 |
{
|
| 774 |
+
"epoch": 15.31,
|
| 775 |
+
"learning_rate": 0.0005440082114447011,
|
| 776 |
+
"loss": 0.0079,
|
| 777 |
+
"step": 39780
|
| 778 |
},
|
| 779 |
{
|
| 780 |
+
"epoch": 15.46,
|
| 781 |
+
"learning_rate": 0.0005384483790950304,
|
| 782 |
+
"loss": 0.0081,
|
| 783 |
+
"step": 40170
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"epoch": 15.61,
|
| 787 |
+
"learning_rate": 0.0005328885467453597,
|
| 788 |
+
"loss": 0.0085,
|
| 789 |
+
"step": 40560
|
|
|
|
|
|
|
|
|
|
|
|
|
| 790 |
},
|
| 791 |
{
|
| 792 |
+
"epoch": 15.76,
|
| 793 |
+
"learning_rate": 0.000527328714395689,
|
| 794 |
+
"loss": 0.0088,
|
| 795 |
+
"step": 40950
|
| 796 |
},
|
| 797 |
{
|
| 798 |
+
"epoch": 15.91,
|
| 799 |
+
"learning_rate": 0.0005217688820460184,
|
| 800 |
+
"loss": 0.0083,
|
| 801 |
+
"step": 41340
|
| 802 |
},
|
| 803 |
{
|
| 804 |
+
"epoch": 16.0,
|
| 805 |
+
"eval_loss": 0.08469703793525696,
|
| 806 |
+
"eval_max_distance": 14,
|
| 807 |
+
"eval_mean_distance": 1,
|
| 808 |
+
"eval_runtime": 0.3815,
|
| 809 |
+
"eval_samples_per_second": 131.073,
|
| 810 |
+
"eval_steps_per_second": 5.243,
|
| 811 |
+
"step": 41568
|
| 812 |
},
|
| 813 |
{
|
| 814 |
+
"epoch": 16.06,
|
| 815 |
+
"learning_rate": 0.0005162090496963476,
|
| 816 |
+
"loss": 0.0081,
|
| 817 |
+
"step": 41730
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 16.21,
|
| 821 |
+
"learning_rate": 0.000510649217346677,
|
| 822 |
+
"loss": 0.0069,
|
| 823 |
+
"step": 42120
|
| 824 |
},
|
| 825 |
{
|
| 826 |
+
"epoch": 16.36,
|
| 827 |
+
"learning_rate": 0.0005050893849970063,
|
| 828 |
+
"loss": 0.007,
|
| 829 |
+
"step": 42510
|
| 830 |
+
},
|
| 831 |
+
{
|
| 832 |
+
"epoch": 16.51,
|
| 833 |
+
"learning_rate": 0.0004995295526473355,
|
| 834 |
+
"loss": 0.0071,
|
| 835 |
+
"step": 42900
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 16.66,
|
| 839 |
+
"learning_rate": 0.0004939697202976649,
|
| 840 |
"loss": 0.0073,
|
| 841 |
+
"step": 43290
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"epoch": 16.81,
|
| 845 |
+
"learning_rate": 0.0004884098879479942,
|
| 846 |
+
"loss": 0.0076,
|
| 847 |
+
"step": 43680
|
| 848 |
},
|
| 849 |
{
|
| 850 |
+
"epoch": 16.96,
|
| 851 |
+
"learning_rate": 0.0004828500555983235,
|
| 852 |
+
"loss": 0.0076,
|
| 853 |
+
"step": 44070
|
| 854 |
},
|
| 855 |
{
|
| 856 |
+
"epoch": 17.0,
|
| 857 |
+
"eval_loss": 0.08625645935535431,
|
| 858 |
+
"eval_max_distance": 11,
|
| 859 |
+
"eval_mean_distance": 1,
|
| 860 |
+
"eval_runtime": 0.3551,
|
| 861 |
+
"eval_samples_per_second": 140.8,
|
| 862 |
+
"eval_steps_per_second": 5.632,
|
| 863 |
+
"step": 44166
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 17.11,
|
| 867 |
+
"learning_rate": 0.00047729022324865286,
|
| 868 |
+
"loss": 0.0064,
|
| 869 |
+
"step": 44460
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 17.26,
|
| 873 |
+
"learning_rate": 0.00047173039089898214,
|
| 874 |
+
"loss": 0.0059,
|
| 875 |
+
"step": 44850
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 17.41,
|
| 879 |
+
"learning_rate": 0.0004661705585493115,
|
| 880 |
+
"loss": 0.0064,
|
| 881 |
+
"step": 45240
|
| 882 |
},
|
| 883 |
{
|
| 884 |
+
"epoch": 17.56,
|
| 885 |
+
"learning_rate": 0.0004606107261996408,
|
| 886 |
+
"loss": 0.0068,
|
| 887 |
+
"step": 45630
|
|
|
|
|
|
|
|
|
|
|
|
|
| 888 |
},
|
| 889 |
{
|
| 890 |
+
"epoch": 17.71,
|
| 891 |
+
"learning_rate": 0.00045505089384997004,
|
| 892 |
+
"loss": 0.0066,
|
| 893 |
+
"step": 46020
|
| 894 |
},
|
| 895 |
{
|
| 896 |
+
"epoch": 17.86,
|
| 897 |
+
"learning_rate": 0.00044949106150029937,
|
| 898 |
+
"loss": 0.0064,
|
| 899 |
+
"step": 46410
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 18.0,
|
| 903 |
+
"eval_loss": 0.08296500891447067,
|
| 904 |
+
"eval_max_distance": 14,
|
| 905 |
+
"eval_mean_distance": 1,
|
| 906 |
+
"eval_runtime": 0.3721,
|
| 907 |
+
"eval_samples_per_second": 134.372,
|
| 908 |
+
"eval_steps_per_second": 5.375,
|
| 909 |
+
"step": 46764
|
| 910 |
},
|
| 911 |
{
|
| 912 |
+
"epoch": 18.01,
|
| 913 |
+
"learning_rate": 0.0004439312291506287,
|
| 914 |
+
"loss": 0.0065,
|
| 915 |
+
"step": 46800
|
|
|
|
|
|
|
|
|
|
|
|
|
| 916 |
},
|
| 917 |
{
|
| 918 |
+
"epoch": 18.16,
|
| 919 |
+
"learning_rate": 0.000438371396800958,
|
| 920 |
+
"loss": 0.0055,
|
| 921 |
+
"step": 47190
|
| 922 |
},
|
| 923 |
{
|
| 924 |
+
"epoch": 18.31,
|
| 925 |
+
"learning_rate": 0.0004328115644512873,
|
| 926 |
+
"loss": 0.0052,
|
| 927 |
+
"step": 47580
|
| 928 |
},
|
| 929 |
{
|
| 930 |
+
"epoch": 18.46,
|
| 931 |
+
"learning_rate": 0.00042725173210161665,
|
| 932 |
+
"loss": 0.0057,
|
| 933 |
+
"step": 47970
|
| 934 |
},
|
| 935 |
{
|
| 936 |
+
"epoch": 18.61,
|
| 937 |
+
"learning_rate": 0.00042169189975194593,
|
| 938 |
+
"loss": 0.006,
|
| 939 |
+
"step": 48360
|
|
|
|
|
|
|
|
|
|
|
|
|
| 940 |
},
|
| 941 |
{
|
| 942 |
+
"epoch": 18.76,
|
| 943 |
+
"learning_rate": 0.00041613206740227527,
|
| 944 |
+
"loss": 0.0055,
|
| 945 |
+
"step": 48750
|
| 946 |
},
|
| 947 |
{
|
| 948 |
+
"epoch": 18.91,
|
| 949 |
+
"learning_rate": 0.0004105722350526046,
|
| 950 |
+
"loss": 0.0054,
|
| 951 |
+
"step": 49140
|
| 952 |
},
|
| 953 |
{
|
| 954 |
+
"epoch": 19.0,
|
| 955 |
+
"eval_loss": 0.08839410543441772,
|
| 956 |
+
"eval_max_distance": 11,
|
| 957 |
+
"eval_mean_distance": 1,
|
| 958 |
+
"eval_runtime": 0.367,
|
| 959 |
+
"eval_samples_per_second": 136.245,
|
| 960 |
+
"eval_steps_per_second": 5.45,
|
| 961 |
+
"step": 49362
|
| 962 |
},
|
| 963 |
{
|
| 964 |
+
"epoch": 19.06,
|
| 965 |
+
"learning_rate": 0.0004050124027029339,
|
| 966 |
+
"loss": 0.0057,
|
| 967 |
+
"step": 49530
|
|
|
|
|
|
|
|
|
|
|
|
|
| 968 |
},
|
| 969 |
{
|
| 970 |
+
"epoch": 19.21,
|
| 971 |
+
"learning_rate": 0.0003994525703532632,
|
| 972 |
"loss": 0.0047,
|
| 973 |
+
"step": 49920
|
| 974 |
},
|
| 975 |
{
|
| 976 |
+
"epoch": 19.36,
|
| 977 |
+
"learning_rate": 0.0003938927380035925,
|
| 978 |
+
"loss": 0.0048,
|
| 979 |
+
"step": 50310
|
| 980 |
},
|
| 981 |
{
|
| 982 |
+
"epoch": 19.52,
|
| 983 |
+
"learning_rate": 0.00038833290565392183,
|
| 984 |
+
"loss": 0.0052,
|
| 985 |
+
"step": 50700
|
| 986 |
},
|
| 987 |
{
|
| 988 |
+
"epoch": 19.67,
|
| 989 |
+
"learning_rate": 0.00038277307330425117,
|
| 990 |
+
"loss": 0.005,
|
| 991 |
+
"step": 51090
|
|
|
|
|
|
|
|
|
|
|
|
|
| 992 |
},
|
| 993 |
{
|
| 994 |
+
"epoch": 19.82,
|
| 995 |
+
"learning_rate": 0.00037721324095458045,
|
| 996 |
+
"loss": 0.0048,
|
| 997 |
+
"step": 51480
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
+
"epoch": 19.97,
|
| 1001 |
+
"learning_rate": 0.0003716534086049098,
|
| 1002 |
+
"loss": 0.0052,
|
| 1003 |
+
"step": 51870
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
+
"epoch": 20.0,
|
| 1007 |
+
"eval_loss": 0.08214738219976425,
|
| 1008 |
+
"eval_max_distance": 10,
|
| 1009 |
+
"eval_mean_distance": 1,
|
| 1010 |
+
"eval_runtime": 0.3692,
|
| 1011 |
+
"eval_samples_per_second": 135.434,
|
| 1012 |
+
"eval_steps_per_second": 5.417,
|
| 1013 |
+
"step": 51960
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
+
"epoch": 20.12,
|
| 1017 |
+
"learning_rate": 0.0003660935762552391,
|
| 1018 |
+
"loss": 0.0049,
|
| 1019 |
+
"step": 52260
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1020 |
},
|
| 1021 |
{
|
| 1022 |
+
"epoch": 20.27,
|
| 1023 |
+
"learning_rate": 0.00036053374390556834,
|
| 1024 |
+
"loss": 0.0043,
|
| 1025 |
+
"step": 52650
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
+
"epoch": 20.42,
|
| 1029 |
+
"learning_rate": 0.0003549739115558977,
|
| 1030 |
+
"loss": 0.0043,
|
| 1031 |
+
"step": 53040
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
+
"epoch": 20.57,
|
| 1035 |
+
"learning_rate": 0.000349414079206227,
|
| 1036 |
+
"loss": 0.0044,
|
| 1037 |
+
"step": 53430
|
| 1038 |
},
|
| 1039 |
{
|
| 1040 |
+
"epoch": 20.72,
|
| 1041 |
+
"learning_rate": 0.0003438542468565563,
|
| 1042 |
+
"loss": 0.0044,
|
| 1043 |
+
"step": 53820
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 20.87,
|
| 1047 |
+
"learning_rate": 0.0003382944145068856,
|
| 1048 |
+
"loss": 0.0045,
|
| 1049 |
+
"step": 54210
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 21.0,
|
| 1053 |
+
"eval_loss": 0.0914614275097847,
|
| 1054 |
+
"eval_max_distance": 14,
|
| 1055 |
+
"eval_mean_distance": 1,
|
| 1056 |
+
"eval_runtime": 0.3653,
|
| 1057 |
+
"eval_samples_per_second": 136.874,
|
| 1058 |
+
"eval_steps_per_second": 5.475,
|
| 1059 |
+
"step": 54558
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
+
"epoch": 21.02,
|
| 1063 |
+
"learning_rate": 0.00033273458215721496,
|
| 1064 |
+
"loss": 0.0041,
|
| 1065 |
+
"step": 54600
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 21.17,
|
| 1069 |
+
"learning_rate": 0.00032717474980754424,
|
| 1070 |
"loss": 0.0035,
|
| 1071 |
+
"step": 54990
|
| 1072 |
},
|
| 1073 |
{
|
| 1074 |
+
"epoch": 21.32,
|
| 1075 |
+
"learning_rate": 0.0003216149174578736,
|
| 1076 |
+
"loss": 0.0038,
|
| 1077 |
+
"step": 55380
|
| 1078 |
},
|
| 1079 |
{
|
| 1080 |
+
"epoch": 21.47,
|
| 1081 |
+
"learning_rate": 0.0003160550851082029,
|
| 1082 |
+
"loss": 0.0038,
|
| 1083 |
+
"step": 55770
|
| 1084 |
},
|
| 1085 |
{
|
| 1086 |
+
"epoch": 21.62,
|
| 1087 |
+
"learning_rate": 0.0003104952527585322,
|
| 1088 |
+
"loss": 0.0041,
|
| 1089 |
+
"step": 56160
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1090 |
},
|
| 1091 |
{
|
| 1092 |
+
"epoch": 21.77,
|
| 1093 |
+
"learning_rate": 0.0003049354204088615,
|
| 1094 |
+
"loss": 0.004,
|
| 1095 |
+
"step": 56550
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
+
"epoch": 21.92,
|
| 1099 |
+
"learning_rate": 0.00029937558805919086,
|
| 1100 |
+
"loss": 0.0037,
|
| 1101 |
+
"step": 56940
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
+
"epoch": 22.0,
|
| 1105 |
+
"eval_loss": 0.09314610809087753,
|
| 1106 |
+
"eval_max_distance": 14,
|
| 1107 |
+
"eval_mean_distance": 1,
|
| 1108 |
+
"eval_runtime": 0.3634,
|
| 1109 |
+
"eval_samples_per_second": 137.604,
|
| 1110 |
+
"eval_steps_per_second": 5.504,
|
| 1111 |
+
"step": 57156
|
| 1112 |
},
|
| 1113 |
{
|
| 1114 |
+
"epoch": 22.07,
|
| 1115 |
+
"learning_rate": 0.00029381575570952014,
|
| 1116 |
+
"loss": 0.0037,
|
| 1117 |
+
"step": 57330
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1118 |
},
|
| 1119 |
{
|
| 1120 |
+
"epoch": 22.22,
|
| 1121 |
+
"learning_rate": 0.0002882559233598495,
|
| 1122 |
+
"loss": 0.0033,
|
| 1123 |
+
"step": 57720
|
| 1124 |
},
|
| 1125 |
{
|
| 1126 |
+
"epoch": 22.37,
|
| 1127 |
+
"learning_rate": 0.0002826960910101788,
|
| 1128 |
+
"loss": 0.0034,
|
| 1129 |
+
"step": 58110
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
+
"epoch": 22.52,
|
| 1133 |
+
"learning_rate": 0.0002771362586605081,
|
| 1134 |
+
"loss": 0.0034,
|
| 1135 |
+
"step": 58500
|
| 1136 |
},
|
| 1137 |
{
|
| 1138 |
+
"epoch": 22.67,
|
| 1139 |
+
"learning_rate": 0.0002715764263108374,
|
| 1140 |
+
"loss": 0.0035,
|
| 1141 |
+
"step": 58890
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1142 |
},
|
| 1143 |
{
|
| 1144 |
+
"epoch": 22.82,
|
| 1145 |
+
"learning_rate": 0.0002660165939611667,
|
| 1146 |
+
"loss": 0.0034,
|
| 1147 |
+
"step": 59280
|
| 1148 |
},
|
| 1149 |
{
|
| 1150 |
+
"epoch": 22.97,
|
| 1151 |
+
"learning_rate": 0.000260456761611496,
|
| 1152 |
+
"loss": 0.0036,
|
| 1153 |
+
"step": 59670
|
| 1154 |
},
|
| 1155 |
{
|
| 1156 |
+
"epoch": 23.0,
|
| 1157 |
+
"eval_loss": 0.09405915439128876,
|
| 1158 |
+
"eval_max_distance": 9,
|
| 1159 |
+
"eval_mean_distance": 1,
|
| 1160 |
+
"eval_runtime": 0.3715,
|
| 1161 |
+
"eval_samples_per_second": 134.573,
|
| 1162 |
+
"eval_steps_per_second": 5.383,
|
| 1163 |
+
"step": 59754
|
| 1164 |
},
|
| 1165 |
{
|
| 1166 |
+
"epoch": 23.12,
|
| 1167 |
+
"learning_rate": 0.0002548969292618253,
|
| 1168 |
+
"loss": 0.003,
|
| 1169 |
+
"step": 60060
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1170 |
},
|
| 1171 |
{
|
| 1172 |
+
"epoch": 23.27,
|
| 1173 |
+
"learning_rate": 0.00024933709691215465,
|
| 1174 |
+
"loss": 0.0031,
|
| 1175 |
+
"step": 60450
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
+
"epoch": 23.42,
|
| 1179 |
+
"learning_rate": 0.00024377726456248396,
|
| 1180 |
+
"loss": 0.003,
|
| 1181 |
+
"step": 60840
|
| 1182 |
},
|
| 1183 |
{
|
| 1184 |
+
"epoch": 23.57,
|
| 1185 |
+
"learning_rate": 0.00023821743221281327,
|
| 1186 |
+
"loss": 0.0029,
|
| 1187 |
+
"step": 61230
|
| 1188 |
},
|
| 1189 |
{
|
| 1190 |
+
"epoch": 23.72,
|
| 1191 |
+
"learning_rate": 0.00023265759986314258,
|
| 1192 |
+
"loss": 0.0028,
|
| 1193 |
+
"step": 61620
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1194 |
},
|
| 1195 |
{
|
| 1196 |
+
"epoch": 23.87,
|
| 1197 |
+
"learning_rate": 0.0002270977675134719,
|
| 1198 |
+
"loss": 0.0028,
|
| 1199 |
+
"step": 62010
|
| 1200 |
},
|
| 1201 |
{
|
| 1202 |
+
"epoch": 24.0,
|
| 1203 |
+
"eval_loss": 0.08611776679754257,
|
| 1204 |
+
"eval_max_distance": 13,
|
| 1205 |
+
"eval_mean_distance": 1,
|
| 1206 |
+
"eval_runtime": 0.3594,
|
| 1207 |
+
"eval_samples_per_second": 139.139,
|
| 1208 |
+
"eval_steps_per_second": 5.566,
|
| 1209 |
+
"step": 62352
|
| 1210 |
},
|
| 1211 |
{
|
| 1212 |
+
"epoch": 24.02,
|
| 1213 |
+
"learning_rate": 0.00022153793516380122,
|
| 1214 |
+
"loss": 0.0027,
|
| 1215 |
+
"step": 62400
|
| 1216 |
},
|
| 1217 |
{
|
| 1218 |
+
"epoch": 24.17,
|
| 1219 |
+
"learning_rate": 0.00021597810281413053,
|
| 1220 |
+
"loss": 0.0026,
|
| 1221 |
+
"step": 62790
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1222 |
},
|
| 1223 |
{
|
| 1224 |
+
"epoch": 24.32,
|
| 1225 |
+
"learning_rate": 0.00021041827046445986,
|
| 1226 |
+
"loss": 0.0027,
|
| 1227 |
+
"step": 63180
|
| 1228 |
},
|
| 1229 |
{
|
| 1230 |
+
"epoch": 24.47,
|
| 1231 |
+
"learning_rate": 0.00020485843811478917,
|
| 1232 |
+
"loss": 0.0027,
|
| 1233 |
+
"step": 63570
|
| 1234 |
},
|
| 1235 |
{
|
| 1236 |
+
"epoch": 24.62,
|
| 1237 |
+
"learning_rate": 0.00019929860576511847,
|
| 1238 |
+
"loss": 0.0027,
|
| 1239 |
+
"step": 63960
|
| 1240 |
},
|
| 1241 |
{
|
| 1242 |
+
"epoch": 24.77,
|
| 1243 |
+
"learning_rate": 0.00019373877341544778,
|
| 1244 |
+
"loss": 0.0024,
|
| 1245 |
+
"step": 64350
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1246 |
},
|
| 1247 |
{
|
| 1248 |
+
"epoch": 24.92,
|
| 1249 |
+
"learning_rate": 0.0001881789410657771,
|
| 1250 |
+
"loss": 0.0026,
|
| 1251 |
+
"step": 64740
|
| 1252 |
},
|
| 1253 |
{
|
| 1254 |
+
"epoch": 25.0,
|
| 1255 |
+
"eval_loss": 0.09115344285964966,
|
| 1256 |
+
"eval_max_distance": 12,
|
| 1257 |
+
"eval_mean_distance": 1,
|
| 1258 |
+
"eval_runtime": 0.3622,
|
| 1259 |
+
"eval_samples_per_second": 138.045,
|
| 1260 |
+
"eval_steps_per_second": 5.522,
|
| 1261 |
+
"step": 64950
|
| 1262 |
},
|
| 1263 |
{
|
| 1264 |
+
"epoch": 25.07,
|
| 1265 |
+
"learning_rate": 0.0001826191087161064,
|
| 1266 |
+
"loss": 0.0026,
|
| 1267 |
+
"step": 65130
|
| 1268 |
},
|
| 1269 |
{
|
| 1270 |
+
"epoch": 25.22,
|
| 1271 |
+
"learning_rate": 0.00017705927636643573,
|
| 1272 |
+
"loss": 0.0023,
|
| 1273 |
+
"step": 65520
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1274 |
},
|
| 1275 |
{
|
| 1276 |
+
"epoch": 25.37,
|
| 1277 |
+
"learning_rate": 0.00017149944401676504,
|
| 1278 |
+
"loss": 0.0023,
|
| 1279 |
+
"step": 65910
|
| 1280 |
},
|
| 1281 |
{
|
| 1282 |
+
"epoch": 25.52,
|
| 1283 |
+
"learning_rate": 0.00016593961166709435,
|
| 1284 |
+
"loss": 0.0021,
|
| 1285 |
+
"step": 66300
|
| 1286 |
},
|
| 1287 |
{
|
| 1288 |
+
"epoch": 25.67,
|
| 1289 |
+
"learning_rate": 0.00016037977931742368,
|
| 1290 |
+
"loss": 0.0021,
|
| 1291 |
+
"step": 66690
|
| 1292 |
},
|
| 1293 |
{
|
| 1294 |
+
"epoch": 25.82,
|
| 1295 |
+
"learning_rate": 0.000154819946967753,
|
| 1296 |
+
"loss": 0.0024,
|
| 1297 |
+
"step": 67080
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 25.97,
|
| 1301 |
+
"learning_rate": 0.0001492601146180823,
|
| 1302 |
+
"loss": 0.0024,
|
| 1303 |
+
"step": 67470
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 26.0,
|
| 1307 |
+
"eval_loss": 0.09158334881067276,
|
| 1308 |
+
"eval_max_distance": 9,
|
| 1309 |
"eval_mean_distance": 0,
|
| 1310 |
+
"eval_runtime": 0.3618,
|
| 1311 |
+
"eval_samples_per_second": 138.208,
|
| 1312 |
+
"eval_steps_per_second": 5.528,
|
| 1313 |
+
"step": 67548
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
+
"epoch": 26.12,
|
| 1317 |
+
"learning_rate": 0.0001437002822684116,
|
| 1318 |
+
"loss": 0.0021,
|
| 1319 |
+
"step": 67860
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 26.27,
|
| 1323 |
+
"learning_rate": 0.0001381404499187409,
|
| 1324 |
+
"loss": 0.0019,
|
| 1325 |
+
"step": 68250
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 26.42,
|
| 1329 |
+
"learning_rate": 0.00013258061756907022,
|
| 1330 |
+
"loss": 0.002,
|
| 1331 |
+
"step": 68640
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 26.57,
|
| 1335 |
+
"learning_rate": 0.00012702078521939955,
|
| 1336 |
+
"loss": 0.0019,
|
| 1337 |
+
"step": 69030
|
| 1338 |
},
|
| 1339 |
{
|
| 1340 |
+
"epoch": 26.72,
|
| 1341 |
+
"learning_rate": 0.00012146095286972886,
|
| 1342 |
+
"loss": 0.0021,
|
| 1343 |
+
"step": 69420
|
| 1344 |
},
|
| 1345 |
{
|
| 1346 |
+
"epoch": 26.87,
|
| 1347 |
+
"learning_rate": 0.00011590112052005817,
|
| 1348 |
+
"loss": 0.002,
|
| 1349 |
+
"step": 69810
|
| 1350 |
},
|
| 1351 |
{
|
| 1352 |
+
"epoch": 27.0,
|
| 1353 |
+
"eval_loss": 0.08878373354673386,
|
| 1354 |
+
"eval_max_distance": 9,
|
| 1355 |
"eval_mean_distance": 0,
|
| 1356 |
+
"eval_runtime": 0.3454,
|
| 1357 |
+
"eval_samples_per_second": 144.754,
|
| 1358 |
+
"eval_steps_per_second": 5.79,
|
| 1359 |
+
"step": 70146
|
| 1360 |
},
|
| 1361 |
{
|
| 1362 |
+
"epoch": 27.02,
|
| 1363 |
+
"learning_rate": 0.00011034128817038747,
|
| 1364 |
+
"loss": 0.0021,
|
| 1365 |
+
"step": 70200
|
| 1366 |
},
|
| 1367 |
{
|
| 1368 |
+
"epoch": 27.17,
|
| 1369 |
+
"learning_rate": 0.0001047814558207168,
|
| 1370 |
+
"loss": 0.0017,
|
| 1371 |
+
"step": 70590
|
| 1372 |
},
|
| 1373 |
{
|
| 1374 |
+
"epoch": 27.32,
|
| 1375 |
+
"learning_rate": 9.92216234710461e-05,
|
| 1376 |
+
"loss": 0.0018,
|
| 1377 |
+
"step": 70980
|
| 1378 |
},
|
| 1379 |
{
|
| 1380 |
+
"epoch": 27.47,
|
| 1381 |
+
"learning_rate": 9.366179112137542e-05,
|
| 1382 |
+
"loss": 0.0017,
|
| 1383 |
+
"step": 71370
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1384 |
},
|
| 1385 |
{
|
| 1386 |
+
"epoch": 27.62,
|
| 1387 |
+
"learning_rate": 8.810195877170473e-05,
|
| 1388 |
+
"loss": 0.0016,
|
| 1389 |
+
"step": 71760
|
| 1390 |
},
|
| 1391 |
{
|
| 1392 |
+
"epoch": 27.77,
|
| 1393 |
+
"learning_rate": 8.254212642203404e-05,
|
| 1394 |
+
"loss": 0.002,
|
| 1395 |
+
"step": 72150
|
| 1396 |
},
|
| 1397 |
{
|
| 1398 |
+
"epoch": 27.92,
|
| 1399 |
+
"learning_rate": 7.698229407236336e-05,
|
| 1400 |
+
"loss": 0.0017,
|
| 1401 |
+
"step": 72540
|
| 1402 |
},
|
| 1403 |
{
|
| 1404 |
+
"epoch": 28.0,
|
| 1405 |
+
"eval_loss": 0.08879587054252625,
|
| 1406 |
+
"eval_max_distance": 9,
|
| 1407 |
"eval_mean_distance": 0,
|
| 1408 |
+
"eval_runtime": 0.3476,
|
| 1409 |
+
"eval_samples_per_second": 143.846,
|
| 1410 |
+
"eval_steps_per_second": 5.754,
|
| 1411 |
+
"step": 72744
|
| 1412 |
},
|
| 1413 |
{
|
| 1414 |
+
"epoch": 28.07,
|
| 1415 |
+
"learning_rate": 7.142246172269268e-05,
|
| 1416 |
+
"loss": 0.0016,
|
| 1417 |
+
"step": 72930
|
| 1418 |
},
|
| 1419 |
{
|
| 1420 |
+
"epoch": 28.22,
|
| 1421 |
+
"learning_rate": 6.586262937302199e-05,
|
| 1422 |
+
"loss": 0.0016,
|
| 1423 |
+
"step": 73320
|
| 1424 |
},
|
| 1425 |
{
|
| 1426 |
+
"epoch": 28.37,
|
| 1427 |
+
"learning_rate": 6.03027970233513e-05,
|
| 1428 |
+
"loss": 0.0016,
|
| 1429 |
+
"step": 73710
|
| 1430 |
},
|
| 1431 |
{
|
| 1432 |
+
"epoch": 28.52,
|
| 1433 |
+
"learning_rate": 5.474296467368061e-05,
|
| 1434 |
+
"loss": 0.0016,
|
| 1435 |
+
"step": 74100
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1436 |
},
|
| 1437 |
{
|
| 1438 |
+
"epoch": 28.67,
|
| 1439 |
+
"learning_rate": 4.9183132324009924e-05,
|
| 1440 |
+
"loss": 0.0016,
|
| 1441 |
+
"step": 74490
|
| 1442 |
},
|
| 1443 |
{
|
| 1444 |
+
"epoch": 28.82,
|
| 1445 |
+
"learning_rate": 4.362329997433924e-05,
|
| 1446 |
+
"loss": 0.0017,
|
| 1447 |
+
"step": 74880
|
| 1448 |
},
|
| 1449 |
{
|
| 1450 |
+
"epoch": 28.97,
|
| 1451 |
+
"learning_rate": 3.806346762466855e-05,
|
| 1452 |
+
"loss": 0.0017,
|
| 1453 |
+
"step": 75270
|
| 1454 |
},
|
| 1455 |
{
|
| 1456 |
+
"epoch": 29.0,
|
| 1457 |
+
"eval_loss": 0.09515639394521713,
|
| 1458 |
+
"eval_max_distance": 9,
|
| 1459 |
"eval_mean_distance": 0,
|
| 1460 |
+
"eval_runtime": 0.343,
|
| 1461 |
+
"eval_samples_per_second": 145.752,
|
| 1462 |
+
"eval_steps_per_second": 5.83,
|
| 1463 |
+
"step": 75342
|
| 1464 |
},
|
| 1465 |
{
|
| 1466 |
+
"epoch": 29.12,
|
| 1467 |
+
"learning_rate": 3.250363527499786e-05,
|
| 1468 |
+
"loss": 0.0015,
|
| 1469 |
+
"step": 75660
|
| 1470 |
},
|
| 1471 |
{
|
| 1472 |
+
"epoch": 29.27,
|
| 1473 |
+
"learning_rate": 2.6943802925327177e-05,
|
| 1474 |
+
"loss": 0.0016,
|
| 1475 |
+
"step": 76050
|
| 1476 |
},
|
| 1477 |
{
|
| 1478 |
+
"epoch": 29.42,
|
| 1479 |
+
"learning_rate": 2.1383970575656488e-05,
|
| 1480 |
+
"loss": 0.0014,
|
| 1481 |
+
"step": 76440
|
| 1482 |
},
|
| 1483 |
{
|
| 1484 |
+
"epoch": 29.57,
|
| 1485 |
+
"learning_rate": 1.5824138225985802e-05,
|
| 1486 |
+
"loss": 0.0013,
|
| 1487 |
+
"step": 76830
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 29.72,
|
| 1491 |
+
"learning_rate": 1.0264305876315115e-05,
|
| 1492 |
+
"loss": 0.0014,
|
| 1493 |
+
"step": 77220
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 29.87,
|
| 1497 |
+
"learning_rate": 4.704473526644427e-06,
|
| 1498 |
+
"loss": 0.0014,
|
| 1499 |
+
"step": 77610
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 30.0,
|
| 1503 |
+
"eval_loss": 0.09847646951675415,
|
| 1504 |
+
"eval_max_distance": 9,
|
| 1505 |
"eval_mean_distance": 0,
|
| 1506 |
+
"eval_runtime": 0.3435,
|
| 1507 |
+
"eval_samples_per_second": 145.564,
|
| 1508 |
+
"eval_steps_per_second": 5.823,
|
| 1509 |
+
"step": 77940
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 30.0,
|
| 1513 |
+
"step": 77940,
|
| 1514 |
+
"total_flos": 4.517674593940685e+16,
|
| 1515 |
+
"train_loss": 0.053724035134690526,
|
| 1516 |
+
"train_runtime": 6582.4117,
|
| 1517 |
+
"train_samples_per_second": 355.137,
|
| 1518 |
+
"train_steps_per_second": 11.841
|
| 1519 |
}
|
| 1520 |
],
|
| 1521 |
+
"logging_steps": 390,
|
| 1522 |
+
"max_steps": 77940,
|
| 1523 |
+
"num_train_epochs": 30,
|
| 1524 |
+
"save_steps": 780,
|
| 1525 |
+
"total_flos": 4.517674593940685e+16,
|
| 1526 |
"trial_name": null,
|
| 1527 |
"trial_params": null
|
| 1528 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4091
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8559fb7fc2610f478f8fb2eefabd395825b089696477a52ed7cf7234f686c78f
|
| 3 |
size 4091
|