Model save
Browse files- README.md +17 -13
- adapter_model.safetensors +1 -1
- all_results.json +6 -11
- runs/Apr21_23-51-41_060407ee4835/events.out.tfevents.1713758215.060407ee4835.3304.0 +2 -2
- train_results.json +6 -6
- trainer_state.json +569 -529
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-7b
|
| 10 |
datasets:
|
| 11 |
-
-
|
| 12 |
model-index:
|
| 13 |
- name: coding_llamaduo_result3
|
| 14 |
results: []
|
|
@@ -19,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
|
| 20 |
# coding_llamaduo_result3
|
| 21 |
|
| 22 |
-
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
-
- Loss: 1.
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -45,24 +44,29 @@ The following hyperparameters were used during training:
|
|
| 45 |
- eval_batch_size: 2
|
| 46 |
- seed: 42
|
| 47 |
- distributed_type: multi-GPU
|
| 48 |
-
- num_devices:
|
| 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
|
| 55 |
-
- num_epochs:
|
| 56 |
|
| 57 |
### Training results
|
| 58 |
|
| 59 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 60 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
|
| 67 |
|
| 68 |
### Framework versions
|
|
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: google/gemma-7b
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: coding_llamaduo_result3
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# coding_llamaduo_result3
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.7502
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 44 |
- eval_batch_size: 2
|
| 45 |
- seed: 42
|
| 46 |
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 4
|
| 48 |
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 16
|
| 50 |
+
- total_eval_batch_size: 8
|
| 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
|
| 54 |
+
- num_epochs: 10
|
| 55 |
|
| 56 |
### Training results
|
| 57 |
|
| 58 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 0.987 | 1.0 | 82 | 1.2808 |
|
| 61 |
+
| 0.6859 | 2.0 | 164 | 1.1719 |
|
| 62 |
+
| 0.5836 | 3.0 | 246 | 1.1480 |
|
| 63 |
+
| 0.5178 | 4.0 | 328 | 1.1717 |
|
| 64 |
+
| 0.4668 | 5.0 | 410 | 1.2044 |
|
| 65 |
+
| 0.3955 | 6.0 | 492 | 1.3252 |
|
| 66 |
+
| 0.3233 | 7.0 | 574 | 1.4225 |
|
| 67 |
+
| 0.2669 | 8.0 | 656 | 1.6119 |
|
| 68 |
+
| 0.2591 | 9.0 | 738 | 1.7353 |
|
| 69 |
+
| 0.2367 | 10.0 | 820 | 1.7502 |
|
| 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 100060536
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:34aea7a58143aa2cce083b76d541e9e03828d7a2593e5a15e290bc6ec5443756
|
| 3 |
size 100060536
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 1.843,
|
| 7 |
-
"eval_steps_per_second": 0.922,
|
| 8 |
-
"total_flos": 6.28853392705323e+17,
|
| 9 |
-
"train_loss": 1.6822980254161648,
|
| 10 |
-
"train_runtime": 5646.8844,
|
| 11 |
"train_samples": 19039,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.257706785410646e+18,
|
| 4 |
+
"train_loss": 1.5742560074096772,
|
| 5 |
+
"train_runtime": 5911.531,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 19039,
|
| 7 |
+
"train_samples_per_second": 2.214,
|
| 8 |
+
"train_steps_per_second": 0.139
|
| 9 |
}
|
runs/Apr21_23-51-41_060407ee4835/events.out.tfevents.1713758215.060407ee4835.3304.0
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:f2fcf23e3c754eb90edaa64da5c0d8ad4995495d6c98c2136469e9476515a067
|
| 3 |
+
size 43185
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 19039,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 1.257706785410646e+18,
|
| 4 |
+
"train_loss": 1.5742560074096772,
|
| 5 |
+
"train_runtime": 5911.531,
|
| 6 |
"train_samples": 19039,
|
| 7 |
+
"train_samples_per_second": 2.214,
|
| 8 |
+
"train_steps_per_second": 0.139
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 820,
|
| 7 |
"is_hyper_param_search": false,
|
|
@@ -9,1216 +9,1256 @@
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 2.4390243902439027e-06,
|
| 15 |
-
"loss":
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 1.2195121951219513e-05,
|
| 22 |
-
"loss":
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch": 0.
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 2.4390243902439026e-05,
|
| 29 |
-
"loss": 35.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch": 0.
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 3.6585365853658535e-05,
|
| 36 |
-
"loss":
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch": 0.
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 4.878048780487805e-05,
|
| 43 |
-
"loss":
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"grad_norm":
|
| 49 |
"learning_rate": 6.097560975609756e-05,
|
| 50 |
-
"loss": 16.
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch": 0.
|
| 55 |
-
"grad_norm": 4.
|
| 56 |
"learning_rate": 7.317073170731707e-05,
|
| 57 |
-
"loss": 15.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch": 0.
|
| 62 |
-
"grad_norm":
|
| 63 |
"learning_rate": 8.53658536585366e-05,
|
| 64 |
-
"loss":
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 0.
|
| 69 |
-
"grad_norm":
|
| 70 |
"learning_rate": 9.75609756097561e-05,
|
| 71 |
-
"loss": 13.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
-
"epoch": 0.
|
| 76 |
-
"grad_norm":
|
| 77 |
"learning_rate": 0.00010975609756097563,
|
| 78 |
-
"loss":
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
-
"epoch": 0.
|
| 83 |
-
"grad_norm":
|
| 84 |
"learning_rate": 0.00012195121951219512,
|
| 85 |
-
"loss": 2.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch": 0.
|
| 90 |
-
"grad_norm":
|
| 91 |
"learning_rate": 0.00013414634146341464,
|
| 92 |
-
"loss": 1.
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"epoch": 0.
|
| 97 |
-
"grad_norm":
|
| 98 |
"learning_rate": 0.00014634146341463414,
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"epoch": 0.
|
| 104 |
-
"grad_norm":
|
| 105 |
"learning_rate": 0.00015853658536585366,
|
| 106 |
-
"loss": 1.
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
-
"epoch": 0.
|
| 111 |
-
"grad_norm":
|
| 112 |
"learning_rate": 0.0001707317073170732,
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
-
"epoch": 0.
|
| 118 |
-
"grad_norm":
|
| 119 |
"learning_rate": 0.0001829268292682927,
|
| 120 |
-
"loss":
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
-
"epoch": 0.
|
| 125 |
-
"grad_norm":
|
| 126 |
"learning_rate": 0.0001951219512195122,
|
| 127 |
-
"loss": 0.
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
-
"epoch": 0
|
| 132 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 133 |
"learning_rate": 0.00019999184556954776,
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch":
|
| 139 |
-
"grad_norm": 1.
|
| 140 |
"learning_rate": 0.0001999420177550043,
|
| 141 |
-
"loss": 0.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
-
"epoch":
|
| 146 |
-
"grad_norm":
|
| 147 |
"learning_rate": 0.00019984691491033906,
|
| 148 |
-
"loss": 0.
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch":
|
| 153 |
-
"grad_norm":
|
| 154 |
"learning_rate": 0.00019970658011837404,
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
-
"epoch":
|
| 160 |
-
"grad_norm":
|
| 161 |
"learning_rate": 0.00019952107695258992,
|
| 162 |
-
"loss": 0.
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"epoch":
|
| 167 |
-
"grad_norm":
|
| 168 |
"learning_rate": 0.00019929048944832638,
|
| 169 |
-
"loss": 0.
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
-
"epoch":
|
| 174 |
-
"grad_norm": 2.
|
| 175 |
"learning_rate": 0.00019901492206471325,
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
-
"epoch":
|
| 181 |
-
"grad_norm":
|
| 182 |
"learning_rate": 0.00019869449963734893,
|
| 183 |
-
"loss": 0.
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
-
"epoch":
|
| 188 |
-
"grad_norm": 1.
|
| 189 |
"learning_rate": 0.00019832936732174834,
|
| 190 |
-
"loss": 0.
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
-
"epoch":
|
| 195 |
-
"grad_norm":
|
| 196 |
"learning_rate": 0.00019791969052758562,
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
-
"epoch":
|
| 202 |
-
"grad_norm":
|
| 203 |
"learning_rate": 0.00019746565484376132,
|
| 204 |
-
"loss": 0.
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
-
"epoch":
|
| 209 |
-
"grad_norm":
|
| 210 |
"learning_rate": 0.00019696746595432828,
|
| 211 |
-
"loss": 0.
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
-
"epoch":
|
| 216 |
-
"grad_norm":
|
| 217 |
"learning_rate": 0.0001964253495453141,
|
| 218 |
-
"loss": 0.
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
-
"epoch":
|
| 223 |
-
"grad_norm": 1.
|
| 224 |
"learning_rate": 0.00019583955120248237,
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
-
"epoch":
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
"learning_rate": 0.00019521033630007928,
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"grad_norm":
|
| 238 |
"learning_rate": 0.00019453798988061535,
|
| 239 |
-
"loss": 0.
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
-
"epoch":
|
| 244 |
-
"eval_loss": 1.
|
| 245 |
-
"eval_runtime": 1.
|
| 246 |
-
"eval_samples_per_second": 1.
|
| 247 |
-
"eval_steps_per_second": 0.
|
| 248 |
"step": 164
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch":
|
| 252 |
-
"grad_norm": 0.
|
| 253 |
"learning_rate": 0.00019382281652573785,
|
| 254 |
-
"loss": 0.
|
| 255 |
"step": 165
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
| 259 |
-
"grad_norm":
|
| 260 |
"learning_rate": 0.00019306514021825118,
|
| 261 |
-
"loss": 0.
|
| 262 |
"step": 170
|
| 263 |
},
|
| 264 |
{
|
| 265 |
-
"epoch":
|
| 266 |
-
"grad_norm":
|
| 267 |
"learning_rate": 0.00019226530419534833,
|
| 268 |
-
"loss": 0.
|
| 269 |
"step": 175
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"grad_norm":
|
| 274 |
"learning_rate": 0.00019142367079312021,
|
| 275 |
-
"loss": 0.
|
| 276 |
"step": 180
|
| 277 |
},
|
| 278 |
{
|
| 279 |
-
"epoch":
|
| 280 |
-
"grad_norm": 0.
|
| 281 |
"learning_rate": 0.00019054062128241264,
|
| 282 |
-
"loss": 0.
|
| 283 |
"step": 185
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch":
|
| 287 |
-
"grad_norm": 1.
|
| 288 |
"learning_rate": 0.00018961655569610557,
|
| 289 |
-
"loss": 0.
|
| 290 |
"step": 190
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch":
|
| 294 |
-
"grad_norm": 0.
|
| 295 |
"learning_rate": 0.0001886518926478932,
|
| 296 |
-
"loss": 0.
|
| 297 |
"step": 195
|
| 298 |
},
|
| 299 |
{
|
| 300 |
-
"epoch":
|
| 301 |
-
"grad_norm": 0.
|
| 302 |
"learning_rate": 0.00018764706914264635,
|
| 303 |
-
"loss": 0.
|
| 304 |
"step": 200
|
| 305 |
},
|
| 306 |
{
|
| 307 |
-
"epoch":
|
| 308 |
-
"grad_norm":
|
| 309 |
"learning_rate": 0.00018660254037844388,
|
| 310 |
-
"loss": 0.
|
| 311 |
"step": 205
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch":
|
| 315 |
-
"grad_norm":
|
| 316 |
"learning_rate": 0.00018551877954036162,
|
| 317 |
-
"loss": 0.
|
| 318 |
"step": 210
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"epoch":
|
| 322 |
-
"grad_norm": 0.
|
| 323 |
"learning_rate": 0.00018439627758611385,
|
| 324 |
-
"loss": 0.
|
| 325 |
"step": 215
|
| 326 |
},
|
| 327 |
{
|
| 328 |
-
"epoch":
|
| 329 |
-
"grad_norm": 0
|
| 330 |
"learning_rate": 0.00018323554302364272,
|
| 331 |
-
"loss": 0.
|
| 332 |
"step": 220
|
| 333 |
},
|
| 334 |
{
|
| 335 |
-
"epoch":
|
| 336 |
-
"grad_norm":
|
| 337 |
"learning_rate": 0.00018203710168075788,
|
| 338 |
-
"loss": 0.
|
| 339 |
"step": 225
|
| 340 |
},
|
| 341 |
{
|
| 342 |
-
"epoch":
|
| 343 |
-
"grad_norm": 1.
|
| 344 |
"learning_rate": 0.0001808014964669293,
|
| 345 |
-
"loss": 0.
|
| 346 |
"step": 230
|
| 347 |
},
|
| 348 |
{
|
| 349 |
-
"epoch":
|
| 350 |
-
"grad_norm": 0.
|
| 351 |
"learning_rate": 0.00017952928712734268,
|
| 352 |
-
"loss": 0.
|
| 353 |
"step": 235
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch":
|
| 357 |
-
"grad_norm":
|
| 358 |
"learning_rate": 0.00017822104998932713,
|
| 359 |
-
"loss": 0.
|
| 360 |
"step": 240
|
| 361 |
},
|
| 362 |
{
|
| 363 |
-
"epoch":
|
| 364 |
-
"grad_norm": 0.
|
| 365 |
"learning_rate": 0.00017687737770127185,
|
| 366 |
-
"loss": 0.
|
| 367 |
"step": 245
|
| 368 |
},
|
| 369 |
{
|
| 370 |
-
"epoch":
|
| 371 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 372 |
"learning_rate": 0.00017549887896414851,
|
| 373 |
-
"loss": 0.
|
| 374 |
"step": 250
|
| 375 |
},
|
| 376 |
{
|
| 377 |
-
"epoch":
|
| 378 |
-
"grad_norm":
|
| 379 |
"learning_rate": 0.0001740861782557618,
|
| 380 |
-
"loss": 0.
|
| 381 |
"step": 255
|
| 382 |
},
|
| 383 |
{
|
| 384 |
-
"epoch":
|
| 385 |
-
"grad_norm": 0.
|
| 386 |
"learning_rate": 0.0001726399155478529,
|
| 387 |
-
"loss": 0.
|
| 388 |
"step": 260
|
| 389 |
},
|
| 390 |
{
|
| 391 |
-
"epoch":
|
| 392 |
-
"grad_norm": 0.
|
| 393 |
"learning_rate": 0.00017116074601618417,
|
| 394 |
-
"loss": 0.
|
| 395 |
"step": 265
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch":
|
| 399 |
-
"grad_norm": 0.
|
| 400 |
"learning_rate": 0.0001696493397437357,
|
| 401 |
-
"loss": 0.
|
| 402 |
"step": 270
|
| 403 |
},
|
| 404 |
{
|
| 405 |
-
"epoch":
|
| 406 |
-
"grad_norm":
|
| 407 |
"learning_rate": 0.00016810638141714934,
|
| 408 |
-
"loss": 0.
|
| 409 |
"step": 275
|
| 410 |
},
|
| 411 |
{
|
| 412 |
-
"epoch":
|
| 413 |
-
"grad_norm":
|
| 414 |
"learning_rate": 0.00016653257001655652,
|
| 415 |
-
"loss": 0.
|
| 416 |
"step": 280
|
| 417 |
},
|
| 418 |
{
|
| 419 |
-
"epoch":
|
| 420 |
-
"grad_norm": 0.
|
| 421 |
"learning_rate": 0.0001649286184989315,
|
| 422 |
-
"loss": 0.
|
| 423 |
"step": 285
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"epoch":
|
| 427 |
-
"grad_norm":
|
| 428 |
"learning_rate": 0.0001632952534751122,
|
| 429 |
-
"loss": 0.
|
| 430 |
"step": 290
|
| 431 |
},
|
| 432 |
{
|
| 433 |
-
"epoch":
|
| 434 |
-
"grad_norm": 0.
|
| 435 |
"learning_rate": 0.00016163321488063637,
|
| 436 |
-
"loss": 0.
|
| 437 |
"step": 295
|
| 438 |
},
|
| 439 |
{
|
| 440 |
-
"epoch":
|
| 441 |
-
"grad_norm": 0.
|
| 442 |
"learning_rate": 0.00015994325564054122,
|
| 443 |
-
"loss": 0.
|
| 444 |
"step": 300
|
| 445 |
},
|
| 446 |
{
|
| 447 |
-
"epoch":
|
| 448 |
-
"grad_norm": 0.
|
| 449 |
"learning_rate": 0.00015822614132827837,
|
| 450 |
-
"loss": 0.
|
| 451 |
"step": 305
|
| 452 |
},
|
| 453 |
{
|
| 454 |
-
"epoch":
|
| 455 |
-
"grad_norm": 0.
|
| 456 |
"learning_rate": 0.00015648264981889934,
|
| 457 |
-
"loss": 0.
|
| 458 |
"step": 310
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch":
|
| 462 |
-
"grad_norm": 0.
|
| 463 |
"learning_rate": 0.00015471357093666804,
|
| 464 |
-
"loss": 0.
|
| 465 |
"step": 315
|
| 466 |
},
|
| 467 |
{
|
| 468 |
-
"epoch":
|
| 469 |
-
"grad_norm": 0.
|
| 470 |
"learning_rate": 0.00015291970609726007,
|
| 471 |
-
"loss": 0.
|
| 472 |
"step": 320
|
| 473 |
},
|
| 474 |
{
|
| 475 |
-
"epoch":
|
| 476 |
-
"grad_norm": 0.
|
| 477 |
"learning_rate": 0.00015110186794471103,
|
| 478 |
-
"loss": 0.
|
| 479 |
"step": 325
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch":
|
| 483 |
-
"eval_loss": 1.
|
| 484 |
-
"eval_runtime": 1.
|
| 485 |
-
"eval_samples_per_second": 1.
|
| 486 |
-
"eval_steps_per_second": 0.
|
| 487 |
"step": 328
|
| 488 |
},
|
| 489 |
{
|
| 490 |
-
"epoch":
|
| 491 |
-
"grad_norm": 0.
|
| 492 |
"learning_rate": 0.00014926087998327837,
|
| 493 |
-
"loss": 0.
|
| 494 |
"step": 330
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"grad_norm": 0.
|
| 499 |
"learning_rate": 0.00014739757620438307,
|
| 500 |
-
"loss": 0.
|
| 501 |
"step": 335
|
| 502 |
},
|
| 503 |
{
|
| 504 |
-
"epoch":
|
| 505 |
-
"grad_norm":
|
| 506 |
"learning_rate": 0.0001455128007088009,
|
| 507 |
-
"loss": 0.
|
| 508 |
"step": 340
|
| 509 |
},
|
| 510 |
{
|
| 511 |
-
"epoch":
|
| 512 |
-
"grad_norm": 0.
|
| 513 |
"learning_rate": 0.00014360740732427367,
|
| 514 |
-
"loss": 0.
|
| 515 |
"step": 345
|
| 516 |
},
|
| 517 |
{
|
| 518 |
-
"epoch":
|
| 519 |
-
"grad_norm": 0.
|
| 520 |
"learning_rate": 0.00014168225921871433,
|
| 521 |
-
"loss": 0.
|
| 522 |
"step": 350
|
| 523 |
},
|
| 524 |
{
|
| 525 |
-
"epoch":
|
| 526 |
-
"grad_norm": 0.
|
| 527 |
"learning_rate": 0.00013973822850918055,
|
| 528 |
-
"loss": 0.
|
| 529 |
"step": 355
|
| 530 |
},
|
| 531 |
{
|
| 532 |
-
"epoch":
|
| 533 |
-
"grad_norm": 0.
|
| 534 |
"learning_rate": 0.0001377761958667946,
|
| 535 |
-
"loss": 0.
|
| 536 |
"step": 360
|
| 537 |
},
|
| 538 |
{
|
| 539 |
-
"epoch":
|
| 540 |
-
"grad_norm": 0.
|
| 541 |
"learning_rate": 0.00013579705011778766,
|
| 542 |
-
"loss": 0.
|
| 543 |
"step": 365
|
| 544 |
},
|
| 545 |
{
|
| 546 |
-
"epoch":
|
| 547 |
-
"grad_norm": 0.
|
| 548 |
"learning_rate": 0.00013380168784085027,
|
| 549 |
-
"loss": 0.
|
| 550 |
"step": 370
|
| 551 |
},
|
| 552 |
{
|
| 553 |
-
"epoch":
|
| 554 |
-
"grad_norm": 0.
|
| 555 |
"learning_rate": 0.00013179101296097035,
|
| 556 |
-
"loss": 0.
|
| 557 |
"step": 375
|
| 558 |
},
|
| 559 |
{
|
| 560 |
-
"epoch":
|
| 561 |
-
"grad_norm": 0.
|
| 562 |
"learning_rate": 0.00012976593633994346,
|
| 563 |
-
"loss": 0.
|
| 564 |
"step": 380
|
| 565 |
},
|
| 566 |
{
|
| 567 |
-
"epoch":
|
| 568 |
-
"grad_norm": 0.
|
| 569 |
"learning_rate": 0.0001277273753637408,
|
| 570 |
-
"loss": 0.
|
| 571 |
"step": 385
|
| 572 |
},
|
| 573 |
{
|
| 574 |
-
"epoch":
|
| 575 |
-
"grad_norm": 0.
|
| 576 |
"learning_rate": 0.00012567625352692127,
|
| 577 |
-
"loss": 0.
|
| 578 |
"step": 390
|
| 579 |
},
|
| 580 |
{
|
| 581 |
-
"epoch":
|
| 582 |
-
"grad_norm":
|
| 583 |
"learning_rate": 0.0001236135000142765,
|
| 584 |
-
"loss": 0.
|
| 585 |
"step": 395
|
| 586 |
},
|
| 587 |
{
|
| 588 |
-
"epoch":
|
| 589 |
-
"grad_norm": 0.
|
| 590 |
"learning_rate": 0.00012154004927989815,
|
| 591 |
-
"loss": 0.
|
| 592 |
"step": 400
|
| 593 |
},
|
| 594 |
{
|
| 595 |
-
"epoch":
|
| 596 |
-
"grad_norm": 0.
|
| 597 |
"learning_rate": 0.00011945684062385803,
|
| 598 |
-
"loss": 0.
|
| 599 |
"step": 405
|
| 600 |
},
|
| 601 |
{
|
| 602 |
-
"epoch":
|
| 603 |
-
"grad_norm": 0.
|
| 604 |
"learning_rate": 0.00011736481776669306,
|
| 605 |
-
"loss": 0.
|
| 606 |
"step": 410
|
| 607 |
},
|
| 608 |
{
|
| 609 |
-
"epoch":
|
| 610 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 611 |
"learning_rate": 0.00011526492842188745,
|
| 612 |
-
"loss": 0.
|
| 613 |
"step": 415
|
| 614 |
},
|
| 615 |
{
|
| 616 |
-
"epoch":
|
| 617 |
-
"grad_norm": 0.
|
| 618 |
"learning_rate": 0.0001131581238665465,
|
| 619 |
-
"loss": 0.
|
| 620 |
"step": 420
|
| 621 |
},
|
| 622 |
{
|
| 623 |
-
"epoch":
|
| 624 |
-
"grad_norm": 0.
|
| 625 |
"learning_rate": 0.00011104535851045539,
|
| 626 |
-
"loss": 0.
|
| 627 |
"step": 425
|
| 628 |
},
|
| 629 |
{
|
| 630 |
-
"epoch":
|
| 631 |
-
"grad_norm": 0.
|
| 632 |
"learning_rate": 0.00010892758946371944,
|
| 633 |
-
"loss": 0.
|
| 634 |
"step": 430
|
| 635 |
},
|
| 636 |
{
|
| 637 |
-
"epoch":
|
| 638 |
-
"grad_norm": 0.
|
| 639 |
"learning_rate": 0.00010680577610318072,
|
| 640 |
-
"loss": 0.
|
| 641 |
"step": 435
|
| 642 |
},
|
| 643 |
{
|
| 644 |
-
"epoch":
|
| 645 |
-
"grad_norm": 0.
|
| 646 |
"learning_rate": 0.00010468087963780789,
|
| 647 |
-
"loss": 0.
|
| 648 |
"step": 440
|
| 649 |
},
|
| 650 |
{
|
| 651 |
-
"epoch":
|
| 652 |
-
"grad_norm": 0.
|
| 653 |
"learning_rate": 0.00010255386267325602,
|
| 654 |
-
"loss": 0.
|
| 655 |
"step": 445
|
| 656 |
},
|
| 657 |
{
|
| 658 |
-
"epoch":
|
| 659 |
-
"grad_norm": 0.
|
| 660 |
"learning_rate": 0.00010042568877579388,
|
| 661 |
-
"loss": 0.
|
| 662 |
"step": 450
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch":
|
| 666 |
-
"grad_norm": 0.
|
| 667 |
"learning_rate": 9.829732203579584e-05,
|
| 668 |
-
"loss": 0.
|
| 669 |
"step": 455
|
| 670 |
},
|
| 671 |
{
|
| 672 |
-
"epoch":
|
| 673 |
-
"grad_norm": 0.
|
| 674 |
"learning_rate": 9.616972663099647e-05,
|
| 675 |
-
"loss": 0.
|
| 676 |
"step": 460
|
| 677 |
},
|
| 678 |
{
|
| 679 |
-
"epoch":
|
| 680 |
-
"grad_norm": 0.
|
| 681 |
"learning_rate": 9.404386638970542e-05,
|
| 682 |
-
"loss": 0.
|
| 683 |
"step": 465
|
| 684 |
},
|
| 685 |
{
|
| 686 |
-
"epoch":
|
| 687 |
-
"grad_norm": 0.
|
| 688 |
"learning_rate": 9.192070435418079e-05,
|
| 689 |
-
"loss": 0.
|
| 690 |
"step": 470
|
| 691 |
},
|
| 692 |
{
|
| 693 |
-
"epoch":
|
| 694 |
-
"grad_norm": 0.
|
| 695 |
"learning_rate": 8.980120234435849e-05,
|
| 696 |
-
"loss": 0.
|
| 697 |
"step": 475
|
| 698 |
},
|
| 699 |
{
|
| 700 |
-
"epoch":
|
| 701 |
-
"grad_norm": 0.
|
| 702 |
"learning_rate": 8.768632052213531e-05,
|
| 703 |
-
"loss": 0.
|
| 704 |
"step": 480
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch":
|
| 708 |
-
"grad_norm": 0.
|
| 709 |
"learning_rate": 8.557701695640321e-05,
|
| 710 |
-
"loss": 0.
|
| 711 |
"step": 485
|
| 712 |
},
|
| 713 |
{
|
| 714 |
-
"epoch":
|
| 715 |
-
"grad_norm": 0.
|
| 716 |
"learning_rate": 8.347424718903151e-05,
|
| 717 |
-
"loss": 0.
|
| 718 |
"step": 490
|
| 719 |
},
|
| 720 |
{
|
| 721 |
-
"epoch":
|
| 722 |
-
"eval_loss": 1.
|
| 723 |
-
"eval_runtime": 1.
|
| 724 |
-
"eval_samples_per_second": 1.
|
| 725 |
-
"eval_steps_per_second": 0.
|
| 726 |
"step": 492
|
| 727 |
},
|
| 728 |
{
|
| 729 |
-
"epoch":
|
| 730 |
-
"grad_norm": 0.
|
| 731 |
"learning_rate": 8.13789638019942e-05,
|
| 732 |
-
"loss": 0.
|
| 733 |
"step": 495
|
| 734 |
},
|
| 735 |
{
|
| 736 |
-
"epoch":
|
| 737 |
-
"grad_norm": 0.
|
| 738 |
"learning_rate": 7.929211598583794e-05,
|
| 739 |
-
"loss": 0.
|
| 740 |
"step": 500
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch":
|
| 744 |
-
"grad_norm": 0.
|
| 745 |
"learning_rate": 7.721464910968627e-05,
|
| 746 |
-
"loss": 0.
|
| 747 |
"step": 505
|
| 748 |
},
|
| 749 |
{
|
| 750 |
-
"epoch":
|
| 751 |
-
"grad_norm": 0.
|
| 752 |
"learning_rate": 7.514750429297528e-05,
|
| 753 |
-
"loss": 0.
|
| 754 |
"step": 510
|
| 755 |
},
|
| 756 |
{
|
| 757 |
-
"epoch":
|
| 758 |
-
"grad_norm": 0.
|
| 759 |
"learning_rate": 7.309161797911441e-05,
|
| 760 |
-
"loss": 0.
|
| 761 |
"step": 515
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch":
|
| 765 |
-
"grad_norm": 0.
|
| 766 |
"learning_rate": 7.104792151126515e-05,
|
| 767 |
-
"loss": 0.
|
| 768 |
"step": 520
|
| 769 |
},
|
| 770 |
{
|
| 771 |
-
"epoch":
|
| 772 |
-
"grad_norm": 0.
|
| 773 |
"learning_rate": 6.901734071043071e-05,
|
| 774 |
-
"loss": 0.
|
| 775 |
"step": 525
|
| 776 |
},
|
| 777 |
{
|
| 778 |
-
"epoch":
|
| 779 |
-
"grad_norm": 0.
|
| 780 |
"learning_rate": 6.700079545604708e-05,
|
| 781 |
-
"loss": 0.
|
| 782 |
"step": 530
|
| 783 |
},
|
| 784 |
{
|
| 785 |
-
"epoch":
|
| 786 |
-
"grad_norm": 0.
|
| 787 |
"learning_rate": 6.499919926926566e-05,
|
| 788 |
-
"loss": 0.
|
| 789 |
"step": 535
|
| 790 |
},
|
| 791 |
{
|
| 792 |
-
"epoch":
|
| 793 |
-
"grad_norm": 0.
|
| 794 |
"learning_rate": 6.301345889911637e-05,
|
| 795 |
-
"loss": 0.
|
| 796 |
"step": 540
|
| 797 |
},
|
| 798 |
{
|
| 799 |
-
"epoch":
|
| 800 |
-
"grad_norm": 0.
|
| 801 |
"learning_rate": 6.104447391173858e-05,
|
| 802 |
-
"loss": 0.
|
| 803 |
"step": 545
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"epoch":
|
| 807 |
-
"grad_norm": 0.
|
| 808 |
"learning_rate": 5.909313628286601e-05,
|
| 809 |
-
"loss": 0.
|
| 810 |
"step": 550
|
| 811 |
},
|
| 812 |
{
|
| 813 |
-
"epoch":
|
| 814 |
-
"grad_norm": 0.
|
| 815 |
"learning_rate": 5.716032999375006e-05,
|
| 816 |
-
"loss": 0.
|
| 817 |
"step": 555
|
| 818 |
},
|
| 819 |
{
|
| 820 |
-
"epoch":
|
| 821 |
-
"grad_norm": 0.
|
| 822 |
"learning_rate": 5.524693063070492e-05,
|
| 823 |
-
"loss": 0.
|
| 824 |
"step": 560
|
| 825 |
},
|
| 826 |
{
|
| 827 |
-
"epoch":
|
| 828 |
-
"grad_norm": 0.
|
| 829 |
"learning_rate": 5.335380498845559e-05,
|
| 830 |
-
"loss": 0.
|
| 831 |
"step": 565
|
| 832 |
},
|
| 833 |
{
|
| 834 |
-
"epoch":
|
| 835 |
-
"grad_norm": 0.
|
| 836 |
"learning_rate": 5.148181067746862e-05,
|
| 837 |
-
"loss": 0.
|
| 838 |
"step": 570
|
| 839 |
},
|
| 840 |
{
|
| 841 |
-
"epoch":
|
| 842 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 843 |
"learning_rate": 4.963179573544357e-05,
|
| 844 |
-
"loss": 0.
|
| 845 |
"step": 575
|
| 846 |
},
|
| 847 |
{
|
| 848 |
-
"epoch":
|
| 849 |
-
"grad_norm": 0.
|
| 850 |
"learning_rate": 4.7804598243140666e-05,
|
| 851 |
-
"loss": 0.
|
| 852 |
"step": 580
|
| 853 |
},
|
| 854 |
{
|
| 855 |
-
"epoch":
|
| 856 |
-
"grad_norm": 0.
|
| 857 |
"learning_rate": 4.60010459447196e-05,
|
| 858 |
-
"loss": 0.
|
| 859 |
"step": 585
|
| 860 |
},
|
| 861 |
{
|
| 862 |
-
"epoch":
|
| 863 |
-
"grad_norm": 0.
|
| 864 |
"learning_rate": 4.422195587276058e-05,
|
| 865 |
-
"loss": 0.
|
| 866 |
"step": 590
|
| 867 |
},
|
| 868 |
{
|
| 869 |
-
"epoch":
|
| 870 |
-
"grad_norm": 0.
|
| 871 |
"learning_rate": 4.2468133978137945e-05,
|
| 872 |
-
"loss": 0.
|
| 873 |
"step": 595
|
| 874 |
},
|
| 875 |
{
|
| 876 |
-
"epoch":
|
| 877 |
-
"grad_norm": 0.
|
| 878 |
"learning_rate": 4.0740374764914136e-05,
|
| 879 |
-
"loss": 0.
|
| 880 |
"step": 600
|
| 881 |
},
|
| 882 |
{
|
| 883 |
-
"epoch":
|
| 884 |
-
"grad_norm": 0.
|
| 885 |
"learning_rate": 3.903946093041877e-05,
|
| 886 |
-
"loss": 0.
|
| 887 |
"step": 605
|
| 888 |
},
|
| 889 |
{
|
| 890 |
-
"epoch":
|
| 891 |
-
"grad_norm": 0.
|
| 892 |
"learning_rate": 3.736616301067694e-05,
|
| 893 |
-
"loss": 0.
|
| 894 |
"step": 610
|
| 895 |
},
|
| 896 |
{
|
| 897 |
-
"epoch":
|
| 898 |
-
"grad_norm": 0.
|
| 899 |
"learning_rate": 3.5721239031346066e-05,
|
| 900 |
-
"loss": 0.
|
| 901 |
"step": 615
|
| 902 |
},
|
| 903 |
{
|
| 904 |
-
"epoch":
|
| 905 |
-
"grad_norm": 0.
|
| 906 |
"learning_rate": 3.410543416432069e-05,
|
| 907 |
-
"loss": 0.
|
| 908 |
"step": 620
|
| 909 |
},
|
| 910 |
{
|
| 911 |
-
"epoch":
|
| 912 |
-
"grad_norm": 0.
|
| 913 |
"learning_rate": 3.2519480390159806e-05,
|
| 914 |
-
"loss": 0.
|
| 915 |
"step": 625
|
| 916 |
},
|
| 917 |
{
|
| 918 |
-
"epoch":
|
| 919 |
-
"grad_norm": 0.
|
| 920 |
"learning_rate": 3.096409616649023e-05,
|
| 921 |
-
"loss": 0.
|
| 922 |
"step": 630
|
| 923 |
},
|
| 924 |
{
|
| 925 |
-
"epoch":
|
| 926 |
-
"grad_norm": 0.
|
| 927 |
"learning_rate": 2.9439986102536043e-05,
|
| 928 |
-
"loss": 0.
|
| 929 |
"step": 635
|
| 930 |
},
|
| 931 |
{
|
| 932 |
-
"epoch":
|
| 933 |
-
"grad_norm": 0.
|
| 934 |
"learning_rate": 2.794784063992131e-05,
|
| 935 |
-
"loss": 0.
|
| 936 |
"step": 640
|
| 937 |
},
|
| 938 |
{
|
| 939 |
-
"epoch":
|
| 940 |
-
"grad_norm": 0.
|
| 941 |
"learning_rate": 2.6488335739891178e-05,
|
| 942 |
-
"loss": 0.
|
| 943 |
"step": 645
|
| 944 |
},
|
| 945 |
{
|
| 946 |
-
"epoch":
|
| 947 |
-
"grad_norm": 0.
|
| 948 |
"learning_rate": 2.50621325770927e-05,
|
| 949 |
-
"loss": 0.
|
| 950 |
"step": 650
|
| 951 |
},
|
| 952 |
{
|
| 953 |
-
"epoch":
|
| 954 |
-
"grad_norm": 0.
|
| 955 |
"learning_rate": 2.366987724005404e-05,
|
| 956 |
-
"loss": 0.
|
| 957 |
"step": 655
|
| 958 |
},
|
| 959 |
{
|
| 960 |
-
"epoch":
|
| 961 |
-
"eval_loss": 1.
|
| 962 |
-
"eval_runtime": 1.
|
| 963 |
-
"eval_samples_per_second": 1.
|
| 964 |
-
"eval_steps_per_second": 0.
|
| 965 |
"step": 656
|
| 966 |
},
|
| 967 |
{
|
| 968 |
-
"epoch":
|
| 969 |
-
"grad_norm": 0.
|
| 970 |
"learning_rate": 2.2312200438498043e-05,
|
| 971 |
-
"loss": 0.
|
| 972 |
"step": 660
|
| 973 |
},
|
| 974 |
{
|
| 975 |
-
"epoch":
|
| 976 |
-
"grad_norm": 0.
|
| 977 |
"learning_rate": 2.0989717217622652e-05,
|
| 978 |
-
"loss": 0.
|
| 979 |
"step": 665
|
| 980 |
},
|
| 981 |
{
|
| 982 |
-
"epoch":
|
| 983 |
-
"grad_norm": 0.
|
| 984 |
"learning_rate": 1.9703026679477256e-05,
|
| 985 |
-
"loss": 0.
|
| 986 |
"step": 670
|
| 987 |
},
|
| 988 |
{
|
| 989 |
-
"epoch":
|
| 990 |
-
"grad_norm": 0.
|
| 991 |
"learning_rate": 1.8452711711561842e-05,
|
| 992 |
-
"loss": 0.
|
| 993 |
"step": 675
|
| 994 |
},
|
| 995 |
{
|
| 996 |
-
"epoch":
|
| 997 |
-
"grad_norm": 0.
|
| 998 |
"learning_rate": 1.7239338722771327e-05,
|
| 999 |
-
"loss": 0.
|
| 1000 |
"step": 680
|
| 1001 |
},
|
| 1002 |
{
|
| 1003 |
-
"epoch":
|
| 1004 |
-
"grad_norm": 0.
|
| 1005 |
"learning_rate": 1.6063457386805004e-05,
|
| 1006 |
-
"loss": 0.
|
| 1007 |
"step": 685
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
-
"epoch":
|
| 1011 |
-
"grad_norm": 0.
|
| 1012 |
"learning_rate": 1.4925600393157324e-05,
|
| 1013 |
-
"loss": 0.
|
| 1014 |
"step": 690
|
| 1015 |
},
|
| 1016 |
{
|
| 1017 |
-
"epoch":
|
| 1018 |
-
"grad_norm": 0.
|
| 1019 |
"learning_rate": 1.3826283205802427e-05,
|
| 1020 |
-
"loss": 0.
|
| 1021 |
"step": 695
|
| 1022 |
},
|
| 1023 |
{
|
| 1024 |
-
"epoch":
|
| 1025 |
-
"grad_norm": 0.
|
| 1026 |
"learning_rate": 1.2766003829682505e-05,
|
| 1027 |
-
"loss": 0.
|
| 1028 |
"step": 700
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
-
"epoch":
|
| 1032 |
-
"grad_norm": 0.
|
| 1033 |
"learning_rate": 1.1745242585104955e-05,
|
| 1034 |
-
"loss": 0.
|
| 1035 |
"step": 705
|
| 1036 |
},
|
| 1037 |
{
|
| 1038 |
-
"epoch":
|
| 1039 |
-
"grad_norm": 0.
|
| 1040 |
"learning_rate": 1.0764461890151112e-05,
|
| 1041 |
-
"loss": 0.
|
| 1042 |
"step": 710
|
| 1043 |
},
|
| 1044 |
{
|
| 1045 |
-
"epoch":
|
| 1046 |
-
"grad_norm": 0.
|
| 1047 |
"learning_rate": 9.824106051194859e-06,
|
| 1048 |
-
"loss": 0.
|
| 1049 |
"step": 715
|
| 1050 |
},
|
| 1051 |
{
|
| 1052 |
-
"epoch":
|
| 1053 |
-
"grad_norm": 0.
|
| 1054 |
"learning_rate": 8.924601061626048e-06,
|
| 1055 |
-
"loss": 0.
|
| 1056 |
"step": 720
|
| 1057 |
},
|
| 1058 |
{
|
| 1059 |
-
"epoch":
|
| 1060 |
-
"grad_norm": 0.
|
| 1061 |
"learning_rate": 8.066354408870048e-06,
|
| 1062 |
-
"loss": 0.
|
| 1063 |
"step": 725
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
-
"epoch":
|
| 1067 |
-
"grad_norm": 0.
|
| 1068 |
"learning_rate": 7.249754889790539e-06,
|
| 1069 |
-
"loss": 0.
|
| 1070 |
"step": 730
|
| 1071 |
},
|
| 1072 |
{
|
| 1073 |
-
"epoch":
|
| 1074 |
-
"grad_norm": 0.
|
| 1075 |
"learning_rate": 6.475172434559573e-06,
|
| 1076 |
-
"loss": 0.
|
| 1077 |
"step": 735
|
| 1078 |
},
|
| 1079 |
{
|
| 1080 |
-
"epoch":
|
| 1081 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1082 |
"learning_rate": 5.742957939074412e-06,
|
| 1083 |
-
"loss": 0.
|
| 1084 |
"step": 740
|
| 1085 |
},
|
| 1086 |
{
|
| 1087 |
-
"epoch":
|
| 1088 |
-
"grad_norm": 0.
|
| 1089 |
"learning_rate": 5.0534431059970685e-06,
|
| 1090 |
-
"loss": 0.
|
| 1091 |
"step": 745
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
-
"epoch":
|
| 1095 |
-
"grad_norm": 0.
|
| 1096 |
"learning_rate": 4.40694029448877e-06,
|
| 1097 |
-
"loss": 0.
|
| 1098 |
"step": 750
|
| 1099 |
},
|
| 1100 |
{
|
| 1101 |
-
"epoch":
|
| 1102 |
-
"grad_norm": 0.
|
| 1103 |
"learning_rate": 3.803742378707198e-06,
|
| 1104 |
-
"loss": 0.
|
| 1105 |
"step": 755
|
| 1106 |
},
|
| 1107 |
{
|
| 1108 |
-
"epoch":
|
| 1109 |
-
"grad_norm": 0.
|
| 1110 |
"learning_rate": 3.2441226151306404e-06,
|
| 1111 |
-
"loss": 0.
|
| 1112 |
"step": 760
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
-
"epoch":
|
| 1116 |
-
"grad_norm": 0.
|
| 1117 |
"learning_rate": 2.7283345187693264e-06,
|
| 1118 |
-
"loss": 0.
|
| 1119 |
"step": 765
|
| 1120 |
},
|
| 1121 |
{
|
| 1122 |
-
"epoch":
|
| 1123 |
-
"grad_norm": 0.
|
| 1124 |
"learning_rate": 2.256611748319792e-06,
|
| 1125 |
-
"loss": 0.
|
| 1126 |
"step": 770
|
| 1127 |
},
|
| 1128 |
{
|
| 1129 |
-
"epoch":
|
| 1130 |
-
"grad_norm": 0.
|
| 1131 |
"learning_rate": 1.8291680003145073e-06,
|
| 1132 |
-
"loss": 0.
|
| 1133 |
"step": 775
|
| 1134 |
},
|
| 1135 |
{
|
| 1136 |
-
"epoch":
|
| 1137 |
-
"grad_norm": 0.
|
| 1138 |
"learning_rate": 1.4461969123145457e-06,
|
| 1139 |
-
"loss": 0.
|
| 1140 |
"step": 780
|
| 1141 |
},
|
| 1142 |
{
|
| 1143 |
-
"epoch":
|
| 1144 |
-
"grad_norm": 0.
|
| 1145 |
"learning_rate": 1.107871975189234e-06,
|
| 1146 |
-
"loss": 0.
|
| 1147 |
"step": 785
|
| 1148 |
},
|
| 1149 |
{
|
| 1150 |
-
"epoch":
|
| 1151 |
-
"grad_norm": 0.
|
| 1152 |
"learning_rate": 8.143464545226298e-07,
|
| 1153 |
-
"loss": 0.
|
| 1154 |
"step": 790
|
| 1155 |
},
|
| 1156 |
{
|
| 1157 |
-
"epoch":
|
| 1158 |
-
"grad_norm": 0.
|
| 1159 |
"learning_rate": 5.657533211820942e-07,
|
| 1160 |
-
"loss": 0.
|
| 1161 |
"step": 795
|
| 1162 |
},
|
| 1163 |
{
|
| 1164 |
-
"epoch":
|
| 1165 |
-
"grad_norm": 0.
|
| 1166 |
"learning_rate": 3.622051910808666e-07,
|
| 1167 |
-
"loss": 0.
|
| 1168 |
"step": 800
|
| 1169 |
},
|
| 1170 |
{
|
| 1171 |
-
"epoch":
|
| 1172 |
-
"grad_norm": 0.
|
| 1173 |
"learning_rate": 2.037942741615617e-07,
|
| 1174 |
-
"loss": 0.
|
| 1175 |
"step": 805
|
| 1176 |
},
|
| 1177 |
{
|
| 1178 |
-
"epoch":
|
| 1179 |
-
"grad_norm": 0.
|
| 1180 |
"learning_rate": 9.059233262386225e-08,
|
| 1181 |
-
"loss": 0.
|
| 1182 |
"step": 810
|
| 1183 |
},
|
| 1184 |
{
|
| 1185 |
-
"epoch":
|
| 1186 |
-
"grad_norm": 0.
|
| 1187 |
"learning_rate": 2.2650648415334376e-08,
|
| 1188 |
-
"loss": 0.
|
| 1189 |
"step": 815
|
| 1190 |
},
|
| 1191 |
{
|
| 1192 |
-
"epoch":
|
| 1193 |
-
"grad_norm": 0.
|
| 1194 |
"learning_rate": 0.0,
|
| 1195 |
-
"loss": 0.
|
| 1196 |
"step": 820
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
-
"epoch":
|
| 1200 |
-
"eval_loss": 1.
|
| 1201 |
-
"eval_runtime": 1.
|
| 1202 |
-
"eval_samples_per_second": 1.
|
| 1203 |
-
"eval_steps_per_second": 0.
|
| 1204 |
"step": 820
|
| 1205 |
},
|
| 1206 |
{
|
| 1207 |
-
"epoch":
|
| 1208 |
"step": 820,
|
| 1209 |
-
"total_flos":
|
| 1210 |
-
"train_loss": 1.
|
| 1211 |
-
"train_runtime":
|
| 1212 |
-
"train_samples_per_second":
|
| 1213 |
-
"train_steps_per_second": 0.
|
| 1214 |
}
|
| 1215 |
],
|
| 1216 |
"logging_steps": 5,
|
| 1217 |
"max_steps": 820,
|
| 1218 |
"num_input_tokens_seen": 0,
|
| 1219 |
-
"num_train_epochs":
|
| 1220 |
"save_steps": 100,
|
| 1221 |
-
"total_flos":
|
| 1222 |
"train_batch_size": 2,
|
| 1223 |
"trial_name": null,
|
| 1224 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
"global_step": 820,
|
| 7 |
"is_hyper_param_search": false,
|
|
|
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.012195121951219513,
|
| 13 |
+
"grad_norm": 460.0,
|
| 14 |
"learning_rate": 2.4390243902439027e-06,
|
| 15 |
+
"loss": 48.1093,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.06097560975609756,
|
| 20 |
+
"grad_norm": 360.0,
|
| 21 |
"learning_rate": 1.2195121951219513e-05,
|
| 22 |
+
"loss": 42.5766,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.12195121951219512,
|
| 27 |
+
"grad_norm": 187.0,
|
| 28 |
"learning_rate": 2.4390243902439026e-05,
|
| 29 |
+
"loss": 35.1457,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.18292682926829268,
|
| 34 |
+
"grad_norm": 31.375,
|
| 35 |
"learning_rate": 3.6585365853658535e-05,
|
| 36 |
+
"loss": 21.8793,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.24390243902439024,
|
| 41 |
+
"grad_norm": 17.375,
|
| 42 |
"learning_rate": 4.878048780487805e-05,
|
| 43 |
+
"loss": 17.5136,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.3048780487804878,
|
| 48 |
+
"grad_norm": 7.59375,
|
| 49 |
"learning_rate": 6.097560975609756e-05,
|
| 50 |
+
"loss": 16.4183,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.36585365853658536,
|
| 55 |
+
"grad_norm": 4.09375,
|
| 56 |
"learning_rate": 7.317073170731707e-05,
|
| 57 |
+
"loss": 15.0038,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.4268292682926829,
|
| 62 |
+
"grad_norm": 8.25,
|
| 63 |
"learning_rate": 8.53658536585366e-05,
|
| 64 |
+
"loss": 14.2277,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.4878048780487805,
|
| 69 |
+
"grad_norm": 16.5,
|
| 70 |
"learning_rate": 9.75609756097561e-05,
|
| 71 |
+
"loss": 13.0434,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
+
"epoch": 0.5487804878048781,
|
| 76 |
+
"grad_norm": 35.25,
|
| 77 |
"learning_rate": 0.00010975609756097563,
|
| 78 |
+
"loss": 8.9174,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
+
"epoch": 0.6097560975609756,
|
| 83 |
+
"grad_norm": 11.125,
|
| 84 |
"learning_rate": 0.00012195121951219512,
|
| 85 |
+
"loss": 2.4207,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.6707317073170732,
|
| 90 |
+
"grad_norm": 1.5625,
|
| 91 |
"learning_rate": 0.00013414634146341464,
|
| 92 |
+
"loss": 1.5637,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"epoch": 0.7317073170731707,
|
| 97 |
+
"grad_norm": 1.9921875,
|
| 98 |
"learning_rate": 0.00014634146341463414,
|
| 99 |
+
"loss": 1.3611,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"epoch": 0.7926829268292683,
|
| 104 |
+
"grad_norm": 1.5,
|
| 105 |
"learning_rate": 0.00015853658536585366,
|
| 106 |
+
"loss": 1.1715,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 0.8536585365853658,
|
| 111 |
+
"grad_norm": 2.15625,
|
| 112 |
"learning_rate": 0.0001707317073170732,
|
| 113 |
+
"loss": 1.0679,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
+
"epoch": 0.9146341463414634,
|
| 118 |
+
"grad_norm": 1.15625,
|
| 119 |
"learning_rate": 0.0001829268292682927,
|
| 120 |
+
"loss": 0.9835,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
+
"epoch": 0.975609756097561,
|
| 125 |
+
"grad_norm": 2.078125,
|
| 126 |
"learning_rate": 0.0001951219512195122,
|
| 127 |
+
"loss": 0.987,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 1.0,
|
| 132 |
+
"eval_loss": 1.2807745933532715,
|
| 133 |
+
"eval_runtime": 1.1183,
|
| 134 |
+
"eval_samples_per_second": 1.788,
|
| 135 |
+
"eval_steps_per_second": 0.894,
|
| 136 |
+
"step": 82
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.0365853658536586,
|
| 140 |
+
"grad_norm": 1.421875,
|
| 141 |
"learning_rate": 0.00019999184556954776,
|
| 142 |
+
"loss": 0.887,
|
| 143 |
"step": 85
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"epoch": 1.0975609756097562,
|
| 147 |
+
"grad_norm": 1.6328125,
|
| 148 |
"learning_rate": 0.0001999420177550043,
|
| 149 |
+
"loss": 0.849,
|
| 150 |
"step": 90
|
| 151 |
},
|
| 152 |
{
|
| 153 |
+
"epoch": 1.1585365853658536,
|
| 154 |
+
"grad_norm": 4.1875,
|
| 155 |
"learning_rate": 0.00019984691491033906,
|
| 156 |
+
"loss": 0.8057,
|
| 157 |
"step": 95
|
| 158 |
},
|
| 159 |
{
|
| 160 |
+
"epoch": 1.2195121951219512,
|
| 161 |
+
"grad_norm": 3.34375,
|
| 162 |
"learning_rate": 0.00019970658011837404,
|
| 163 |
+
"loss": 0.8035,
|
| 164 |
"step": 100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
+
"epoch": 1.2804878048780488,
|
| 168 |
+
"grad_norm": 1.4375,
|
| 169 |
"learning_rate": 0.00019952107695258992,
|
| 170 |
+
"loss": 0.7717,
|
| 171 |
"step": 105
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 1.3414634146341464,
|
| 175 |
+
"grad_norm": 1.3046875,
|
| 176 |
"learning_rate": 0.00019929048944832638,
|
| 177 |
+
"loss": 0.7583,
|
| 178 |
"step": 110
|
| 179 |
},
|
| 180 |
{
|
| 181 |
+
"epoch": 1.4024390243902438,
|
| 182 |
+
"grad_norm": 2.265625,
|
| 183 |
"learning_rate": 0.00019901492206471325,
|
| 184 |
+
"loss": 0.7363,
|
| 185 |
"step": 115
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 1.4634146341463414,
|
| 189 |
+
"grad_norm": 0.5546875,
|
| 190 |
"learning_rate": 0.00019869449963734893,
|
| 191 |
+
"loss": 0.7397,
|
| 192 |
"step": 120
|
| 193 |
},
|
| 194 |
{
|
| 195 |
+
"epoch": 1.524390243902439,
|
| 196 |
+
"grad_norm": 1.5859375,
|
| 197 |
"learning_rate": 0.00019832936732174834,
|
| 198 |
+
"loss": 0.7268,
|
| 199 |
"step": 125
|
| 200 |
},
|
| 201 |
{
|
| 202 |
+
"epoch": 1.5853658536585367,
|
| 203 |
+
"grad_norm": 2.796875,
|
| 204 |
"learning_rate": 0.00019791969052758562,
|
| 205 |
+
"loss": 0.7103,
|
| 206 |
"step": 130
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 1.6463414634146343,
|
| 210 |
+
"grad_norm": 1.6640625,
|
| 211 |
"learning_rate": 0.00019746565484376132,
|
| 212 |
+
"loss": 0.7166,
|
| 213 |
"step": 135
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"epoch": 1.7073170731707317,
|
| 217 |
+
"grad_norm": 1.4296875,
|
| 218 |
"learning_rate": 0.00019696746595432828,
|
| 219 |
+
"loss": 0.7174,
|
| 220 |
"step": 140
|
| 221 |
},
|
| 222 |
{
|
| 223 |
+
"epoch": 1.7682926829268293,
|
| 224 |
+
"grad_norm": 2.296875,
|
| 225 |
"learning_rate": 0.0001964253495453141,
|
| 226 |
+
"loss": 0.6822,
|
| 227 |
"step": 145
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 1.8292682926829267,
|
| 231 |
+
"grad_norm": 1.8515625,
|
| 232 |
"learning_rate": 0.00019583955120248237,
|
| 233 |
+
"loss": 0.679,
|
| 234 |
"step": 150
|
| 235 |
},
|
| 236 |
{
|
| 237 |
+
"epoch": 1.8902439024390243,
|
| 238 |
+
"grad_norm": 0.42578125,
|
| 239 |
"learning_rate": 0.00019521033630007928,
|
| 240 |
+
"loss": 0.6708,
|
| 241 |
"step": 155
|
| 242 |
},
|
| 243 |
{
|
| 244 |
+
"epoch": 1.951219512195122,
|
| 245 |
+
"grad_norm": 0.490234375,
|
| 246 |
"learning_rate": 0.00019453798988061535,
|
| 247 |
+
"loss": 0.6859,
|
| 248 |
"step": 160
|
| 249 |
},
|
| 250 |
{
|
| 251 |
+
"epoch": 2.0,
|
| 252 |
+
"eval_loss": 1.1719168424606323,
|
| 253 |
+
"eval_runtime": 1.1192,
|
| 254 |
+
"eval_samples_per_second": 1.787,
|
| 255 |
+
"eval_steps_per_second": 0.893,
|
| 256 |
"step": 164
|
| 257 |
},
|
| 258 |
{
|
| 259 |
+
"epoch": 2.0121951219512195,
|
| 260 |
+
"grad_norm": 0.5703125,
|
| 261 |
"learning_rate": 0.00019382281652573785,
|
| 262 |
+
"loss": 0.6606,
|
| 263 |
"step": 165
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 2.073170731707317,
|
| 267 |
+
"grad_norm": 3.296875,
|
| 268 |
"learning_rate": 0.00019306514021825118,
|
| 269 |
+
"loss": 0.6528,
|
| 270 |
"step": 170
|
| 271 |
},
|
| 272 |
{
|
| 273 |
+
"epoch": 2.1341463414634148,
|
| 274 |
+
"grad_norm": 2.609375,
|
| 275 |
"learning_rate": 0.00019226530419534833,
|
| 276 |
+
"loss": 0.6424,
|
| 277 |
"step": 175
|
| 278 |
},
|
| 279 |
{
|
| 280 |
+
"epoch": 2.1951219512195124,
|
| 281 |
+
"grad_norm": 1.5703125,
|
| 282 |
"learning_rate": 0.00019142367079312021,
|
| 283 |
+
"loss": 0.6061,
|
| 284 |
"step": 180
|
| 285 |
},
|
| 286 |
{
|
| 287 |
+
"epoch": 2.2560975609756095,
|
| 288 |
+
"grad_norm": 0.5234375,
|
| 289 |
"learning_rate": 0.00019054062128241264,
|
| 290 |
+
"loss": 0.6186,
|
| 291 |
"step": 185
|
| 292 |
},
|
| 293 |
{
|
| 294 |
+
"epoch": 2.317073170731707,
|
| 295 |
+
"grad_norm": 1.2890625,
|
| 296 |
"learning_rate": 0.00018961655569610557,
|
| 297 |
+
"loss": 0.6099,
|
| 298 |
"step": 190
|
| 299 |
},
|
| 300 |
{
|
| 301 |
+
"epoch": 2.3780487804878048,
|
| 302 |
+
"grad_norm": 0.89453125,
|
| 303 |
"learning_rate": 0.0001886518926478932,
|
| 304 |
+
"loss": 0.6184,
|
| 305 |
"step": 195
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 2.4390243902439024,
|
| 309 |
+
"grad_norm": 0.6796875,
|
| 310 |
"learning_rate": 0.00018764706914264635,
|
| 311 |
+
"loss": 0.5949,
|
| 312 |
"step": 200
|
| 313 |
},
|
| 314 |
{
|
| 315 |
+
"epoch": 2.5,
|
| 316 |
+
"grad_norm": 0.8046875,
|
| 317 |
"learning_rate": 0.00018660254037844388,
|
| 318 |
+
"loss": 0.5993,
|
| 319 |
"step": 205
|
| 320 |
},
|
| 321 |
{
|
| 322 |
+
"epoch": 2.5609756097560976,
|
| 323 |
+
"grad_norm": 1.3125,
|
| 324 |
"learning_rate": 0.00018551877954036162,
|
| 325 |
+
"loss": 0.6041,
|
| 326 |
"step": 210
|
| 327 |
},
|
| 328 |
{
|
| 329 |
+
"epoch": 2.6219512195121952,
|
| 330 |
+
"grad_norm": 0.9921875,
|
| 331 |
"learning_rate": 0.00018439627758611385,
|
| 332 |
+
"loss": 0.6005,
|
| 333 |
"step": 215
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"epoch": 2.682926829268293,
|
| 337 |
+
"grad_norm": 1.0,
|
| 338 |
"learning_rate": 0.00018323554302364272,
|
| 339 |
+
"loss": 0.5917,
|
| 340 |
"step": 220
|
| 341 |
},
|
| 342 |
{
|
| 343 |
+
"epoch": 2.7439024390243905,
|
| 344 |
+
"grad_norm": 1.8515625,
|
| 345 |
"learning_rate": 0.00018203710168075788,
|
| 346 |
+
"loss": 0.5894,
|
| 347 |
"step": 225
|
| 348 |
},
|
| 349 |
{
|
| 350 |
+
"epoch": 2.8048780487804876,
|
| 351 |
+
"grad_norm": 1.15625,
|
| 352 |
"learning_rate": 0.0001808014964669293,
|
| 353 |
+
"loss": 0.5884,
|
| 354 |
"step": 230
|
| 355 |
},
|
| 356 |
{
|
| 357 |
+
"epoch": 2.8658536585365852,
|
| 358 |
+
"grad_norm": 0.83984375,
|
| 359 |
"learning_rate": 0.00017952928712734268,
|
| 360 |
+
"loss": 0.5967,
|
| 361 |
"step": 235
|
| 362 |
},
|
| 363 |
{
|
| 364 |
+
"epoch": 2.926829268292683,
|
| 365 |
+
"grad_norm": 0.3359375,
|
| 366 |
"learning_rate": 0.00017822104998932713,
|
| 367 |
+
"loss": 0.6064,
|
| 368 |
"step": 240
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 2.9878048780487805,
|
| 372 |
+
"grad_norm": 0.51171875,
|
| 373 |
"learning_rate": 0.00017687737770127185,
|
| 374 |
+
"loss": 0.5836,
|
| 375 |
"step": 245
|
| 376 |
},
|
| 377 |
{
|
| 378 |
+
"epoch": 3.0,
|
| 379 |
+
"eval_loss": 1.1479520797729492,
|
| 380 |
+
"eval_runtime": 1.1346,
|
| 381 |
+
"eval_samples_per_second": 1.763,
|
| 382 |
+
"eval_steps_per_second": 0.881,
|
| 383 |
+
"step": 246
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.048780487804878,
|
| 387 |
+
"grad_norm": 1.078125,
|
| 388 |
"learning_rate": 0.00017549887896414851,
|
| 389 |
+
"loss": 0.5409,
|
| 390 |
"step": 250
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"epoch": 3.1097560975609757,
|
| 394 |
+
"grad_norm": 0.361328125,
|
| 395 |
"learning_rate": 0.0001740861782557618,
|
| 396 |
+
"loss": 0.5316,
|
| 397 |
"step": 255
|
| 398 |
},
|
| 399 |
{
|
| 400 |
+
"epoch": 3.1707317073170733,
|
| 401 |
+
"grad_norm": 0.59765625,
|
| 402 |
"learning_rate": 0.0001726399155478529,
|
| 403 |
+
"loss": 0.5168,
|
| 404 |
"step": 260
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 3.231707317073171,
|
| 408 |
+
"grad_norm": 0.40234375,
|
| 409 |
"learning_rate": 0.00017116074601618417,
|
| 410 |
+
"loss": 0.5243,
|
| 411 |
"step": 265
|
| 412 |
},
|
| 413 |
{
|
| 414 |
+
"epoch": 3.292682926829268,
|
| 415 |
+
"grad_norm": 0.7890625,
|
| 416 |
"learning_rate": 0.0001696493397437357,
|
| 417 |
+
"loss": 0.5076,
|
| 418 |
"step": 270
|
| 419 |
},
|
| 420 |
{
|
| 421 |
+
"epoch": 3.3536585365853657,
|
| 422 |
+
"grad_norm": 0.96875,
|
| 423 |
"learning_rate": 0.00016810638141714934,
|
| 424 |
+
"loss": 0.5251,
|
| 425 |
"step": 275
|
| 426 |
},
|
| 427 |
{
|
| 428 |
+
"epoch": 3.4146341463414633,
|
| 429 |
+
"grad_norm": 0.890625,
|
| 430 |
"learning_rate": 0.00016653257001655652,
|
| 431 |
+
"loss": 0.5379,
|
| 432 |
"step": 280
|
| 433 |
},
|
| 434 |
{
|
| 435 |
+
"epoch": 3.475609756097561,
|
| 436 |
+
"grad_norm": 0.7109375,
|
| 437 |
"learning_rate": 0.0001649286184989315,
|
| 438 |
+
"loss": 0.5421,
|
| 439 |
"step": 285
|
| 440 |
},
|
| 441 |
{
|
| 442 |
+
"epoch": 3.5365853658536586,
|
| 443 |
+
"grad_norm": 0.59765625,
|
| 444 |
"learning_rate": 0.0001632952534751122,
|
| 445 |
+
"loss": 0.5148,
|
| 446 |
"step": 290
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 3.597560975609756,
|
| 450 |
+
"grad_norm": 0.490234375,
|
| 451 |
"learning_rate": 0.00016163321488063637,
|
| 452 |
+
"loss": 0.5222,
|
| 453 |
"step": 295
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"epoch": 3.658536585365854,
|
| 457 |
+
"grad_norm": 0.55078125,
|
| 458 |
"learning_rate": 0.00015994325564054122,
|
| 459 |
+
"loss": 0.5229,
|
| 460 |
"step": 300
|
| 461 |
},
|
| 462 |
{
|
| 463 |
+
"epoch": 3.7195121951219514,
|
| 464 |
+
"grad_norm": 0.515625,
|
| 465 |
"learning_rate": 0.00015822614132827837,
|
| 466 |
+
"loss": 0.532,
|
| 467 |
"step": 305
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 3.7804878048780486,
|
| 471 |
+
"grad_norm": 0.458984375,
|
| 472 |
"learning_rate": 0.00015648264981889934,
|
| 473 |
+
"loss": 0.5401,
|
| 474 |
"step": 310
|
| 475 |
},
|
| 476 |
{
|
| 477 |
+
"epoch": 3.841463414634146,
|
| 478 |
+
"grad_norm": 0.453125,
|
| 479 |
"learning_rate": 0.00015471357093666804,
|
| 480 |
+
"loss": 0.5311,
|
| 481 |
"step": 315
|
| 482 |
},
|
| 483 |
{
|
| 484 |
+
"epoch": 3.902439024390244,
|
| 485 |
+
"grad_norm": 0.396484375,
|
| 486 |
"learning_rate": 0.00015291970609726007,
|
| 487 |
+
"loss": 0.525,
|
| 488 |
"step": 320
|
| 489 |
},
|
| 490 |
{
|
| 491 |
+
"epoch": 3.9634146341463414,
|
| 492 |
+
"grad_norm": 0.359375,
|
| 493 |
"learning_rate": 0.00015110186794471103,
|
| 494 |
+
"loss": 0.5178,
|
| 495 |
"step": 325
|
| 496 |
},
|
| 497 |
{
|
| 498 |
+
"epoch": 4.0,
|
| 499 |
+
"eval_loss": 1.171655535697937,
|
| 500 |
+
"eval_runtime": 1.117,
|
| 501 |
+
"eval_samples_per_second": 1.791,
|
| 502 |
+
"eval_steps_per_second": 0.895,
|
| 503 |
"step": 328
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 4.024390243902439,
|
| 507 |
+
"grad_norm": 0.8125,
|
| 508 |
"learning_rate": 0.00014926087998327837,
|
| 509 |
+
"loss": 0.5073,
|
| 510 |
"step": 330
|
| 511 |
},
|
| 512 |
{
|
| 513 |
+
"epoch": 4.085365853658536,
|
| 514 |
+
"grad_norm": 0.3671875,
|
| 515 |
"learning_rate": 0.00014739757620438307,
|
| 516 |
+
"loss": 0.4417,
|
| 517 |
"step": 335
|
| 518 |
},
|
| 519 |
{
|
| 520 |
+
"epoch": 4.146341463414634,
|
| 521 |
+
"grad_norm": 0.7265625,
|
| 522 |
"learning_rate": 0.0001455128007088009,
|
| 523 |
+
"loss": 0.4321,
|
| 524 |
"step": 340
|
| 525 |
},
|
| 526 |
{
|
| 527 |
+
"epoch": 4.2073170731707314,
|
| 528 |
+
"grad_norm": 0.4921875,
|
| 529 |
"learning_rate": 0.00014360740732427367,
|
| 530 |
+
"loss": 0.4481,
|
| 531 |
"step": 345
|
| 532 |
},
|
| 533 |
{
|
| 534 |
+
"epoch": 4.2682926829268295,
|
| 535 |
+
"grad_norm": 0.75390625,
|
| 536 |
"learning_rate": 0.00014168225921871433,
|
| 537 |
+
"loss": 0.4489,
|
| 538 |
"step": 350
|
| 539 |
},
|
| 540 |
{
|
| 541 |
+
"epoch": 4.329268292682927,
|
| 542 |
+
"grad_norm": 0.52734375,
|
| 543 |
"learning_rate": 0.00013973822850918055,
|
| 544 |
+
"loss": 0.4517,
|
| 545 |
"step": 355
|
| 546 |
},
|
| 547 |
{
|
| 548 |
+
"epoch": 4.390243902439025,
|
| 549 |
+
"grad_norm": 0.78125,
|
| 550 |
"learning_rate": 0.0001377761958667946,
|
| 551 |
+
"loss": 0.4441,
|
| 552 |
"step": 360
|
| 553 |
},
|
| 554 |
{
|
| 555 |
+
"epoch": 4.451219512195122,
|
| 556 |
+
"grad_norm": 0.59375,
|
| 557 |
"learning_rate": 0.00013579705011778766,
|
| 558 |
+
"loss": 0.4539,
|
| 559 |
"step": 365
|
| 560 |
},
|
| 561 |
{
|
| 562 |
+
"epoch": 4.512195121951219,
|
| 563 |
+
"grad_norm": 0.578125,
|
| 564 |
"learning_rate": 0.00013380168784085027,
|
| 565 |
+
"loss": 0.4489,
|
| 566 |
"step": 370
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 4.573170731707317,
|
| 570 |
+
"grad_norm": 0.70703125,
|
| 571 |
"learning_rate": 0.00013179101296097035,
|
| 572 |
+
"loss": 0.4443,
|
| 573 |
"step": 375
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"epoch": 4.634146341463414,
|
| 577 |
+
"grad_norm": 0.6796875,
|
| 578 |
"learning_rate": 0.00012976593633994346,
|
| 579 |
+
"loss": 0.4664,
|
| 580 |
"step": 380
|
| 581 |
},
|
| 582 |
{
|
| 583 |
+
"epoch": 4.695121951219512,
|
| 584 |
+
"grad_norm": 0.43359375,
|
| 585 |
"learning_rate": 0.0001277273753637408,
|
| 586 |
+
"loss": 0.4572,
|
| 587 |
"step": 385
|
| 588 |
},
|
| 589 |
{
|
| 590 |
+
"epoch": 4.7560975609756095,
|
| 591 |
+
"grad_norm": 0.84765625,
|
| 592 |
"learning_rate": 0.00012567625352692127,
|
| 593 |
+
"loss": 0.4675,
|
| 594 |
"step": 390
|
| 595 |
},
|
| 596 |
{
|
| 597 |
+
"epoch": 4.817073170731708,
|
| 598 |
+
"grad_norm": 0.486328125,
|
| 599 |
"learning_rate": 0.0001236135000142765,
|
| 600 |
+
"loss": 0.45,
|
| 601 |
"step": 395
|
| 602 |
},
|
| 603 |
{
|
| 604 |
+
"epoch": 4.878048780487805,
|
| 605 |
+
"grad_norm": 0.4140625,
|
| 606 |
"learning_rate": 0.00012154004927989815,
|
| 607 |
+
"loss": 0.4538,
|
| 608 |
"step": 400
|
| 609 |
},
|
| 610 |
{
|
| 611 |
+
"epoch": 4.939024390243903,
|
| 612 |
+
"grad_norm": 0.3984375,
|
| 613 |
"learning_rate": 0.00011945684062385803,
|
| 614 |
+
"loss": 0.4565,
|
| 615 |
"step": 405
|
| 616 |
},
|
| 617 |
{
|
| 618 |
+
"epoch": 5.0,
|
| 619 |
+
"grad_norm": 0.4140625,
|
| 620 |
"learning_rate": 0.00011736481776669306,
|
| 621 |
+
"loss": 0.4668,
|
| 622 |
"step": 410
|
| 623 |
},
|
| 624 |
{
|
| 625 |
+
"epoch": 5.0,
|
| 626 |
+
"eval_loss": 1.2044252157211304,
|
| 627 |
+
"eval_runtime": 1.1171,
|
| 628 |
+
"eval_samples_per_second": 1.79,
|
| 629 |
+
"eval_steps_per_second": 0.895,
|
| 630 |
+
"step": 410
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 5.060975609756097,
|
| 634 |
+
"grad_norm": 0.63671875,
|
| 635 |
"learning_rate": 0.00011526492842188745,
|
| 636 |
+
"loss": 0.3794,
|
| 637 |
"step": 415
|
| 638 |
},
|
| 639 |
{
|
| 640 |
+
"epoch": 5.121951219512195,
|
| 641 |
+
"grad_norm": 0.83984375,
|
| 642 |
"learning_rate": 0.0001131581238665465,
|
| 643 |
+
"loss": 0.376,
|
| 644 |
"step": 420
|
| 645 |
},
|
| 646 |
{
|
| 647 |
+
"epoch": 5.182926829268292,
|
| 648 |
+
"grad_norm": 0.453125,
|
| 649 |
"learning_rate": 0.00011104535851045539,
|
| 650 |
+
"loss": 0.3721,
|
| 651 |
"step": 425
|
| 652 |
},
|
| 653 |
{
|
| 654 |
+
"epoch": 5.2439024390243905,
|
| 655 |
+
"grad_norm": 0.470703125,
|
| 656 |
"learning_rate": 0.00010892758946371944,
|
| 657 |
+
"loss": 0.3812,
|
| 658 |
"step": 430
|
| 659 |
},
|
| 660 |
{
|
| 661 |
+
"epoch": 5.304878048780488,
|
| 662 |
+
"grad_norm": 0.78125,
|
| 663 |
"learning_rate": 0.00010680577610318072,
|
| 664 |
+
"loss": 0.3748,
|
| 665 |
"step": 435
|
| 666 |
},
|
| 667 |
{
|
| 668 |
+
"epoch": 5.365853658536586,
|
| 669 |
+
"grad_norm": 0.7578125,
|
| 670 |
"learning_rate": 0.00010468087963780789,
|
| 671 |
+
"loss": 0.3698,
|
| 672 |
"step": 440
|
| 673 |
},
|
| 674 |
{
|
| 675 |
+
"epoch": 5.426829268292683,
|
| 676 |
+
"grad_norm": 0.48828125,
|
| 677 |
"learning_rate": 0.00010255386267325602,
|
| 678 |
+
"loss": 0.3849,
|
| 679 |
"step": 445
|
| 680 |
},
|
| 681 |
{
|
| 682 |
+
"epoch": 5.487804878048781,
|
| 683 |
+
"grad_norm": 0.45703125,
|
| 684 |
"learning_rate": 0.00010042568877579388,
|
| 685 |
+
"loss": 0.3831,
|
| 686 |
"step": 450
|
| 687 |
},
|
| 688 |
{
|
| 689 |
+
"epoch": 5.548780487804878,
|
| 690 |
+
"grad_norm": 0.470703125,
|
| 691 |
"learning_rate": 9.829732203579584e-05,
|
| 692 |
+
"loss": 0.3765,
|
| 693 |
"step": 455
|
| 694 |
},
|
| 695 |
{
|
| 696 |
+
"epoch": 5.609756097560975,
|
| 697 |
+
"grad_norm": 0.58984375,
|
| 698 |
"learning_rate": 9.616972663099647e-05,
|
| 699 |
+
"loss": 0.4006,
|
| 700 |
"step": 460
|
| 701 |
},
|
| 702 |
{
|
| 703 |
+
"epoch": 5.670731707317073,
|
| 704 |
+
"grad_norm": 0.4609375,
|
| 705 |
"learning_rate": 9.404386638970542e-05,
|
| 706 |
+
"loss": 0.384,
|
| 707 |
"step": 465
|
| 708 |
},
|
| 709 |
{
|
| 710 |
+
"epoch": 5.7317073170731705,
|
| 711 |
+
"grad_norm": 0.41796875,
|
| 712 |
"learning_rate": 9.192070435418079e-05,
|
| 713 |
+
"loss": 0.3804,
|
| 714 |
"step": 470
|
| 715 |
},
|
| 716 |
{
|
| 717 |
+
"epoch": 5.7926829268292686,
|
| 718 |
+
"grad_norm": 0.56640625,
|
| 719 |
"learning_rate": 8.980120234435849e-05,
|
| 720 |
+
"loss": 0.39,
|
| 721 |
"step": 475
|
| 722 |
},
|
| 723 |
{
|
| 724 |
+
"epoch": 5.853658536585366,
|
| 725 |
+
"grad_norm": 0.7265625,
|
| 726 |
"learning_rate": 8.768632052213531e-05,
|
| 727 |
+
"loss": 0.3881,
|
| 728 |
"step": 480
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 5.914634146341464,
|
| 732 |
+
"grad_norm": 0.44921875,
|
| 733 |
"learning_rate": 8.557701695640321e-05,
|
| 734 |
+
"loss": 0.3908,
|
| 735 |
"step": 485
|
| 736 |
},
|
| 737 |
{
|
| 738 |
+
"epoch": 5.975609756097561,
|
| 739 |
+
"grad_norm": 0.4140625,
|
| 740 |
"learning_rate": 8.347424718903151e-05,
|
| 741 |
+
"loss": 0.3955,
|
| 742 |
"step": 490
|
| 743 |
},
|
| 744 |
{
|
| 745 |
+
"epoch": 6.0,
|
| 746 |
+
"eval_loss": 1.325190782546997,
|
| 747 |
+
"eval_runtime": 1.1255,
|
| 748 |
+
"eval_samples_per_second": 1.777,
|
| 749 |
+
"eval_steps_per_second": 0.888,
|
| 750 |
"step": 492
|
| 751 |
},
|
| 752 |
{
|
| 753 |
+
"epoch": 6.036585365853658,
|
| 754 |
+
"grad_norm": 0.3984375,
|
| 755 |
"learning_rate": 8.13789638019942e-05,
|
| 756 |
+
"loss": 0.3366,
|
| 757 |
"step": 495
|
| 758 |
},
|
| 759 |
{
|
| 760 |
+
"epoch": 6.097560975609756,
|
| 761 |
+
"grad_norm": 0.408203125,
|
| 762 |
"learning_rate": 7.929211598583794e-05,
|
| 763 |
+
"loss": 0.3141,
|
| 764 |
"step": 500
|
| 765 |
},
|
| 766 |
{
|
| 767 |
+
"epoch": 6.158536585365853,
|
| 768 |
+
"grad_norm": 0.54296875,
|
| 769 |
"learning_rate": 7.721464910968627e-05,
|
| 770 |
+
"loss": 0.3234,
|
| 771 |
"step": 505
|
| 772 |
},
|
| 773 |
{
|
| 774 |
+
"epoch": 6.219512195121951,
|
| 775 |
+
"grad_norm": 0.423828125,
|
| 776 |
"learning_rate": 7.514750429297528e-05,
|
| 777 |
+
"loss": 0.3232,
|
| 778 |
"step": 510
|
| 779 |
},
|
| 780 |
{
|
| 781 |
+
"epoch": 6.280487804878049,
|
| 782 |
+
"grad_norm": 0.4921875,
|
| 783 |
"learning_rate": 7.309161797911441e-05,
|
| 784 |
+
"loss": 0.3198,
|
| 785 |
"step": 515
|
| 786 |
},
|
| 787 |
{
|
| 788 |
+
"epoch": 6.341463414634147,
|
| 789 |
+
"grad_norm": 0.43359375,
|
| 790 |
"learning_rate": 7.104792151126515e-05,
|
| 791 |
+
"loss": 0.3236,
|
| 792 |
"step": 520
|
| 793 |
},
|
| 794 |
{
|
| 795 |
+
"epoch": 6.402439024390244,
|
| 796 |
+
"grad_norm": 0.447265625,
|
| 797 |
"learning_rate": 6.901734071043071e-05,
|
| 798 |
+
"loss": 0.3126,
|
| 799 |
"step": 525
|
| 800 |
},
|
| 801 |
{
|
| 802 |
+
"epoch": 6.463414634146342,
|
| 803 |
+
"grad_norm": 0.41015625,
|
| 804 |
"learning_rate": 6.700079545604708e-05,
|
| 805 |
+
"loss": 0.3167,
|
| 806 |
"step": 530
|
| 807 |
},
|
| 808 |
{
|
| 809 |
+
"epoch": 6.524390243902439,
|
| 810 |
+
"grad_norm": 0.52734375,
|
| 811 |
"learning_rate": 6.499919926926566e-05,
|
| 812 |
+
"loss": 0.3412,
|
| 813 |
"step": 535
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"epoch": 6.585365853658536,
|
| 817 |
+
"grad_norm": 0.5703125,
|
| 818 |
"learning_rate": 6.301345889911637e-05,
|
| 819 |
+
"loss": 0.3269,
|
| 820 |
"step": 540
|
| 821 |
},
|
| 822 |
{
|
| 823 |
+
"epoch": 6.646341463414634,
|
| 824 |
+
"grad_norm": 0.423828125,
|
| 825 |
"learning_rate": 6.104447391173858e-05,
|
| 826 |
+
"loss": 0.3309,
|
| 827 |
"step": 545
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 6.7073170731707314,
|
| 831 |
+
"grad_norm": 0.482421875,
|
| 832 |
"learning_rate": 5.909313628286601e-05,
|
| 833 |
+
"loss": 0.3172,
|
| 834 |
"step": 550
|
| 835 |
},
|
| 836 |
{
|
| 837 |
+
"epoch": 6.7682926829268295,
|
| 838 |
+
"grad_norm": 0.53125,
|
| 839 |
"learning_rate": 5.716032999375006e-05,
|
| 840 |
+
"loss": 0.326,
|
| 841 |
"step": 555
|
| 842 |
},
|
| 843 |
{
|
| 844 |
+
"epoch": 6.829268292682927,
|
| 845 |
+
"grad_norm": 0.439453125,
|
| 846 |
"learning_rate": 5.524693063070492e-05,
|
| 847 |
+
"loss": 0.3207,
|
| 848 |
"step": 560
|
| 849 |
},
|
| 850 |
{
|
| 851 |
+
"epoch": 6.890243902439025,
|
| 852 |
+
"grad_norm": 0.55859375,
|
| 853 |
"learning_rate": 5.335380498845559e-05,
|
| 854 |
+
"loss": 0.3295,
|
| 855 |
"step": 565
|
| 856 |
},
|
| 857 |
{
|
| 858 |
+
"epoch": 6.951219512195122,
|
| 859 |
+
"grad_norm": 0.47265625,
|
| 860 |
"learning_rate": 5.148181067746862e-05,
|
| 861 |
+
"loss": 0.3233,
|
| 862 |
"step": 570
|
| 863 |
},
|
| 864 |
{
|
| 865 |
+
"epoch": 7.0,
|
| 866 |
+
"eval_loss": 1.4225410223007202,
|
| 867 |
+
"eval_runtime": 1.1188,
|
| 868 |
+
"eval_samples_per_second": 1.788,
|
| 869 |
+
"eval_steps_per_second": 0.894,
|
| 870 |
+
"step": 574
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 7.012195121951219,
|
| 874 |
+
"grad_norm": 0.421875,
|
| 875 |
"learning_rate": 4.963179573544357e-05,
|
| 876 |
+
"loss": 0.3129,
|
| 877 |
"step": 575
|
| 878 |
},
|
| 879 |
{
|
| 880 |
+
"epoch": 7.073170731707317,
|
| 881 |
+
"grad_norm": 0.57421875,
|
| 882 |
"learning_rate": 4.7804598243140666e-05,
|
| 883 |
+
"loss": 0.2764,
|
| 884 |
"step": 580
|
| 885 |
},
|
| 886 |
{
|
| 887 |
+
"epoch": 7.134146341463414,
|
| 888 |
+
"grad_norm": 0.5,
|
| 889 |
"learning_rate": 4.60010459447196e-05,
|
| 890 |
+
"loss": 0.2792,
|
| 891 |
"step": 585
|
| 892 |
},
|
| 893 |
{
|
| 894 |
+
"epoch": 7.195121951219512,
|
| 895 |
+
"grad_norm": 0.486328125,
|
| 896 |
"learning_rate": 4.422195587276058e-05,
|
| 897 |
+
"loss": 0.2799,
|
| 898 |
"step": 590
|
| 899 |
},
|
| 900 |
{
|
| 901 |
+
"epoch": 7.2560975609756095,
|
| 902 |
+
"grad_norm": 0.47265625,
|
| 903 |
"learning_rate": 4.2468133978137945e-05,
|
| 904 |
+
"loss": 0.2759,
|
| 905 |
"step": 595
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 7.317073170731708,
|
| 909 |
+
"grad_norm": 0.48046875,
|
| 910 |
"learning_rate": 4.0740374764914136e-05,
|
| 911 |
+
"loss": 0.2697,
|
| 912 |
"step": 600
|
| 913 |
},
|
| 914 |
{
|
| 915 |
+
"epoch": 7.378048780487805,
|
| 916 |
+
"grad_norm": 0.439453125,
|
| 917 |
"learning_rate": 3.903946093041877e-05,
|
| 918 |
+
"loss": 0.2917,
|
| 919 |
"step": 605
|
| 920 |
},
|
| 921 |
{
|
| 922 |
+
"epoch": 7.439024390243903,
|
| 923 |
+
"grad_norm": 0.47265625,
|
| 924 |
"learning_rate": 3.736616301067694e-05,
|
| 925 |
+
"loss": 0.2748,
|
| 926 |
"step": 610
|
| 927 |
},
|
| 928 |
{
|
| 929 |
+
"epoch": 7.5,
|
| 930 |
+
"grad_norm": 0.50390625,
|
| 931 |
"learning_rate": 3.5721239031346066e-05,
|
| 932 |
+
"loss": 0.2771,
|
| 933 |
"step": 615
|
| 934 |
},
|
| 935 |
{
|
| 936 |
+
"epoch": 7.560975609756097,
|
| 937 |
+
"grad_norm": 0.5,
|
| 938 |
"learning_rate": 3.410543416432069e-05,
|
| 939 |
+
"loss": 0.2788,
|
| 940 |
"step": 620
|
| 941 |
},
|
| 942 |
{
|
| 943 |
+
"epoch": 7.621951219512195,
|
| 944 |
+
"grad_norm": 0.44921875,
|
| 945 |
"learning_rate": 3.2519480390159806e-05,
|
| 946 |
+
"loss": 0.2829,
|
| 947 |
"step": 625
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 7.682926829268292,
|
| 951 |
+
"grad_norm": 0.494140625,
|
| 952 |
"learning_rate": 3.096409616649023e-05,
|
| 953 |
+
"loss": 0.2708,
|
| 954 |
"step": 630
|
| 955 |
},
|
| 956 |
{
|
| 957 |
+
"epoch": 7.7439024390243905,
|
| 958 |
+
"grad_norm": 0.46875,
|
| 959 |
"learning_rate": 2.9439986102536043e-05,
|
| 960 |
+
"loss": 0.2945,
|
| 961 |
"step": 635
|
| 962 |
},
|
| 963 |
{
|
| 964 |
+
"epoch": 7.804878048780488,
|
| 965 |
+
"grad_norm": 0.482421875,
|
| 966 |
"learning_rate": 2.794784063992131e-05,
|
| 967 |
+
"loss": 0.2833,
|
| 968 |
"step": 640
|
| 969 |
},
|
| 970 |
{
|
| 971 |
+
"epoch": 7.865853658536586,
|
| 972 |
+
"grad_norm": 0.4375,
|
| 973 |
"learning_rate": 2.6488335739891178e-05,
|
| 974 |
+
"loss": 0.2767,
|
| 975 |
"step": 645
|
| 976 |
},
|
| 977 |
{
|
| 978 |
+
"epoch": 7.926829268292683,
|
| 979 |
+
"grad_norm": 0.46875,
|
| 980 |
"learning_rate": 2.50621325770927e-05,
|
| 981 |
+
"loss": 0.2725,
|
| 982 |
"step": 650
|
| 983 |
},
|
| 984 |
{
|
| 985 |
+
"epoch": 7.987804878048781,
|
| 986 |
+
"grad_norm": 0.4453125,
|
| 987 |
"learning_rate": 2.366987724005404e-05,
|
| 988 |
+
"loss": 0.2669,
|
| 989 |
"step": 655
|
| 990 |
},
|
| 991 |
{
|
| 992 |
+
"epoch": 8.0,
|
| 993 |
+
"eval_loss": 1.611946702003479,
|
| 994 |
+
"eval_runtime": 1.1206,
|
| 995 |
+
"eval_samples_per_second": 1.785,
|
| 996 |
+
"eval_steps_per_second": 0.892,
|
| 997 |
"step": 656
|
| 998 |
},
|
| 999 |
{
|
| 1000 |
+
"epoch": 8.048780487804878,
|
| 1001 |
+
"grad_norm": 0.435546875,
|
| 1002 |
"learning_rate": 2.2312200438498043e-05,
|
| 1003 |
+
"loss": 0.2564,
|
| 1004 |
"step": 660
|
| 1005 |
},
|
| 1006 |
{
|
| 1007 |
+
"epoch": 8.109756097560975,
|
| 1008 |
+
"grad_norm": 0.447265625,
|
| 1009 |
"learning_rate": 2.0989717217622652e-05,
|
| 1010 |
+
"loss": 0.2598,
|
| 1011 |
"step": 665
|
| 1012 |
},
|
| 1013 |
{
|
| 1014 |
+
"epoch": 8.170731707317072,
|
| 1015 |
+
"grad_norm": 0.4921875,
|
| 1016 |
"learning_rate": 1.9703026679477256e-05,
|
| 1017 |
+
"loss": 0.2507,
|
| 1018 |
"step": 670
|
| 1019 |
},
|
| 1020 |
{
|
| 1021 |
+
"epoch": 8.231707317073171,
|
| 1022 |
+
"grad_norm": 0.3984375,
|
| 1023 |
"learning_rate": 1.8452711711561842e-05,
|
| 1024 |
+
"loss": 0.2433,
|
| 1025 |
"step": 675
|
| 1026 |
},
|
| 1027 |
{
|
| 1028 |
+
"epoch": 8.292682926829269,
|
| 1029 |
+
"grad_norm": 0.408203125,
|
| 1030 |
"learning_rate": 1.7239338722771327e-05,
|
| 1031 |
+
"loss": 0.251,
|
| 1032 |
"step": 680
|
| 1033 |
},
|
| 1034 |
{
|
| 1035 |
+
"epoch": 8.353658536585366,
|
| 1036 |
+
"grad_norm": 0.40234375,
|
| 1037 |
"learning_rate": 1.6063457386805004e-05,
|
| 1038 |
+
"loss": 0.252,
|
| 1039 |
"step": 685
|
| 1040 |
},
|
| 1041 |
{
|
| 1042 |
+
"epoch": 8.414634146341463,
|
| 1043 |
+
"grad_norm": 0.404296875,
|
| 1044 |
"learning_rate": 1.4925600393157324e-05,
|
| 1045 |
+
"loss": 0.2471,
|
| 1046 |
"step": 690
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
+
"epoch": 8.475609756097562,
|
| 1050 |
+
"grad_norm": 0.494140625,
|
| 1051 |
"learning_rate": 1.3826283205802427e-05,
|
| 1052 |
+
"loss": 0.2526,
|
| 1053 |
"step": 695
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
+
"epoch": 8.536585365853659,
|
| 1057 |
+
"grad_norm": 0.423828125,
|
| 1058 |
"learning_rate": 1.2766003829682505e-05,
|
| 1059 |
+
"loss": 0.2517,
|
| 1060 |
"step": 700
|
| 1061 |
},
|
| 1062 |
{
|
| 1063 |
+
"epoch": 8.597560975609756,
|
| 1064 |
+
"grad_norm": 0.4375,
|
| 1065 |
"learning_rate": 1.1745242585104955e-05,
|
| 1066 |
+
"loss": 0.2479,
|
| 1067 |
"step": 705
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
+
"epoch": 8.658536585365853,
|
| 1071 |
+
"grad_norm": 0.404296875,
|
| 1072 |
"learning_rate": 1.0764461890151112e-05,
|
| 1073 |
+
"loss": 0.2485,
|
| 1074 |
"step": 710
|
| 1075 |
},
|
| 1076 |
{
|
| 1077 |
+
"epoch": 8.71951219512195,
|
| 1078 |
+
"grad_norm": 0.44140625,
|
| 1079 |
"learning_rate": 9.824106051194859e-06,
|
| 1080 |
+
"loss": 0.2559,
|
| 1081 |
"step": 715
|
| 1082 |
},
|
| 1083 |
{
|
| 1084 |
+
"epoch": 8.78048780487805,
|
| 1085 |
+
"grad_norm": 0.40234375,
|
| 1086 |
"learning_rate": 8.924601061626048e-06,
|
| 1087 |
+
"loss": 0.2588,
|
| 1088 |
"step": 720
|
| 1089 |
},
|
| 1090 |
{
|
| 1091 |
+
"epoch": 8.841463414634147,
|
| 1092 |
+
"grad_norm": 0.40625,
|
| 1093 |
"learning_rate": 8.066354408870048e-06,
|
| 1094 |
+
"loss": 0.2518,
|
| 1095 |
"step": 725
|
| 1096 |
},
|
| 1097 |
{
|
| 1098 |
+
"epoch": 8.902439024390244,
|
| 1099 |
+
"grad_norm": 0.42578125,
|
| 1100 |
"learning_rate": 7.249754889790539e-06,
|
| 1101 |
+
"loss": 0.2542,
|
| 1102 |
"step": 730
|
| 1103 |
},
|
| 1104 |
{
|
| 1105 |
+
"epoch": 8.963414634146341,
|
| 1106 |
+
"grad_norm": 0.453125,
|
| 1107 |
"learning_rate": 6.475172434559573e-06,
|
| 1108 |
+
"loss": 0.2591,
|
| 1109 |
"step": 735
|
| 1110 |
},
|
| 1111 |
{
|
| 1112 |
+
"epoch": 9.0,
|
| 1113 |
+
"eval_loss": 1.73529052734375,
|
| 1114 |
+
"eval_runtime": 1.1218,
|
| 1115 |
+
"eval_samples_per_second": 1.783,
|
| 1116 |
+
"eval_steps_per_second": 0.891,
|
| 1117 |
+
"step": 738
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 9.024390243902438,
|
| 1121 |
+
"grad_norm": 0.419921875,
|
| 1122 |
"learning_rate": 5.742957939074412e-06,
|
| 1123 |
+
"loss": 0.2467,
|
| 1124 |
"step": 740
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
+
"epoch": 9.085365853658537,
|
| 1128 |
+
"grad_norm": 0.384765625,
|
| 1129 |
"learning_rate": 5.0534431059970685e-06,
|
| 1130 |
+
"loss": 0.2449,
|
| 1131 |
"step": 745
|
| 1132 |
},
|
| 1133 |
{
|
| 1134 |
+
"epoch": 9.146341463414634,
|
| 1135 |
+
"grad_norm": 0.38671875,
|
| 1136 |
"learning_rate": 4.40694029448877e-06,
|
| 1137 |
+
"loss": 0.244,
|
| 1138 |
"step": 750
|
| 1139 |
},
|
| 1140 |
{
|
| 1141 |
+
"epoch": 9.207317073170731,
|
| 1142 |
+
"grad_norm": 0.419921875,
|
| 1143 |
"learning_rate": 3.803742378707198e-06,
|
| 1144 |
+
"loss": 0.2612,
|
| 1145 |
"step": 755
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
+
"epoch": 9.268292682926829,
|
| 1149 |
+
"grad_norm": 0.384765625,
|
| 1150 |
"learning_rate": 3.2441226151306404e-06,
|
| 1151 |
+
"loss": 0.2497,
|
| 1152 |
"step": 760
|
| 1153 |
},
|
| 1154 |
{
|
| 1155 |
+
"epoch": 9.329268292682928,
|
| 1156 |
+
"grad_norm": 0.41796875,
|
| 1157 |
"learning_rate": 2.7283345187693264e-06,
|
| 1158 |
+
"loss": 0.2571,
|
| 1159 |
"step": 765
|
| 1160 |
},
|
| 1161 |
{
|
| 1162 |
+
"epoch": 9.390243902439025,
|
| 1163 |
+
"grad_norm": 0.40234375,
|
| 1164 |
"learning_rate": 2.256611748319792e-06,
|
| 1165 |
+
"loss": 0.2458,
|
| 1166 |
"step": 770
|
| 1167 |
},
|
| 1168 |
{
|
| 1169 |
+
"epoch": 9.451219512195122,
|
| 1170 |
+
"grad_norm": 0.427734375,
|
| 1171 |
"learning_rate": 1.8291680003145073e-06,
|
| 1172 |
+
"loss": 0.2585,
|
| 1173 |
"step": 775
|
| 1174 |
},
|
| 1175 |
{
|
| 1176 |
+
"epoch": 9.512195121951219,
|
| 1177 |
+
"grad_norm": 0.423828125,
|
| 1178 |
"learning_rate": 1.4461969123145457e-06,
|
| 1179 |
+
"loss": 0.2516,
|
| 1180 |
"step": 780
|
| 1181 |
},
|
| 1182 |
{
|
| 1183 |
+
"epoch": 9.573170731707316,
|
| 1184 |
+
"grad_norm": 0.39453125,
|
| 1185 |
"learning_rate": 1.107871975189234e-06,
|
| 1186 |
+
"loss": 0.2396,
|
| 1187 |
"step": 785
|
| 1188 |
},
|
| 1189 |
{
|
| 1190 |
+
"epoch": 9.634146341463415,
|
| 1191 |
+
"grad_norm": 0.380859375,
|
| 1192 |
"learning_rate": 8.143464545226298e-07,
|
| 1193 |
+
"loss": 0.2441,
|
| 1194 |
"step": 790
|
| 1195 |
},
|
| 1196 |
{
|
| 1197 |
+
"epoch": 9.695121951219512,
|
| 1198 |
+
"grad_norm": 0.3984375,
|
| 1199 |
"learning_rate": 5.657533211820942e-07,
|
| 1200 |
+
"loss": 0.2417,
|
| 1201 |
"step": 795
|
| 1202 |
},
|
| 1203 |
{
|
| 1204 |
+
"epoch": 9.75609756097561,
|
| 1205 |
+
"grad_norm": 0.400390625,
|
| 1206 |
"learning_rate": 3.622051910808666e-07,
|
| 1207 |
+
"loss": 0.2425,
|
| 1208 |
"step": 800
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
+
"epoch": 9.817073170731707,
|
| 1212 |
+
"grad_norm": 0.412109375,
|
| 1213 |
"learning_rate": 2.037942741615617e-07,
|
| 1214 |
+
"loss": 0.2447,
|
| 1215 |
"step": 805
|
| 1216 |
},
|
| 1217 |
{
|
| 1218 |
+
"epoch": 9.878048780487806,
|
| 1219 |
+
"grad_norm": 0.39453125,
|
| 1220 |
"learning_rate": 9.059233262386225e-08,
|
| 1221 |
+
"loss": 0.2512,
|
| 1222 |
"step": 810
|
| 1223 |
},
|
| 1224 |
{
|
| 1225 |
+
"epoch": 9.939024390243903,
|
| 1226 |
+
"grad_norm": 0.39453125,
|
| 1227 |
"learning_rate": 2.2650648415334376e-08,
|
| 1228 |
+
"loss": 0.2365,
|
| 1229 |
"step": 815
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
+
"epoch": 10.0,
|
| 1233 |
+
"grad_norm": 0.3828125,
|
| 1234 |
"learning_rate": 0.0,
|
| 1235 |
+
"loss": 0.2367,
|
| 1236 |
"step": 820
|
| 1237 |
},
|
| 1238 |
{
|
| 1239 |
+
"epoch": 10.0,
|
| 1240 |
+
"eval_loss": 1.750213384628296,
|
| 1241 |
+
"eval_runtime": 1.1197,
|
| 1242 |
+
"eval_samples_per_second": 1.786,
|
| 1243 |
+
"eval_steps_per_second": 0.893,
|
| 1244 |
"step": 820
|
| 1245 |
},
|
| 1246 |
{
|
| 1247 |
+
"epoch": 10.0,
|
| 1248 |
"step": 820,
|
| 1249 |
+
"total_flos": 1.257706785410646e+18,
|
| 1250 |
+
"train_loss": 1.5742560074096772,
|
| 1251 |
+
"train_runtime": 5911.531,
|
| 1252 |
+
"train_samples_per_second": 2.214,
|
| 1253 |
+
"train_steps_per_second": 0.139
|
| 1254 |
}
|
| 1255 |
],
|
| 1256 |
"logging_steps": 5,
|
| 1257 |
"max_steps": 820,
|
| 1258 |
"num_input_tokens_seen": 0,
|
| 1259 |
+
"num_train_epochs": 10,
|
| 1260 |
"save_steps": 100,
|
| 1261 |
+
"total_flos": 1.257706785410646e+18,
|
| 1262 |
"train_batch_size": 2,
|
| 1263 |
"trial_name": null,
|
| 1264 |
"trial_params": null
|