Model save
Browse files- README.md +19 -23
- adapter_model.safetensors +1 -1
- all_results.json +6 -11
- runs/Jun10_05-12-11_48ddfe8e991f/events.out.tfevents.1717996367.48ddfe8e991f.142677.0 +2 -2
- train_results.json +6 -6
- trainer_state.json +368 -644
README.md
CHANGED
|
@@ -2,13 +2,12 @@
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
base_model: google/gemma-2b
|
| 10 |
datasets:
|
| 11 |
-
-
|
| 12 |
model-index:
|
| 13 |
- name: gemma2b-summarize-gemini1_5flash-64k
|
| 14 |
results: []
|
|
@@ -19,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
|
| 20 |
# gemma2b-summarize-gemini1_5flash-64k
|
| 21 |
|
| 22 |
-
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
-
- Loss: 2.
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -41,14 +40,14 @@ More information needed
|
|
| 41 |
|
| 42 |
The following hyperparameters were used during training:
|
| 43 |
- learning_rate: 0.0002
|
| 44 |
-
- train_batch_size:
|
| 45 |
-
- eval_batch_size:
|
| 46 |
- seed: 42
|
| 47 |
- distributed_type: multi-GPU
|
| 48 |
- num_devices: 8
|
| 49 |
- gradient_accumulation_steps: 2
|
| 50 |
-
- total_train_batch_size:
|
| 51 |
-
- total_eval_batch_size:
|
| 52 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 53 |
- lr_scheduler_type: cosine
|
| 54 |
- lr_scheduler_warmup_ratio: 0.1
|
|
@@ -56,21 +55,18 @@ The following hyperparameters were used during training:
|
|
| 56 |
|
| 57 |
### Training results
|
| 58 |
|
| 59 |
-
| Training Loss | Epoch
|
| 60 |
-
|
| 61 |
-
| 1.
|
| 62 |
-
| 1.
|
| 63 |
-
| 1.
|
| 64 |
-
|
|
| 65 |
-
|
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.9208 | 10.9905 | 577 | 2.7079 |
|
| 72 |
-
| 0.9195 | 12.0 | 630 | 2.7148 |
|
| 73 |
-
| 0.9212 | 12.9905 | 682 | 2.7154 |
|
| 74 |
|
| 75 |
|
| 76 |
### Framework versions
|
|
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: google/gemma-2b
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: gemma2b-summarize-gemini1_5flash-64k
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# gemma2b-summarize-gemini1_5flash-64k
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 2.5156
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 40 |
|
| 41 |
The following hyperparameters were used during training:
|
| 42 |
- learning_rate: 0.0002
|
| 43 |
+
- train_batch_size: 8
|
| 44 |
+
- eval_batch_size: 8
|
| 45 |
- seed: 42
|
| 46 |
- distributed_type: multi-GPU
|
| 47 |
- num_devices: 8
|
| 48 |
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 128
|
| 50 |
+
- total_eval_batch_size: 64
|
| 51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
- lr_scheduler_type: cosine
|
| 53 |
- lr_scheduler_warmup_ratio: 0.1
|
|
|
|
| 55 |
|
| 56 |
### Training results
|
| 57 |
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 1.3387 | 0.9905 | 52 | 2.5711 |
|
| 61 |
+
| 1.0869 | 2.0 | 105 | 2.4883 |
|
| 62 |
+
| 1.0363 | 2.9905 | 157 | 2.4850 |
|
| 63 |
+
| 0.9967 | 4.0 | 210 | 2.4809 |
|
| 64 |
+
| 0.98 | 4.9905 | 262 | 2.4904 |
|
| 65 |
+
| 0.9574 | 6.0 | 315 | 2.4995 |
|
| 66 |
+
| 0.9392 | 6.9905 | 367 | 2.5062 |
|
| 67 |
+
| 0.9415 | 8.0 | 420 | 2.5143 |
|
| 68 |
+
| 0.9367 | 8.9905 | 472 | 2.5155 |
|
| 69 |
+
| 0.9348 | 9.9048 | 520 | 2.5156 |
|
|
|
|
|
|
|
|
|
|
| 70 |
|
| 71 |
|
| 72 |
### Framework versions
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 39256960
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4711620e971ab843c174762480b72259ca41cd9ba5e1123d1b7c569ba006191
|
| 3 |
size 39256960
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 24.482,
|
| 7 |
-
"eval_steps_per_second": 1.166,
|
| 8 |
-
"total_flos": 1.1018283868225536e+18,
|
| 9 |
-
"train_loss": 0.0,
|
| 10 |
-
"train_runtime": 3.5395,
|
| 11 |
"train_samples": 63353,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second":
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.904761904761905,
|
| 3 |
+
"total_flos": 8.18501087353897e+17,
|
| 4 |
+
"train_loss": 1.098491345002101,
|
| 5 |
+
"train_runtime": 2780.3493,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 63353,
|
| 7 |
+
"train_samples_per_second": 23.972,
|
| 8 |
+
"train_steps_per_second": 0.187
|
| 9 |
}
|
runs/Jun10_05-12-11_48ddfe8e991f/events.out.tfevents.1717996367.48ddfe8e991f.142677.0
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:ed27c036d85716b87752449c452c5de6cd3beb6bedfcc4b976658c6ff3aea995
|
| 3 |
+
size 30709
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 63353,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second":
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.904761904761905,
|
| 3 |
+
"total_flos": 8.18501087353897e+17,
|
| 4 |
+
"train_loss": 1.098491345002101,
|
| 5 |
+
"train_runtime": 2780.3493,
|
| 6 |
"train_samples": 63353,
|
| 7 |
+
"train_samples_per_second": 23.972,
|
| 8 |
+
"train_steps_per_second": 0.187
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,1112 +1,836 @@
|
|
| 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": 0.01904761904761905,
|
| 13 |
-
"grad_norm": 3.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 3.
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09523809523809523,
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate": 1.
|
| 22 |
-
"loss": 3.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.19047619047619047,
|
| 27 |
-
"grad_norm": 2.
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss":
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2857142857142857,
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.38095238095238093,
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 2.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.47619047619047616,
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.5714285714285714,
|
| 55 |
-
"grad_norm":
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss":
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.6666666666666666,
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.7619047619047619,
|
| 69 |
-
"grad_norm":
|
| 70 |
-
"learning_rate": 0.
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.8571428571428571,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
-
"learning_rate": 0.
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.9523809523809523,
|
| 83 |
-
"grad_norm":
|
| 84 |
-
"learning_rate": 0.
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.9904761904761905,
|
| 90 |
-
"eval_loss": 2.
|
| 91 |
-
"eval_runtime": 0.
|
| 92 |
-
"eval_samples_per_second":
|
| 93 |
-
"eval_steps_per_second":
|
| 94 |
"step": 52
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 1.0476190476190477,
|
| 98 |
-
"grad_norm":
|
| 99 |
-
"learning_rate": 0.
|
| 100 |
-
"loss": 1.
|
| 101 |
"step": 55
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 1.1428571428571428,
|
| 105 |
-
"grad_norm": 0.
|
| 106 |
-
"learning_rate": 0.
|
| 107 |
-
"loss": 1.
|
| 108 |
"step": 60
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 1.2380952380952381,
|
| 112 |
-
"grad_norm": 0.
|
| 113 |
-
"learning_rate": 0.
|
| 114 |
-
"loss": 1.
|
| 115 |
"step": 65
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 1.3333333333333333,
|
| 119 |
-
"grad_norm": 0.
|
| 120 |
-
"learning_rate": 0.
|
| 121 |
-
"loss": 1.
|
| 122 |
"step": 70
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 1.4285714285714286,
|
| 126 |
-
"grad_norm": 0.
|
| 127 |
-
"learning_rate": 0.
|
| 128 |
-
"loss": 1.
|
| 129 |
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 1.5238095238095237,
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
-
"learning_rate": 0.
|
| 135 |
-
"loss": 1.
|
| 136 |
"step": 80
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 1.619047619047619,
|
| 140 |
-
"grad_norm": 0.
|
| 141 |
-
"learning_rate": 0.
|
| 142 |
-
"loss": 1.
|
| 143 |
"step": 85
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 1.7142857142857144,
|
| 147 |
-
"grad_norm": 0.
|
| 148 |
-
"learning_rate": 0.
|
| 149 |
-
"loss": 1.
|
| 150 |
"step": 90
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 1.8095238095238095,
|
| 154 |
-
"grad_norm": 0.
|
| 155 |
-
"learning_rate": 0.
|
| 156 |
-
"loss": 1.
|
| 157 |
"step": 95
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 1.9047619047619047,
|
| 161 |
-
"grad_norm": 0.
|
| 162 |
-
"learning_rate": 0.
|
| 163 |
-
"loss": 1.
|
| 164 |
"step": 100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 2.0,
|
| 168 |
-
"grad_norm": 0.
|
| 169 |
-
"learning_rate": 0.
|
| 170 |
-
"loss": 1.
|
| 171 |
"step": 105
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 2.0,
|
| 175 |
-
"eval_loss": 2.
|
| 176 |
-
"eval_runtime": 0.
|
| 177 |
-
"eval_samples_per_second":
|
| 178 |
-
"eval_steps_per_second": 2.
|
| 179 |
"step": 105
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 2.0952380952380953,
|
| 183 |
-
"grad_norm": 0.
|
| 184 |
-
"learning_rate": 0.
|
| 185 |
-
"loss": 1.
|
| 186 |
"step": 110
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 2.1904761904761907,
|
| 190 |
-
"grad_norm": 0.
|
| 191 |
-
"learning_rate": 0.
|
| 192 |
-
"loss": 1.
|
| 193 |
"step": 115
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 2.2857142857142856,
|
| 197 |
-
"grad_norm": 0.
|
| 198 |
-
"learning_rate": 0.
|
| 199 |
-
"loss": 1.
|
| 200 |
"step": 120
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 2.380952380952381,
|
| 204 |
-
"grad_norm": 0.
|
| 205 |
-
"learning_rate": 0.
|
| 206 |
-
"loss": 1.
|
| 207 |
"step": 125
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 2.4761904761904763,
|
| 211 |
-
"grad_norm": 0.
|
| 212 |
-
"learning_rate": 0.
|
| 213 |
-
"loss": 1.
|
| 214 |
"step": 130
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 2.571428571428571,
|
| 218 |
-
"grad_norm": 0.
|
| 219 |
-
"learning_rate": 0.
|
| 220 |
-
"loss": 1.
|
| 221 |
"step": 135
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 2.6666666666666665,
|
| 225 |
-
"grad_norm": 0.
|
| 226 |
-
"learning_rate": 0.
|
| 227 |
-
"loss": 1.
|
| 228 |
"step": 140
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 2.761904761904762,
|
| 232 |
-
"grad_norm": 0.
|
| 233 |
-
"learning_rate": 0.
|
| 234 |
-
"loss": 1.
|
| 235 |
"step": 145
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 2.857142857142857,
|
| 239 |
-
"grad_norm": 0.
|
| 240 |
-
"learning_rate": 0.
|
| 241 |
-
"loss": 1.
|
| 242 |
"step": 150
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 2.9523809523809526,
|
| 246 |
-
"grad_norm": 0.
|
| 247 |
-
"learning_rate": 0.
|
| 248 |
-
"loss": 1.
|
| 249 |
"step": 155
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 2.9904761904761905,
|
| 253 |
-
"eval_loss": 2.
|
| 254 |
-
"eval_runtime": 0.
|
| 255 |
-
"eval_samples_per_second":
|
| 256 |
-
"eval_steps_per_second":
|
| 257 |
"step": 157
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 3.0476190476190474,
|
| 261 |
-
"grad_norm": 0.
|
| 262 |
-
"learning_rate": 0.
|
| 263 |
-
"loss": 1.
|
| 264 |
"step": 160
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 3.142857142857143,
|
| 268 |
-
"grad_norm": 0.
|
| 269 |
-
"learning_rate": 0.
|
| 270 |
-
"loss": 1.
|
| 271 |
"step": 165
|
| 272 |
},
|
| 273 |
{
|
| 274 |
"epoch": 3.238095238095238,
|
| 275 |
-
"grad_norm": 0.
|
| 276 |
-
"learning_rate": 0.
|
| 277 |
-
"loss": 1.
|
| 278 |
"step": 170
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 3.3333333333333335,
|
| 282 |
-
"grad_norm": 0.
|
| 283 |
-
"learning_rate": 0.
|
| 284 |
-
"loss": 1.
|
| 285 |
"step": 175
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 3.4285714285714284,
|
| 289 |
-
"grad_norm": 0.
|
| 290 |
-
"learning_rate": 0.
|
| 291 |
-
"loss": 1.
|
| 292 |
"step": 180
|
| 293 |
},
|
| 294 |
{
|
| 295 |
"epoch": 3.5238095238095237,
|
| 296 |
-
"grad_norm":
|
| 297 |
-
"learning_rate": 0.
|
| 298 |
-
"loss": 1.
|
| 299 |
"step": 185
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 3.619047619047619,
|
| 303 |
-
"grad_norm": 0.
|
| 304 |
-
"learning_rate": 0.
|
| 305 |
-
"loss": 1.
|
| 306 |
"step": 190
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 3.7142857142857144,
|
| 310 |
-
"grad_norm": 0.
|
| 311 |
-
"learning_rate": 0.
|
| 312 |
-
"loss": 1.
|
| 313 |
"step": 195
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 3.8095238095238093,
|
| 317 |
-
"grad_norm": 0.
|
| 318 |
-
"learning_rate": 0.
|
| 319 |
-
"loss":
|
| 320 |
"step": 200
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 3.9047619047619047,
|
| 324 |
-
"grad_norm": 0.
|
| 325 |
-
"learning_rate": 0.
|
| 326 |
-
"loss": 1.
|
| 327 |
"step": 205
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 4.0,
|
| 331 |
-
"grad_norm": 0.
|
| 332 |
-
"learning_rate": 0.
|
| 333 |
-
"loss":
|
| 334 |
"step": 210
|
| 335 |
},
|
| 336 |
{
|
| 337 |
"epoch": 4.0,
|
| 338 |
-
"eval_loss": 2.
|
| 339 |
-
"eval_runtime": 0.
|
| 340 |
-
"eval_samples_per_second":
|
| 341 |
-
"eval_steps_per_second": 2.
|
| 342 |
"step": 210
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 4.095238095238095,
|
| 346 |
-
"grad_norm": 0.
|
| 347 |
-
"learning_rate": 0.
|
| 348 |
-
"loss":
|
| 349 |
"step": 215
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 4.190476190476191,
|
| 353 |
-
"grad_norm": 0.
|
| 354 |
-
"learning_rate": 0.
|
| 355 |
-
"loss":
|
| 356 |
"step": 220
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 4.285714285714286,
|
| 360 |
-
"grad_norm": 0.
|
| 361 |
-
"learning_rate": 0.
|
| 362 |
-
"loss":
|
| 363 |
"step": 225
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 4.380952380952381,
|
| 367 |
-
"grad_norm": 0.
|
| 368 |
-
"learning_rate": 0.
|
| 369 |
-
"loss":
|
| 370 |
"step": 230
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 4.476190476190476,
|
| 374 |
-
"grad_norm": 0.
|
| 375 |
-
"learning_rate": 0.
|
| 376 |
-
"loss":
|
| 377 |
"step": 235
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 4.571428571428571,
|
| 381 |
-
"grad_norm": 0.
|
| 382 |
-
"learning_rate": 0.
|
| 383 |
-
"loss":
|
| 384 |
"step": 240
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 4.666666666666667,
|
| 388 |
-
"grad_norm": 0.
|
| 389 |
-
"learning_rate": 0.
|
| 390 |
-
"loss":
|
| 391 |
"step": 245
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 4.761904761904762,
|
| 395 |
-
"grad_norm": 0.
|
| 396 |
-
"learning_rate": 0.
|
| 397 |
-
"loss":
|
| 398 |
"step": 250
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 4.857142857142857,
|
| 402 |
-
"grad_norm": 0.
|
| 403 |
-
"learning_rate": 0.
|
| 404 |
-
"loss":
|
| 405 |
"step": 255
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 4.9523809523809526,
|
| 409 |
-
"grad_norm": 0.
|
| 410 |
-
"learning_rate": 0.
|
| 411 |
-
"loss":
|
| 412 |
"step": 260
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 4.9904761904761905,
|
| 416 |
-
"eval_loss": 2.
|
| 417 |
-
"eval_runtime": 0.
|
| 418 |
-
"eval_samples_per_second":
|
| 419 |
-
"eval_steps_per_second":
|
| 420 |
"step": 262
|
| 421 |
},
|
| 422 |
{
|
| 423 |
"epoch": 5.0476190476190474,
|
| 424 |
-
"grad_norm": 0.
|
| 425 |
-
"learning_rate": 0.
|
| 426 |
-
"loss": 0.
|
| 427 |
"step": 265
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 5.142857142857143,
|
| 431 |
-
"grad_norm": 0.
|
| 432 |
-
"learning_rate": 0.
|
| 433 |
-
"loss": 0.
|
| 434 |
"step": 270
|
| 435 |
},
|
| 436 |
{
|
| 437 |
"epoch": 5.238095238095238,
|
| 438 |
-
"grad_norm": 0.
|
| 439 |
-
"learning_rate": 0.
|
| 440 |
-
"loss": 0.
|
| 441 |
"step": 275
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 5.333333333333333,
|
| 445 |
-
"grad_norm": 0.
|
| 446 |
-
"learning_rate": 0.
|
| 447 |
-
"loss": 0.
|
| 448 |
"step": 280
|
| 449 |
},
|
| 450 |
{
|
| 451 |
"epoch": 5.428571428571429,
|
| 452 |
-
"grad_norm": 0.
|
| 453 |
-
"learning_rate": 0.
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 285
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 5.523809523809524,
|
| 459 |
-
"grad_norm": 0.
|
| 460 |
-
"learning_rate":
|
| 461 |
-
"loss": 0.
|
| 462 |
"step": 290
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 5.619047619047619,
|
| 466 |
-
"grad_norm": 0.
|
| 467 |
-
"learning_rate":
|
| 468 |
-
"loss": 0.
|
| 469 |
"step": 295
|
| 470 |
},
|
| 471 |
{
|
| 472 |
"epoch": 5.714285714285714,
|
| 473 |
-
"grad_norm": 0.
|
| 474 |
-
"learning_rate":
|
| 475 |
-
"loss": 0.
|
| 476 |
"step": 300
|
| 477 |
},
|
| 478 |
{
|
| 479 |
"epoch": 5.809523809523809,
|
| 480 |
-
"grad_norm": 0.
|
| 481 |
-
"learning_rate":
|
| 482 |
-
"loss": 0.
|
| 483 |
"step": 305
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 5.904761904761905,
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
-
"learning_rate":
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 310
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 6.0,
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
-
"learning_rate":
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 315
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 6.0,
|
| 501 |
-
"eval_loss": 2.
|
| 502 |
-
"eval_runtime": 0.
|
| 503 |
-
"eval_samples_per_second":
|
| 504 |
-
"eval_steps_per_second": 2.
|
| 505 |
"step": 315
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 6.095238095238095,
|
| 509 |
-
"grad_norm": 0.
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 320
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 6.190476190476191,
|
| 516 |
-
"grad_norm": 0.
|
| 517 |
-
"learning_rate":
|
| 518 |
-
"loss": 0.
|
| 519 |
"step": 325
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 6.285714285714286,
|
| 523 |
-
"grad_norm": 0.
|
| 524 |
-
"learning_rate":
|
| 525 |
-
"loss": 0.
|
| 526 |
"step": 330
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 6.380952380952381,
|
| 530 |
-
"grad_norm": 0.
|
| 531 |
-
"learning_rate":
|
| 532 |
-
"loss": 0.
|
| 533 |
"step": 335
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 6.476190476190476,
|
| 537 |
-
"grad_norm": 0.
|
| 538 |
-
"learning_rate":
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 340
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 6.571428571428571,
|
| 544 |
-
"grad_norm": 0.
|
| 545 |
-
"learning_rate":
|
| 546 |
-
"loss": 0.
|
| 547 |
"step": 345
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 6.666666666666667,
|
| 551 |
-
"grad_norm": 0.
|
| 552 |
-
"learning_rate":
|
| 553 |
-
"loss": 0.
|
| 554 |
"step": 350
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 6.761904761904762,
|
| 558 |
-
"grad_norm": 0.
|
| 559 |
-
"learning_rate":
|
| 560 |
-
"loss": 0.
|
| 561 |
"step": 355
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 6.857142857142857,
|
| 565 |
-
"grad_norm": 0.
|
| 566 |
-
"learning_rate":
|
| 567 |
-
"loss": 0.
|
| 568 |
"step": 360
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 6.9523809523809526,
|
| 572 |
-
"grad_norm": 0.
|
| 573 |
-
"learning_rate":
|
| 574 |
-
"loss": 0.
|
| 575 |
"step": 365
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 6.9904761904761905,
|
| 579 |
-
"eval_loss": 2.
|
| 580 |
-
"eval_runtime": 0.
|
| 581 |
-
"eval_samples_per_second":
|
| 582 |
-
"eval_steps_per_second": 1.
|
| 583 |
"step": 367
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 7.0476190476190474,
|
| 587 |
-
"grad_norm": 0.
|
| 588 |
-
"learning_rate":
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 370
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 7.142857142857143,
|
| 594 |
-
"grad_norm": 0.
|
| 595 |
-
"learning_rate":
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 375
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 7.238095238095238,
|
| 601 |
-
"grad_norm": 0.
|
| 602 |
-
"learning_rate":
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 380
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 7.333333333333333,
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 385
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 7.428571428571429,
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
-
"learning_rate":
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 390
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 7.523809523809524,
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
-
"learning_rate":
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 395
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 7.619047619047619,
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
-
"learning_rate":
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 400
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 7.714285714285714,
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
-
"learning_rate":
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 405
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 7.809523809523809,
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
-
"learning_rate":
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 410
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 7.904761904761905,
|
| 650 |
-
"grad_norm": 0.
|
| 651 |
-
"learning_rate":
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 415
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 8.0,
|
| 657 |
-
"grad_norm": 0.
|
| 658 |
-
"learning_rate":
|
| 659 |
-
"loss": 0.
|
| 660 |
"step": 420
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 8.0,
|
| 664 |
-
"eval_loss": 2.
|
| 665 |
-
"eval_runtime": 0.
|
| 666 |
-
"eval_samples_per_second":
|
| 667 |
-
"eval_steps_per_second": 2.
|
| 668 |
"step": 420
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 8.095238095238095,
|
| 672 |
-
"grad_norm": 0.
|
| 673 |
-
"learning_rate":
|
| 674 |
-
"loss": 0.
|
| 675 |
"step": 425
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 8.19047619047619,
|
| 679 |
-
"grad_norm": 0.
|
| 680 |
-
"learning_rate":
|
| 681 |
-
"loss": 0.
|
| 682 |
"step": 430
|
| 683 |
},
|
| 684 |
{
|
| 685 |
"epoch": 8.285714285714286,
|
| 686 |
-
"grad_norm": 0.
|
| 687 |
-
"learning_rate":
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 435
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 8.380952380952381,
|
| 693 |
-
"grad_norm": 0.
|
| 694 |
-
"learning_rate":
|
| 695 |
-
"loss": 0.
|
| 696 |
"step": 440
|
| 697 |
},
|
| 698 |
{
|
| 699 |
"epoch": 8.476190476190476,
|
| 700 |
-
"grad_norm": 0.
|
| 701 |
-
"learning_rate":
|
| 702 |
-
"loss": 0.
|
| 703 |
"step": 445
|
| 704 |
},
|
| 705 |
{
|
| 706 |
"epoch": 8.571428571428571,
|
| 707 |
-
"grad_norm": 0.
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 0.
|
| 710 |
"step": 450
|
| 711 |
},
|
| 712 |
{
|
| 713 |
"epoch": 8.666666666666666,
|
| 714 |
-
"grad_norm": 0.
|
| 715 |
-
"learning_rate":
|
| 716 |
-
"loss": 0.
|
| 717 |
"step": 455
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 8.761904761904763,
|
| 721 |
-
"grad_norm": 0.
|
| 722 |
-
"learning_rate": 8.
|
| 723 |
-
"loss": 0.
|
| 724 |
"step": 460
|
| 725 |
},
|
| 726 |
{
|
| 727 |
"epoch": 8.857142857142858,
|
| 728 |
-
"grad_norm": 0.
|
| 729 |
-
"learning_rate":
|
| 730 |
-
"loss": 0.
|
| 731 |
"step": 465
|
| 732 |
},
|
| 733 |
{
|
| 734 |
"epoch": 8.952380952380953,
|
| 735 |
-
"grad_norm": 0.
|
| 736 |
-
"learning_rate":
|
| 737 |
-
"loss": 0.
|
| 738 |
"step": 470
|
| 739 |
},
|
| 740 |
{
|
| 741 |
"epoch": 8.99047619047619,
|
| 742 |
-
"eval_loss": 2.
|
| 743 |
-
"eval_runtime": 0.
|
| 744 |
-
"eval_samples_per_second":
|
| 745 |
-
"eval_steps_per_second": 1.
|
| 746 |
"step": 472
|
| 747 |
},
|
| 748 |
{
|
| 749 |
"epoch": 9.047619047619047,
|
| 750 |
-
"grad_norm": 0.
|
| 751 |
-
"learning_rate":
|
| 752 |
-
"loss": 0.
|
| 753 |
"step": 475
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 9.142857142857142,
|
| 757 |
-
"grad_norm": 0.
|
| 758 |
-
"learning_rate":
|
| 759 |
-
"loss": 0.
|
| 760 |
"step": 480
|
| 761 |
},
|
| 762 |
{
|
| 763 |
"epoch": 9.238095238095237,
|
| 764 |
-
"grad_norm": 0.
|
| 765 |
-
"learning_rate":
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 485
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 9.333333333333334,
|
| 771 |
-
"grad_norm": 0.
|
| 772 |
-
"learning_rate":
|
| 773 |
-
"loss": 0.
|
| 774 |
"step": 490
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 9.428571428571429,
|
| 778 |
-
"grad_norm": 0.
|
| 779 |
-
"learning_rate":
|
| 780 |
-
"loss": 0.
|
| 781 |
"step": 495
|
| 782 |
},
|
| 783 |
{
|
| 784 |
"epoch": 9.523809523809524,
|
| 785 |
-
"grad_norm": 0.
|
| 786 |
-
"learning_rate":
|
| 787 |
-
"loss": 0.
|
| 788 |
"step": 500
|
| 789 |
},
|
| 790 |
{
|
| 791 |
"epoch": 9.619047619047619,
|
| 792 |
-
"grad_norm": 0.
|
| 793 |
-
"learning_rate":
|
| 794 |
-
"loss": 0.
|
| 795 |
"step": 505
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 9.714285714285714,
|
| 799 |
-
"grad_norm": 0.
|
| 800 |
-
"learning_rate":
|
| 801 |
-
"loss": 0.
|
| 802 |
"step": 510
|
| 803 |
},
|
| 804 |
{
|
| 805 |
"epoch": 9.80952380952381,
|
| 806 |
-
"grad_norm": 0.
|
| 807 |
-
"learning_rate":
|
| 808 |
-
"loss": 0.
|
| 809 |
"step": 515
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 9.904761904761905,
|
| 813 |
-
"grad_norm": 0.
|
| 814 |
-
"learning_rate":
|
| 815 |
-
"loss": 0.
|
| 816 |
"step": 520
|
| 817 |
},
|
| 818 |
{
|
| 819 |
-
"epoch":
|
| 820 |
-
"
|
| 821 |
-
"
|
| 822 |
-
"
|
| 823 |
-
"
|
| 824 |
-
|
| 825 |
-
{
|
| 826 |
-
"epoch": 10.0,
|
| 827 |
-
"eval_loss": 2.6975326538085938,
|
| 828 |
-
"eval_runtime": 0.485,
|
| 829 |
-
"eval_samples_per_second": 43.303,
|
| 830 |
-
"eval_steps_per_second": 2.062,
|
| 831 |
-
"step": 525
|
| 832 |
-
},
|
| 833 |
-
{
|
| 834 |
-
"epoch": 10.095238095238095,
|
| 835 |
-
"grad_norm": 0.30078125,
|
| 836 |
-
"learning_rate": 5.6323846910828735e-05,
|
| 837 |
-
"loss": 0.9233,
|
| 838 |
-
"step": 530
|
| 839 |
-
},
|
| 840 |
-
{
|
| 841 |
-
"epoch": 10.19047619047619,
|
| 842 |
-
"grad_norm": 0.265625,
|
| 843 |
-
"learning_rate": 5.432205277496327e-05,
|
| 844 |
-
"loss": 0.9235,
|
| 845 |
-
"step": 535
|
| 846 |
-
},
|
| 847 |
-
{
|
| 848 |
-
"epoch": 10.285714285714286,
|
| 849 |
-
"grad_norm": 0.2734375,
|
| 850 |
-
"learning_rate": 5.234312799786921e-05,
|
| 851 |
-
"loss": 0.9194,
|
| 852 |
-
"step": 540
|
| 853 |
-
},
|
| 854 |
-
{
|
| 855 |
-
"epoch": 10.380952380952381,
|
| 856 |
-
"grad_norm": 0.29296875,
|
| 857 |
-
"learning_rate": 5.0388063358324134e-05,
|
| 858 |
-
"loss": 0.9235,
|
| 859 |
-
"step": 545
|
| 860 |
-
},
|
| 861 |
-
{
|
| 862 |
-
"epoch": 10.476190476190476,
|
| 863 |
-
"grad_norm": 0.279296875,
|
| 864 |
-
"learning_rate": 4.845783768916482e-05,
|
| 865 |
-
"loss": 0.9217,
|
| 866 |
-
"step": 550
|
| 867 |
-
},
|
| 868 |
-
{
|
| 869 |
-
"epoch": 10.571428571428571,
|
| 870 |
-
"grad_norm": 0.265625,
|
| 871 |
-
"learning_rate": 4.6553417387219886e-05,
|
| 872 |
-
"loss": 0.9243,
|
| 873 |
-
"step": 555
|
| 874 |
-
},
|
| 875 |
-
{
|
| 876 |
-
"epoch": 10.666666666666666,
|
| 877 |
-
"grad_norm": 0.306640625,
|
| 878 |
-
"learning_rate": 4.467575592946864e-05,
|
| 879 |
-
"loss": 0.9262,
|
| 880 |
-
"step": 560
|
| 881 |
-
},
|
| 882 |
-
{
|
| 883 |
-
"epoch": 10.761904761904763,
|
| 884 |
-
"grad_norm": 0.255859375,
|
| 885 |
-
"learning_rate": 4.282579339566802e-05,
|
| 886 |
-
"loss": 0.9206,
|
| 887 |
-
"step": 565
|
| 888 |
-
},
|
| 889 |
-
{
|
| 890 |
-
"epoch": 10.857142857142858,
|
| 891 |
-
"grad_norm": 0.2578125,
|
| 892 |
-
"learning_rate": 4.100445599768774e-05,
|
| 893 |
-
"loss": 0.9256,
|
| 894 |
-
"step": 570
|
| 895 |
-
},
|
| 896 |
-
{
|
| 897 |
-
"epoch": 10.952380952380953,
|
| 898 |
-
"grad_norm": 0.265625,
|
| 899 |
-
"learning_rate": 3.9212655615787804e-05,
|
| 900 |
-
"loss": 0.9208,
|
| 901 |
-
"step": 575
|
| 902 |
-
},
|
| 903 |
-
{
|
| 904 |
-
"epoch": 10.99047619047619,
|
| 905 |
-
"eval_loss": 2.707897663116455,
|
| 906 |
-
"eval_runtime": 0.647,
|
| 907 |
-
"eval_samples_per_second": 32.456,
|
| 908 |
-
"eval_steps_per_second": 1.546,
|
| 909 |
-
"step": 577
|
| 910 |
-
},
|
| 911 |
-
{
|
| 912 |
-
"epoch": 11.047619047619047,
|
| 913 |
-
"grad_norm": 0.25,
|
| 914 |
-
"learning_rate": 3.745128934207225e-05,
|
| 915 |
-
"loss": 0.9226,
|
| 916 |
-
"step": 580
|
| 917 |
-
},
|
| 918 |
-
{
|
| 919 |
-
"epoch": 11.142857142857142,
|
| 920 |
-
"grad_norm": 0.263671875,
|
| 921 |
-
"learning_rate": 3.5721239031346066e-05,
|
| 922 |
-
"loss": 0.9121,
|
| 923 |
-
"step": 585
|
| 924 |
-
},
|
| 925 |
-
{
|
| 926 |
-
"epoch": 11.238095238095237,
|
| 927 |
-
"grad_norm": 0.259765625,
|
| 928 |
-
"learning_rate": 3.402337085960119e-05,
|
| 929 |
-
"loss": 0.9155,
|
| 930 |
-
"step": 590
|
| 931 |
-
},
|
| 932 |
-
{
|
| 933 |
-
"epoch": 11.333333333333334,
|
| 934 |
-
"grad_norm": 0.2578125,
|
| 935 |
-
"learning_rate": 3.235853489035241e-05,
|
| 936 |
-
"loss": 0.9194,
|
| 937 |
-
"step": 595
|
| 938 |
-
},
|
| 939 |
-
{
|
| 940 |
-
"epoch": 11.428571428571429,
|
| 941 |
-
"grad_norm": 0.26953125,
|
| 942 |
-
"learning_rate": 3.072756464904006e-05,
|
| 943 |
-
"loss": 0.9202,
|
| 944 |
-
"step": 600
|
| 945 |
-
},
|
| 946 |
-
{
|
| 947 |
-
"epoch": 11.523809523809524,
|
| 948 |
-
"grad_norm": 0.27734375,
|
| 949 |
-
"learning_rate": 2.9131276705713006e-05,
|
| 950 |
-
"loss": 0.9186,
|
| 951 |
-
"step": 605
|
| 952 |
-
},
|
| 953 |
-
{
|
| 954 |
-
"epoch": 11.619047619047619,
|
| 955 |
-
"grad_norm": 0.259765625,
|
| 956 |
-
"learning_rate": 2.7570470266200176e-05,
|
| 957 |
-
"loss": 0.9195,
|
| 958 |
-
"step": 610
|
| 959 |
-
},
|
| 960 |
-
{
|
| 961 |
-
"epoch": 11.714285714285714,
|
| 962 |
-
"grad_norm": 0.2470703125,
|
| 963 |
-
"learning_rate": 2.6045926771976303e-05,
|
| 964 |
-
"loss": 0.9223,
|
| 965 |
-
"step": 615
|
| 966 |
-
},
|
| 967 |
-
{
|
| 968 |
-
"epoch": 11.80952380952381,
|
| 969 |
-
"grad_norm": 0.25,
|
| 970 |
-
"learning_rate": 2.4558409508920986e-05,
|
| 971 |
-
"loss": 0.9139,
|
| 972 |
-
"step": 620
|
| 973 |
-
},
|
| 974 |
-
{
|
| 975 |
-
"epoch": 11.904761904761905,
|
| 976 |
-
"grad_norm": 0.248046875,
|
| 977 |
-
"learning_rate": 2.3108663225168435e-05,
|
| 978 |
-
"loss": 0.9117,
|
| 979 |
-
"step": 625
|
| 980 |
-
},
|
| 981 |
-
{
|
| 982 |
-
"epoch": 12.0,
|
| 983 |
-
"grad_norm": 0.263671875,
|
| 984 |
-
"learning_rate": 2.1697413758237784e-05,
|
| 985 |
-
"loss": 0.9195,
|
| 986 |
-
"step": 630
|
| 987 |
-
},
|
| 988 |
-
{
|
| 989 |
-
"epoch": 12.0,
|
| 990 |
-
"eval_loss": 2.714789867401123,
|
| 991 |
-
"eval_runtime": 0.4906,
|
| 992 |
-
"eval_samples_per_second": 42.808,
|
| 993 |
-
"eval_steps_per_second": 2.038,
|
| 994 |
-
"step": 630
|
| 995 |
-
},
|
| 996 |
-
{
|
| 997 |
-
"epoch": 12.095238095238095,
|
| 998 |
-
"grad_norm": 0.2392578125,
|
| 999 |
-
"learning_rate": 2.032536767163141e-05,
|
| 1000 |
-
"loss": 0.9167,
|
| 1001 |
-
"step": 635
|
| 1002 |
-
},
|
| 1003 |
-
{
|
| 1004 |
-
"epoch": 12.19047619047619,
|
| 1005 |
-
"grad_norm": 0.25,
|
| 1006 |
-
"learning_rate": 1.8993211901083353e-05,
|
| 1007 |
-
"loss": 0.9144,
|
| 1008 |
-
"step": 640
|
| 1009 |
-
},
|
| 1010 |
-
{
|
| 1011 |
-
"epoch": 12.285714285714286,
|
| 1012 |
-
"grad_norm": 0.2578125,
|
| 1013 |
-
"learning_rate": 1.7701613410634365e-05,
|
| 1014 |
-
"loss": 0.915,
|
| 1015 |
-
"step": 645
|
| 1016 |
-
},
|
| 1017 |
-
{
|
| 1018 |
-
"epoch": 12.380952380952381,
|
| 1019 |
-
"grad_norm": 0.248046875,
|
| 1020 |
-
"learning_rate": 1.6451218858706374e-05,
|
| 1021 |
-
"loss": 0.9153,
|
| 1022 |
-
"step": 650
|
| 1023 |
-
},
|
| 1024 |
-
{
|
| 1025 |
-
"epoch": 12.476190476190476,
|
| 1026 |
-
"grad_norm": 0.251953125,
|
| 1027 |
-
"learning_rate": 1.5242654274342894e-05,
|
| 1028 |
-
"loss": 0.9133,
|
| 1029 |
-
"step": 655
|
| 1030 |
-
},
|
| 1031 |
-
{
|
| 1032 |
-
"epoch": 12.571428571428571,
|
| 1033 |
-
"grad_norm": 0.25,
|
| 1034 |
-
"learning_rate": 1.4076524743778319e-05,
|
| 1035 |
-
"loss": 0.9134,
|
| 1036 |
-
"step": 660
|
| 1037 |
-
},
|
| 1038 |
-
{
|
| 1039 |
-
"epoch": 12.666666666666666,
|
| 1040 |
-
"grad_norm": 0.248046875,
|
| 1041 |
-
"learning_rate": 1.295341410749208e-05,
|
| 1042 |
-
"loss": 0.9126,
|
| 1043 |
-
"step": 665
|
| 1044 |
-
},
|
| 1045 |
-
{
|
| 1046 |
-
"epoch": 12.761904761904763,
|
| 1047 |
-
"grad_norm": 0.25,
|
| 1048 |
-
"learning_rate": 1.1873884667900125e-05,
|
| 1049 |
-
"loss": 0.9094,
|
| 1050 |
-
"step": 670
|
| 1051 |
-
},
|
| 1052 |
-
{
|
| 1053 |
-
"epoch": 12.857142857142858,
|
| 1054 |
-
"grad_norm": 0.236328125,
|
| 1055 |
-
"learning_rate": 1.083847690782972e-05,
|
| 1056 |
-
"loss": 0.9183,
|
| 1057 |
-
"step": 675
|
| 1058 |
-
},
|
| 1059 |
-
{
|
| 1060 |
-
"epoch": 12.952380952380953,
|
| 1061 |
-
"grad_norm": 0.255859375,
|
| 1062 |
-
"learning_rate": 9.8477092199184e-06,
|
| 1063 |
-
"loss": 0.9212,
|
| 1064 |
-
"step": 680
|
| 1065 |
-
},
|
| 1066 |
-
{
|
| 1067 |
-
"epoch": 12.99047619047619,
|
| 1068 |
-
"eval_loss": 2.7153878211975098,
|
| 1069 |
-
"eval_runtime": 0.5024,
|
| 1070 |
-
"eval_samples_per_second": 41.796,
|
| 1071 |
-
"eval_steps_per_second": 1.99,
|
| 1072 |
-
"step": 682
|
| 1073 |
-
},
|
| 1074 |
-
{
|
| 1075 |
-
"epoch": 13.047619047619047,
|
| 1076 |
-
"grad_norm": 0.2431640625,
|
| 1077 |
-
"learning_rate": 8.902077647072881e-06,
|
| 1078 |
-
"loss": 0.9139,
|
| 1079 |
-
"step": 685
|
| 1080 |
-
},
|
| 1081 |
-
{
|
| 1082 |
-
"epoch": 13.142857142857142,
|
| 1083 |
-
"grad_norm": 0.2431640625,
|
| 1084 |
-
"learning_rate": 8.002055634117578e-06,
|
| 1085 |
-
"loss": 0.9123,
|
| 1086 |
-
"step": 690
|
| 1087 |
-
},
|
| 1088 |
-
{
|
| 1089 |
-
"epoch": 13.238095238095237,
|
| 1090 |
-
"grad_norm": 0.2470703125,
|
| 1091 |
-
"learning_rate": 7.148093790757371e-06,
|
| 1092 |
-
"loss": 0.9152,
|
| 1093 |
-
"step": 695
|
| 1094 |
},
|
| 1095 |
{
|
| 1096 |
-
"epoch":
|
| 1097 |
-
"
|
| 1098 |
-
"
|
| 1099 |
-
"
|
| 1100 |
-
"
|
| 1101 |
-
|
| 1102 |
-
|
| 1103 |
-
"epoch": 13.333333333333334,
|
| 1104 |
-
"step": 700,
|
| 1105 |
-
"total_flos": 1.1018283868225536e+18,
|
| 1106 |
-
"train_loss": 0.0,
|
| 1107 |
-
"train_runtime": 3.5395,
|
| 1108 |
-
"train_samples_per_second": 37654.826,
|
| 1109 |
-
"train_steps_per_second": 146.913
|
| 1110 |
}
|
| 1111 |
],
|
| 1112 |
"logging_steps": 5,
|
|
@@ -1126,8 +850,8 @@
|
|
| 1126 |
"attributes": {}
|
| 1127 |
}
|
| 1128 |
},
|
| 1129 |
-
"total_flos":
|
| 1130 |
-
"train_batch_size":
|
| 1131 |
"trial_name": null,
|
| 1132 |
"trial_params": null
|
| 1133 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.904761904761905,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 520,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01904761904761905,
|
| 13 |
+
"grad_norm": 3.328125,
|
| 14 |
+
"learning_rate": 3.846153846153847e-06,
|
| 15 |
+
"loss": 3.0068,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09523809523809523,
|
| 20 |
+
"grad_norm": 3.171875,
|
| 21 |
+
"learning_rate": 1.923076923076923e-05,
|
| 22 |
+
"loss": 3.0102,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.19047619047619047,
|
| 27 |
+
"grad_norm": 2.375,
|
| 28 |
+
"learning_rate": 3.846153846153846e-05,
|
| 29 |
+
"loss": 2.9279,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2857142857142857,
|
| 34 |
+
"grad_norm": 1.90625,
|
| 35 |
+
"learning_rate": 5.769230769230769e-05,
|
| 36 |
+
"loss": 2.6349,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.38095238095238093,
|
| 41 |
+
"grad_norm": 1.703125,
|
| 42 |
+
"learning_rate": 7.692307692307693e-05,
|
| 43 |
+
"loss": 2.3632,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.47619047619047616,
|
| 48 |
+
"grad_norm": 1.625,
|
| 49 |
+
"learning_rate": 9.615384615384617e-05,
|
| 50 |
+
"loss": 2.1672,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.5714285714285714,
|
| 55 |
+
"grad_norm": 1.171875,
|
| 56 |
+
"learning_rate": 0.00011538461538461538,
|
| 57 |
+
"loss": 1.9589,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.6666666666666666,
|
| 62 |
+
"grad_norm": 0.69921875,
|
| 63 |
+
"learning_rate": 0.00013461538461538464,
|
| 64 |
+
"loss": 1.7472,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.7619047619047619,
|
| 69 |
+
"grad_norm": 0.84765625,
|
| 70 |
+
"learning_rate": 0.00015384615384615385,
|
| 71 |
+
"loss": 1.5556,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.8571428571428571,
|
| 76 |
+
"grad_norm": 0.35546875,
|
| 77 |
+
"learning_rate": 0.0001730769230769231,
|
| 78 |
+
"loss": 1.4254,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.9523809523809523,
|
| 83 |
+
"grad_norm": 1.3203125,
|
| 84 |
+
"learning_rate": 0.00019230769230769233,
|
| 85 |
+
"loss": 1.3387,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.9904761904761905,
|
| 90 |
+
"eval_loss": 2.571077346801758,
|
| 91 |
+
"eval_runtime": 0.5007,
|
| 92 |
+
"eval_samples_per_second": 19.97,
|
| 93 |
+
"eval_steps_per_second": 1.997,
|
| 94 |
"step": 52
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 1.0476190476190477,
|
| 98 |
+
"grad_norm": 1.546875,
|
| 99 |
+
"learning_rate": 0.00019997972289848503,
|
| 100 |
+
"loss": 1.279,
|
| 101 |
"step": 55
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 1.1428571428571428,
|
| 105 |
+
"grad_norm": 0.466796875,
|
| 106 |
+
"learning_rate": 0.00019985583705641418,
|
| 107 |
+
"loss": 1.2299,
|
| 108 |
"step": 60
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 1.2380952380952381,
|
| 112 |
+
"grad_norm": 0.1943359375,
|
| 113 |
+
"learning_rate": 0.00019961946980917456,
|
| 114 |
+
"loss": 1.1885,
|
| 115 |
"step": 65
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 1.3333333333333333,
|
| 119 |
+
"grad_norm": 0.404296875,
|
| 120 |
+
"learning_rate": 0.0001992708874098054,
|
| 121 |
+
"loss": 1.1735,
|
| 122 |
"step": 70
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 1.4285714285714286,
|
| 126 |
+
"grad_norm": 0.462890625,
|
| 127 |
+
"learning_rate": 0.0001988104825147528,
|
| 128 |
+
"loss": 1.1575,
|
| 129 |
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 1.5238095238095237,
|
| 133 |
+
"grad_norm": 0.443359375,
|
| 134 |
+
"learning_rate": 0.00019823877374156647,
|
| 135 |
+
"loss": 1.1404,
|
| 136 |
"step": 80
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 1.619047619047619,
|
| 140 |
+
"grad_norm": 0.306640625,
|
| 141 |
+
"learning_rate": 0.00019755640508470942,
|
| 142 |
+
"loss": 1.1303,
|
| 143 |
"step": 85
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 1.7142857142857144,
|
| 147 |
+
"grad_norm": 0.21875,
|
| 148 |
+
"learning_rate": 0.00019676414519013781,
|
| 149 |
+
"loss": 1.1185,
|
| 150 |
"step": 90
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 1.8095238095238095,
|
| 154 |
+
"grad_norm": 0.22265625,
|
| 155 |
+
"learning_rate": 0.00019586288648946947,
|
| 156 |
+
"loss": 1.1006,
|
| 157 |
"step": 95
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 1.9047619047619047,
|
| 161 |
+
"grad_norm": 0.30859375,
|
| 162 |
+
"learning_rate": 0.00019485364419471454,
|
| 163 |
+
"loss": 1.1027,
|
| 164 |
"step": 100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 2.0,
|
| 168 |
+
"grad_norm": 0.2236328125,
|
| 169 |
+
"learning_rate": 0.00019373755515470254,
|
| 170 |
+
"loss": 1.0869,
|
| 171 |
"step": 105
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 2.0,
|
| 175 |
+
"eval_loss": 2.4883005619049072,
|
| 176 |
+
"eval_runtime": 0.4918,
|
| 177 |
+
"eval_samples_per_second": 20.333,
|
| 178 |
+
"eval_steps_per_second": 2.033,
|
| 179 |
"step": 105
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 2.0952380952380953,
|
| 183 |
+
"grad_norm": 0.271484375,
|
| 184 |
+
"learning_rate": 0.00019251587657449236,
|
| 185 |
+
"loss": 1.0706,
|
| 186 |
"step": 110
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 2.1904761904761907,
|
| 190 |
+
"grad_norm": 0.373046875,
|
| 191 |
+
"learning_rate": 0.00019118998459920902,
|
| 192 |
+
"loss": 1.0609,
|
| 193 |
"step": 115
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 2.2857142857142856,
|
| 197 |
+
"grad_norm": 0.68359375,
|
| 198 |
+
"learning_rate": 0.0001897613727639014,
|
| 199 |
+
"loss": 1.057,
|
| 200 |
"step": 120
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 2.380952380952381,
|
| 204 |
+
"grad_norm": 0.50390625,
|
| 205 |
+
"learning_rate": 0.0001882316503111678,
|
| 206 |
+
"loss": 1.0538,
|
| 207 |
"step": 125
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 2.4761904761904763,
|
| 211 |
+
"grad_norm": 0.3046875,
|
| 212 |
+
"learning_rate": 0.00018660254037844388,
|
| 213 |
+
"loss": 1.0531,
|
| 214 |
"step": 130
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 2.571428571428571,
|
| 218 |
+
"grad_norm": 0.5703125,
|
| 219 |
+
"learning_rate": 0.00018487587805699526,
|
| 220 |
+
"loss": 1.0403,
|
| 221 |
"step": 135
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 2.6666666666666665,
|
| 225 |
+
"grad_norm": 0.203125,
|
| 226 |
+
"learning_rate": 0.00018305360832480117,
|
| 227 |
+
"loss": 1.0392,
|
| 228 |
"step": 140
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 2.761904761904762,
|
| 232 |
+
"grad_norm": 0.3359375,
|
| 233 |
+
"learning_rate": 0.00018113778385565733,
|
| 234 |
+
"loss": 1.044,
|
| 235 |
"step": 145
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 2.857142857142857,
|
| 239 |
+
"grad_norm": 0.25390625,
|
| 240 |
+
"learning_rate": 0.0001791305627069662,
|
| 241 |
+
"loss": 1.0329,
|
| 242 |
"step": 150
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 2.9523809523809526,
|
| 246 |
+
"grad_norm": 0.279296875,
|
| 247 |
+
"learning_rate": 0.00017703420588881946,
|
| 248 |
+
"loss": 1.0363,
|
| 249 |
"step": 155
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 2.9904761904761905,
|
| 253 |
+
"eval_loss": 2.4850006103515625,
|
| 254 |
+
"eval_runtime": 0.5452,
|
| 255 |
+
"eval_samples_per_second": 18.343,
|
| 256 |
+
"eval_steps_per_second": 1.834,
|
| 257 |
"step": 157
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 3.0476190476190474,
|
| 261 |
+
"grad_norm": 0.244140625,
|
| 262 |
+
"learning_rate": 0.00017485107481711012,
|
| 263 |
+
"loss": 1.024,
|
| 264 |
"step": 160
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 3.142857142857143,
|
| 268 |
+
"grad_norm": 0.37109375,
|
| 269 |
+
"learning_rate": 0.00017258362865354426,
|
| 270 |
+
"loss": 1.0143,
|
| 271 |
"step": 165
|
| 272 |
},
|
| 273 |
{
|
| 274 |
"epoch": 3.238095238095238,
|
| 275 |
+
"grad_norm": 0.34375,
|
| 276 |
+
"learning_rate": 0.00017023442153554777,
|
| 277 |
+
"loss": 1.0159,
|
| 278 |
"step": 170
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 3.3333333333333335,
|
| 282 |
+
"grad_norm": 0.259765625,
|
| 283 |
+
"learning_rate": 0.0001678060996991891,
|
| 284 |
+
"loss": 1.01,
|
| 285 |
"step": 175
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 3.4285714285714284,
|
| 289 |
+
"grad_norm": 0.314453125,
|
| 290 |
+
"learning_rate": 0.0001653013984983585,
|
| 291 |
+
"loss": 1.0035,
|
| 292 |
"step": 180
|
| 293 |
},
|
| 294 |
{
|
| 295 |
"epoch": 3.5238095238095237,
|
| 296 |
+
"grad_norm": 1.265625,
|
| 297 |
+
"learning_rate": 0.00016272313932356162,
|
| 298 |
+
"loss": 1.0042,
|
| 299 |
"step": 185
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 3.619047619047619,
|
| 303 |
+
"grad_norm": 0.81640625,
|
| 304 |
+
"learning_rate": 0.0001600742264237979,
|
| 305 |
+
"loss": 1.0087,
|
| 306 |
"step": 190
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 3.7142857142857144,
|
| 310 |
+
"grad_norm": 0.384765625,
|
| 311 |
+
"learning_rate": 0.0001573576436351046,
|
| 312 |
+
"loss": 1.0057,
|
| 313 |
"step": 195
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 3.8095238095238093,
|
| 317 |
+
"grad_norm": 0.283203125,
|
| 318 |
+
"learning_rate": 0.00015457645101945046,
|
| 319 |
+
"loss": 0.9958,
|
| 320 |
"step": 200
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 3.9047619047619047,
|
| 324 |
+
"grad_norm": 0.2890625,
|
| 325 |
+
"learning_rate": 0.00015173378141776568,
|
| 326 |
+
"loss": 1.0002,
|
| 327 |
"step": 205
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 4.0,
|
| 331 |
+
"grad_norm": 0.298828125,
|
| 332 |
+
"learning_rate": 0.00014883283692099112,
|
| 333 |
+
"loss": 0.9967,
|
| 334 |
"step": 210
|
| 335 |
},
|
| 336 |
{
|
| 337 |
"epoch": 4.0,
|
| 338 |
+
"eval_loss": 2.480916738510132,
|
| 339 |
+
"eval_runtime": 0.4922,
|
| 340 |
+
"eval_samples_per_second": 20.317,
|
| 341 |
+
"eval_steps_per_second": 2.032,
|
| 342 |
"step": 210
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 4.095238095238095,
|
| 346 |
+
"grad_norm": 0.244140625,
|
| 347 |
+
"learning_rate": 0.00014587688526312143,
|
| 348 |
+
"loss": 0.9743,
|
| 349 |
"step": 215
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 4.190476190476191,
|
| 353 |
+
"grad_norm": 0.3359375,
|
| 354 |
+
"learning_rate": 0.00014286925614030542,
|
| 355 |
+
"loss": 0.9753,
|
| 356 |
"step": 220
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 4.285714285714286,
|
| 360 |
+
"grad_norm": 0.25,
|
| 361 |
+
"learning_rate": 0.0001398133374601501,
|
| 362 |
+
"loss": 0.9836,
|
| 363 |
"step": 225
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 4.380952380952381,
|
| 367 |
+
"grad_norm": 0.267578125,
|
| 368 |
+
"learning_rate": 0.00013671257152545277,
|
| 369 |
+
"loss": 0.9794,
|
| 370 |
"step": 230
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 4.476190476190476,
|
| 374 |
+
"grad_norm": 0.25,
|
| 375 |
+
"learning_rate": 0.0001335704511566605,
|
| 376 |
+
"loss": 0.9753,
|
| 377 |
"step": 235
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 4.571428571428571,
|
| 381 |
+
"grad_norm": 0.26953125,
|
| 382 |
+
"learning_rate": 0.0001303905157574247,
|
| 383 |
+
"loss": 0.9812,
|
| 384 |
"step": 240
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 4.666666666666667,
|
| 388 |
+
"grad_norm": 0.25390625,
|
| 389 |
+
"learning_rate": 0.00012717634732768243,
|
| 390 |
+
"loss": 0.9818,
|
| 391 |
"step": 245
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 4.761904761904762,
|
| 395 |
+
"grad_norm": 0.302734375,
|
| 396 |
+
"learning_rate": 0.0001239315664287558,
|
| 397 |
+
"loss": 0.9834,
|
| 398 |
"step": 250
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 4.857142857142857,
|
| 402 |
+
"grad_norm": 0.302734375,
|
| 403 |
+
"learning_rate": 0.00012065982810501404,
|
| 404 |
+
"loss": 0.9737,
|
| 405 |
"step": 255
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 4.9523809523809526,
|
| 409 |
+
"grad_norm": 0.267578125,
|
| 410 |
+
"learning_rate": 0.00011736481776669306,
|
| 411 |
+
"loss": 0.98,
|
| 412 |
"step": 260
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 4.9904761904761905,
|
| 416 |
+
"eval_loss": 2.490432024002075,
|
| 417 |
+
"eval_runtime": 0.7012,
|
| 418 |
+
"eval_samples_per_second": 14.262,
|
| 419 |
+
"eval_steps_per_second": 1.426,
|
| 420 |
"step": 262
|
| 421 |
},
|
| 422 |
{
|
| 423 |
"epoch": 5.0476190476190474,
|
| 424 |
+
"grad_norm": 0.546875,
|
| 425 |
+
"learning_rate": 0.00011405024703850929,
|
| 426 |
+
"loss": 0.9722,
|
| 427 |
"step": 265
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 5.142857142857143,
|
| 431 |
+
"grad_norm": 0.314453125,
|
| 432 |
+
"learning_rate": 0.00011071984957874479,
|
| 433 |
+
"loss": 0.9661,
|
| 434 |
"step": 270
|
| 435 |
},
|
| 436 |
{
|
| 437 |
"epoch": 5.238095238095238,
|
| 438 |
+
"grad_norm": 0.296875,
|
| 439 |
+
"learning_rate": 0.00010737737687351284,
|
| 440 |
+
"loss": 0.9631,
|
| 441 |
"step": 275
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 5.333333333333333,
|
| 445 |
+
"grad_norm": 0.2890625,
|
| 446 |
+
"learning_rate": 0.00010402659401094152,
|
| 447 |
+
"loss": 0.9568,
|
| 448 |
"step": 280
|
| 449 |
},
|
| 450 |
{
|
| 451 |
"epoch": 5.428571428571429,
|
| 452 |
+
"grad_norm": 0.3359375,
|
| 453 |
+
"learning_rate": 0.00010067127544003563,
|
| 454 |
+
"loss": 0.9619,
|
| 455 |
"step": 285
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 5.523809523809524,
|
| 459 |
+
"grad_norm": 0.255859375,
|
| 460 |
+
"learning_rate": 9.73152007189939e-05,
|
| 461 |
+
"loss": 0.9597,
|
| 462 |
"step": 290
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 5.619047619047619,
|
| 466 |
+
"grad_norm": 0.3984375,
|
| 467 |
+
"learning_rate": 9.396215025777139e-05,
|
| 468 |
+
"loss": 0.9619,
|
| 469 |
"step": 295
|
| 470 |
},
|
| 471 |
{
|
| 472 |
"epoch": 5.714285714285714,
|
| 473 |
+
"grad_norm": 0.3515625,
|
| 474 |
+
"learning_rate": 9.061590105968208e-05,
|
| 475 |
+
"loss": 0.9614,
|
| 476 |
"step": 300
|
| 477 |
},
|
| 478 |
{
|
| 479 |
"epoch": 5.809523809523809,
|
| 480 |
+
"grad_norm": 0.435546875,
|
| 481 |
+
"learning_rate": 8.728022246683894e-05,
|
| 482 |
+
"loss": 0.9536,
|
| 483 |
"step": 305
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 5.904761904761905,
|
| 487 |
+
"grad_norm": 0.3359375,
|
| 488 |
+
"learning_rate": 8.395887191422397e-05,
|
| 489 |
+
"loss": 0.9584,
|
| 490 |
"step": 310
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 6.0,
|
| 494 |
+
"grad_norm": 0.2265625,
|
| 495 |
+
"learning_rate": 8.065559069717088e-05,
|
| 496 |
+
"loss": 0.9574,
|
| 497 |
"step": 315
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 6.0,
|
| 501 |
+
"eval_loss": 2.499537706375122,
|
| 502 |
+
"eval_runtime": 0.4857,
|
| 503 |
+
"eval_samples_per_second": 20.591,
|
| 504 |
+
"eval_steps_per_second": 2.059,
|
| 505 |
"step": 315
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 6.095238095238095,
|
| 509 |
+
"grad_norm": 0.24609375,
|
| 510 |
+
"learning_rate": 7.73740997570278e-05,
|
| 511 |
+
"loss": 0.9494,
|
| 512 |
"step": 320
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 6.190476190476191,
|
| 516 |
+
"grad_norm": 0.2490234375,
|
| 517 |
+
"learning_rate": 7.411809548974792e-05,
|
| 518 |
+
"loss": 0.9472,
|
| 519 |
"step": 325
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 6.285714285714286,
|
| 523 |
+
"grad_norm": 0.263671875,
|
| 524 |
+
"learning_rate": 7.089124558212871e-05,
|
| 525 |
+
"loss": 0.9521,
|
| 526 |
"step": 330
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 6.380952380952381,
|
| 530 |
+
"grad_norm": 0.275390625,
|
| 531 |
+
"learning_rate": 6.769718488039023e-05,
|
| 532 |
+
"loss": 0.9466,
|
| 533 |
"step": 335
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 6.476190476190476,
|
| 537 |
+
"grad_norm": 0.240234375,
|
| 538 |
+
"learning_rate": 6.453951129574644e-05,
|
| 539 |
+
"loss": 0.9436,
|
| 540 |
"step": 340
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 6.571428571428571,
|
| 544 |
+
"grad_norm": 0.240234375,
|
| 545 |
+
"learning_rate": 6.142178175158149e-05,
|
| 546 |
+
"loss": 0.9501,
|
| 547 |
"step": 345
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 6.666666666666667,
|
| 551 |
+
"grad_norm": 0.259765625,
|
| 552 |
+
"learning_rate": 5.834750817679606e-05,
|
| 553 |
+
"loss": 0.9484,
|
| 554 |
"step": 350
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 6.761904761904762,
|
| 558 |
+
"grad_norm": 0.234375,
|
| 559 |
+
"learning_rate": 5.5320153549837415e-05,
|
| 560 |
+
"loss": 0.9489,
|
| 561 |
"step": 355
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 6.857142857142857,
|
| 565 |
+
"grad_norm": 0.2236328125,
|
| 566 |
+
"learning_rate": 5.234312799786921e-05,
|
| 567 |
+
"loss": 0.9486,
|
| 568 |
"step": 360
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 6.9523809523809526,
|
| 572 |
+
"grad_norm": 0.2333984375,
|
| 573 |
+
"learning_rate": 4.9419784955474524e-05,
|
| 574 |
+
"loss": 0.9392,
|
| 575 |
"step": 365
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 6.9904761904761905,
|
| 579 |
+
"eval_loss": 2.5062003135681152,
|
| 580 |
+
"eval_runtime": 0.5559,
|
| 581 |
+
"eval_samples_per_second": 17.989,
|
| 582 |
+
"eval_steps_per_second": 1.799,
|
| 583 |
"step": 367
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 7.0476190476190474,
|
| 587 |
+
"grad_norm": 0.2158203125,
|
| 588 |
+
"learning_rate": 4.6553417387219886e-05,
|
| 589 |
+
"loss": 0.9358,
|
| 590 |
"step": 370
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 7.142857142857143,
|
| 594 |
+
"grad_norm": 0.24609375,
|
| 595 |
+
"learning_rate": 4.374725407833532e-05,
|
| 596 |
+
"loss": 0.9313,
|
| 597 |
"step": 375
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 7.238095238095238,
|
| 601 |
+
"grad_norm": 0.2578125,
|
| 602 |
+
"learning_rate": 4.100445599768774e-05,
|
| 603 |
+
"loss": 0.9343,
|
| 604 |
"step": 380
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 7.333333333333333,
|
| 608 |
+
"grad_norm": 0.2236328125,
|
| 609 |
+
"learning_rate": 3.832811273714569e-05,
|
| 610 |
+
"loss": 0.9384,
|
| 611 |
"step": 385
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 7.428571428571429,
|
| 615 |
+
"grad_norm": 0.2490234375,
|
| 616 |
+
"learning_rate": 3.5721239031346066e-05,
|
| 617 |
+
"loss": 0.9387,
|
| 618 |
"step": 390
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 7.523809523809524,
|
| 622 |
+
"grad_norm": 0.2265625,
|
| 623 |
+
"learning_rate": 3.318677136178228e-05,
|
| 624 |
+
"loss": 0.9356,
|
| 625 |
"step": 395
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 7.619047619047619,
|
| 629 |
+
"grad_norm": 0.232421875,
|
| 630 |
+
"learning_rate": 3.072756464904006e-05,
|
| 631 |
+
"loss": 0.9428,
|
| 632 |
"step": 400
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 7.714285714285714,
|
| 636 |
+
"grad_norm": 0.2099609375,
|
| 637 |
+
"learning_rate": 2.8346389036906828e-05,
|
| 638 |
+
"loss": 0.9405,
|
| 639 |
"step": 405
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 7.809523809523809,
|
| 643 |
+
"grad_norm": 0.216796875,
|
| 644 |
+
"learning_rate": 2.6045926771976303e-05,
|
| 645 |
+
"loss": 0.9401,
|
| 646 |
"step": 410
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 7.904761904761905,
|
| 650 |
+
"grad_norm": 0.22265625,
|
| 651 |
+
"learning_rate": 2.382876918226409e-05,
|
| 652 |
+
"loss": 0.9459,
|
| 653 |
"step": 415
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 8.0,
|
| 657 |
+
"grad_norm": 0.2109375,
|
| 658 |
+
"learning_rate": 2.1697413758237784e-05,
|
| 659 |
+
"loss": 0.9415,
|
| 660 |
"step": 420
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 8.0,
|
| 664 |
+
"eval_loss": 2.514326333999634,
|
| 665 |
+
"eval_runtime": 0.4829,
|
| 666 |
+
"eval_samples_per_second": 20.709,
|
| 667 |
+
"eval_steps_per_second": 2.071,
|
| 668 |
"step": 420
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 8.095238095238095,
|
| 672 |
+
"grad_norm": 0.255859375,
|
| 673 |
+
"learning_rate": 1.965426133954854e-05,
|
| 674 |
+
"loss": 0.9302,
|
| 675 |
"step": 425
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 8.19047619047619,
|
| 679 |
+
"grad_norm": 0.2138671875,
|
| 680 |
+
"learning_rate": 1.7701613410634365e-05,
|
| 681 |
+
"loss": 0.941,
|
| 682 |
"step": 430
|
| 683 |
},
|
| 684 |
{
|
| 685 |
"epoch": 8.285714285714286,
|
| 686 |
+
"grad_norm": 0.21484375,
|
| 687 |
+
"learning_rate": 1.584166950824061e-05,
|
| 688 |
+
"loss": 0.939,
|
| 689 |
"step": 435
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 8.380952380952381,
|
| 693 |
+
"grad_norm": 0.2021484375,
|
| 694 |
+
"learning_rate": 1.4076524743778319e-05,
|
| 695 |
+
"loss": 0.9319,
|
| 696 |
"step": 440
|
| 697 |
},
|
| 698 |
{
|
| 699 |
"epoch": 8.476190476190476,
|
| 700 |
+
"grad_norm": 0.212890625,
|
| 701 |
+
"learning_rate": 1.2408167443311214e-05,
|
| 702 |
+
"loss": 0.9314,
|
| 703 |
"step": 445
|
| 704 |
},
|
| 705 |
{
|
| 706 |
"epoch": 8.571428571428571,
|
| 707 |
+
"grad_norm": 0.208984375,
|
| 708 |
+
"learning_rate": 1.083847690782972e-05,
|
| 709 |
+
"loss": 0.9331,
|
| 710 |
"step": 450
|
| 711 |
},
|
| 712 |
{
|
| 713 |
"epoch": 8.666666666666666,
|
| 714 |
+
"grad_norm": 0.2080078125,
|
| 715 |
+
"learning_rate": 9.369221296335006e-06,
|
| 716 |
+
"loss": 0.9392,
|
| 717 |
"step": 455
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 8.761904761904763,
|
| 721 |
+
"grad_norm": 0.2138671875,
|
| 722 |
+
"learning_rate": 8.002055634117578e-06,
|
| 723 |
+
"loss": 0.9305,
|
| 724 |
"step": 460
|
| 725 |
},
|
| 726 |
{
|
| 727 |
"epoch": 8.857142857142858,
|
| 728 |
+
"grad_norm": 0.2060546875,
|
| 729 |
+
"learning_rate": 6.738519948473976e-06,
|
| 730 |
+
"loss": 0.9343,
|
| 731 |
"step": 465
|
| 732 |
},
|
| 733 |
{
|
| 734 |
"epoch": 8.952380952380953,
|
| 735 |
+
"grad_norm": 0.2099609375,
|
| 736 |
+
"learning_rate": 5.580037533961546e-06,
|
| 737 |
+
"loss": 0.9367,
|
| 738 |
"step": 470
|
| 739 |
},
|
| 740 |
{
|
| 741 |
"epoch": 8.99047619047619,
|
| 742 |
+
"eval_loss": 2.515460252761841,
|
| 743 |
+
"eval_runtime": 0.6608,
|
| 744 |
+
"eval_samples_per_second": 15.132,
|
| 745 |
+
"eval_steps_per_second": 1.513,
|
| 746 |
"step": 472
|
| 747 |
},
|
| 748 |
{
|
| 749 |
"epoch": 9.047619047619047,
|
| 750 |
+
"grad_norm": 0.2041015625,
|
| 751 |
+
"learning_rate": 4.527913349145441e-06,
|
| 752 |
+
"loss": 0.9329,
|
| 753 |
"step": 475
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 9.142857142857142,
|
| 757 |
+
"grad_norm": 0.2060546875,
|
| 758 |
+
"learning_rate": 3.5833325466437694e-06,
|
| 759 |
+
"loss": 0.9294,
|
| 760 |
"step": 480
|
| 761 |
},
|
| 762 |
{
|
| 763 |
"epoch": 9.238095238095237,
|
| 764 |
+
"grad_norm": 0.2080078125,
|
| 765 |
+
"learning_rate": 2.7473591381266708e-06,
|
| 766 |
+
"loss": 0.9319,
|
| 767 |
"step": 485
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 9.333333333333334,
|
| 771 |
+
"grad_norm": 0.212890625,
|
| 772 |
+
"learning_rate": 2.0209347957732328e-06,
|
| 773 |
+
"loss": 0.9388,
|
| 774 |
"step": 490
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 9.428571428571429,
|
| 778 |
+
"grad_norm": 0.205078125,
|
| 779 |
+
"learning_rate": 1.404877791536141e-06,
|
| 780 |
+
"loss": 0.9344,
|
| 781 |
"step": 495
|
| 782 |
},
|
| 783 |
{
|
| 784 |
"epoch": 9.523809523809524,
|
| 785 |
+
"grad_norm": 0.20703125,
|
| 786 |
+
"learning_rate": 8.998820754091531e-07,
|
| 787 |
+
"loss": 0.9345,
|
| 788 |
"step": 500
|
| 789 |
},
|
| 790 |
{
|
| 791 |
"epoch": 9.619047619047619,
|
| 792 |
+
"grad_norm": 0.203125,
|
| 793 |
+
"learning_rate": 5.065164937354428e-07,
|
| 794 |
+
"loss": 0.9339,
|
| 795 |
"step": 505
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 9.714285714285714,
|
| 799 |
+
"grad_norm": 0.2021484375,
|
| 800 |
+
"learning_rate": 2.2522414843748618e-07,
|
| 801 |
+
"loss": 0.9356,
|
| 802 |
"step": 510
|
| 803 |
},
|
| 804 |
{
|
| 805 |
"epoch": 9.80952380952381,
|
| 806 |
+
"grad_norm": 0.2021484375,
|
| 807 |
+
"learning_rate": 5.632189789027687e-08,
|
| 808 |
+
"loss": 0.9265,
|
| 809 |
"step": 515
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 9.904761904761905,
|
| 813 |
+
"grad_norm": 0.201171875,
|
| 814 |
+
"learning_rate": 0.0,
|
| 815 |
+
"loss": 0.9348,
|
| 816 |
"step": 520
|
| 817 |
},
|
| 818 |
{
|
| 819 |
+
"epoch": 9.904761904761905,
|
| 820 |
+
"eval_loss": 2.5155773162841797,
|
| 821 |
+
"eval_runtime": 0.4929,
|
| 822 |
+
"eval_samples_per_second": 20.288,
|
| 823 |
+
"eval_steps_per_second": 2.029,
|
| 824 |
+
"step": 520
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 825 |
},
|
| 826 |
{
|
| 827 |
+
"epoch": 9.904761904761905,
|
| 828 |
+
"step": 520,
|
| 829 |
+
"total_flos": 8.18501087353897e+17,
|
| 830 |
+
"train_loss": 1.098491345002101,
|
| 831 |
+
"train_runtime": 2780.3493,
|
| 832 |
+
"train_samples_per_second": 23.972,
|
| 833 |
+
"train_steps_per_second": 0.187
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 834 |
}
|
| 835 |
],
|
| 836 |
"logging_steps": 5,
|
|
|
|
| 850 |
"attributes": {}
|
| 851 |
}
|
| 852 |
},
|
| 853 |
+
"total_flos": 8.18501087353897e+17,
|
| 854 |
+
"train_batch_size": 8,
|
| 855 |
"trial_name": null,
|
| 856 |
"trial_params": null
|
| 857 |
}
|