Model save
Browse files- README.md +6 -12
- all_results.json +4 -4
- config.json +1 -1
- model.safetensors +1 -1
- runs/Aug12_15-26-23_ip-26-0-165-24/events.out.tfevents.1723476428.ip-26-0-165-24.498584.0 +3 -0
- train_results.json +4 -4
- trainer_state.json +301 -301
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -2,18 +2,12 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: HuggingFaceTB/SmolLM-360M
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
-
- trl
|
| 7 |
-
- sft
|
| 8 |
-
- generated_from_trainer
|
| 9 |
- trl
|
| 10 |
- sft
|
|
|
|
| 11 |
- generated_from_trainer
|
| 12 |
datasets:
|
| 13 |
-
-
|
| 14 |
-
- HuggingFaceTB/self-oss-instruct-sc2-H4
|
| 15 |
-
- HuggingFaceTB/OpenHermes-2.5-H4
|
| 16 |
-
- HuggingFaceTB/instruct-data-basics-H4
|
| 17 |
model-index:
|
| 18 |
- name: smollm-350M-instruct-test2
|
| 19 |
results: []
|
|
@@ -22,12 +16,12 @@ model-index:
|
|
| 22 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 23 |
should probably proofread and complete it, then remove this comment. -->
|
| 24 |
|
| 25 |
-
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/loubnabnl/huggingface/runs/
|
| 26 |
# smollm-350M-instruct-test2
|
| 27 |
|
| 28 |
-
This model is a fine-tuned version of [HuggingFaceTB/SmolLM-360M](https://huggingface.co/HuggingFaceTB/SmolLM-360M) on the
|
| 29 |
It achieves the following results on the evaluation set:
|
| 30 |
-
- Loss: 1.
|
| 31 |
|
| 32 |
## Model description
|
| 33 |
|
|
@@ -64,7 +58,7 @@ The following hyperparameters were used during training:
|
|
| 64 |
|
| 65 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 66 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 67 |
-
| 0.
|
| 68 |
|
| 69 |
|
| 70 |
### Framework versions
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: HuggingFaceTB/SmolLM-360M
|
| 4 |
tags:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
+
- alignment-handbook
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
|
|
|
|
|
|
|
|
|
| 11 |
model-index:
|
| 12 |
- name: smollm-350M-instruct-test2
|
| 13 |
results: []
|
|
|
|
| 16 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
|
| 19 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/loubnabnl/huggingface/runs/0gd1f5vs)
|
| 20 |
# smollm-350M-instruct-test2
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [HuggingFaceTB/SmolLM-360M](https://huggingface.co/HuggingFaceTB/SmolLM-360M) on the generator dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 1.2024
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
|
|
| 58 |
|
| 59 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 60 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 61 |
+
| 0.8401 | 1.0 | 816 | 1.2024 |
|
| 62 |
|
| 63 |
|
| 64 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -6,9 +6,9 @@
|
|
| 6 |
"eval_samples_per_second": 189.723,
|
| 7 |
"eval_steps_per_second": 5.934,
|
| 8 |
"total_flos": 80063181619200.0,
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 319078,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 6 |
"eval_samples_per_second": 189.723,
|
| 7 |
"eval_steps_per_second": 5.934,
|
| 8 |
"total_flos": 80063181619200.0,
|
| 9 |
+
"train_loss": 0.914715180794398,
|
| 10 |
+
"train_runtime": 1874.4496,
|
| 11 |
"train_samples": 319078,
|
| 12 |
+
"train_samples_per_second": 55.714,
|
| 13 |
+
"train_steps_per_second": 0.435
|
| 14 |
}
|
config.json
CHANGED
|
@@ -25,6 +25,6 @@
|
|
| 25 |
"tie_word_embeddings": true,
|
| 26 |
"torch_dtype": "bfloat16",
|
| 27 |
"transformers_version": "4.42.3",
|
| 28 |
-
"use_cache":
|
| 29 |
"vocab_size": 49152
|
| 30 |
}
|
|
|
|
| 25 |
"tie_word_embeddings": true,
|
| 26 |
"torch_dtype": "bfloat16",
|
| 27 |
"transformers_version": "4.42.3",
|
| 28 |
+
"use_cache": false,
|
| 29 |
"vocab_size": 49152
|
| 30 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 723674912
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9807c299d8bab57ce4b5ad8aed938ec75a2a49cf79a22d3ee29cca105e2f5d4
|
| 3 |
size 723674912
|
runs/Aug12_15-26-23_ip-26-0-165-24/events.out.tfevents.1723476428.ip-26-0-165-24.498584.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7522cf76d54904b417764e96888e83230877da42b81aee8eb6296aa0d6b01408
|
| 3 |
+
size 40257
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
"total_flos": 80063181619200.0,
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 319078,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 1.0,
|
| 3 |
"total_flos": 80063181619200.0,
|
| 4 |
+
"train_loss": 0.914715180794398,
|
| 5 |
+
"train_runtime": 1874.4496,
|
| 6 |
"train_samples": 319078,
|
| 7 |
+
"train_samples_per_second": 55.714,
|
| 8 |
+
"train_steps_per_second": 0.435
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -10,1168 +10,1168 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0012254901960784314,
|
| 13 |
-
"grad_norm": 1.
|
| 14 |
"learning_rate": 1.2195121951219513e-05,
|
| 15 |
"loss": 1.3541,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.006127450980392157,
|
| 20 |
-
"grad_norm": 1.
|
| 21 |
"learning_rate": 6.097560975609756e-05,
|
| 22 |
"loss": 1.355,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.012254901960784314,
|
| 27 |
-
"grad_norm": 1.
|
| 28 |
"learning_rate": 0.00012195121951219512,
|
| 29 |
"loss": 1.3083,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.01838235294117647,
|
| 34 |
-
"grad_norm": 0.
|
| 35 |
"learning_rate": 0.00018292682926829268,
|
| 36 |
"loss": 1.2226,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.024509803921568627,
|
| 41 |
-
"grad_norm": 0.
|
| 42 |
"learning_rate": 0.00024390243902439024,
|
| 43 |
-
"loss": 1.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.030637254901960783,
|
| 48 |
-
"grad_norm": 0.
|
| 49 |
"learning_rate": 0.0003048780487804878,
|
| 50 |
"loss": 1.0942,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.03676470588235294,
|
| 55 |
-
"grad_norm": 0.
|
| 56 |
"learning_rate": 0.00036585365853658537,
|
| 57 |
"loss": 1.0785,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.0428921568627451,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
"learning_rate": 0.0004268292682926829,
|
| 64 |
"loss": 1.0549,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.049019607843137254,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
"learning_rate": 0.0004878048780487805,
|
| 71 |
"loss": 1.0493,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.05514705882352941,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
"learning_rate": 0.0005487804878048781,
|
| 78 |
"loss": 1.0306,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.061274509803921566,
|
| 83 |
-
"grad_norm": 0.
|
| 84 |
"learning_rate": 0.0006097560975609756,
|
| 85 |
"loss": 1.0204,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.06740196078431372,
|
| 90 |
-
"grad_norm": 0.
|
| 91 |
"learning_rate": 0.0006707317073170732,
|
| 92 |
"loss": 1.0281,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.07352941176470588,
|
| 97 |
-
"grad_norm": 0.
|
| 98 |
"learning_rate": 0.0007317073170731707,
|
| 99 |
-
"loss": 1.
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.07965686274509803,
|
| 104 |
-
"grad_norm": 0.
|
| 105 |
"learning_rate": 0.0007926829268292683,
|
| 106 |
"loss": 1.0019,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.0857843137254902,
|
| 111 |
-
"grad_norm": 0.
|
| 112 |
"learning_rate": 0.0008536585365853659,
|
| 113 |
-
"loss": 1.
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.09191176470588236,
|
| 118 |
-
"grad_norm": 0.
|
| 119 |
"learning_rate": 0.0009146341463414635,
|
| 120 |
"loss": 1.0071,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.09803921568627451,
|
| 125 |
-
"grad_norm": 0.
|
| 126 |
"learning_rate": 0.000975609756097561,
|
| 127 |
"loss": 0.9962,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.10416666666666667,
|
| 132 |
-
"grad_norm": 0.
|
| 133 |
"learning_rate": 0.000999958782259877,
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.11029411764705882,
|
| 139 |
-
"grad_norm": 0.
|
| 140 |
"learning_rate": 0.0009997069206794246,
|
| 141 |
-
"loss": 1.
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.11642156862745098,
|
| 146 |
-
"grad_norm": 0.
|
| 147 |
"learning_rate": 0.0009992262114666653,
|
| 148 |
"loss": 0.9904,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.12254901960784313,
|
| 153 |
-
"grad_norm": 0.
|
| 154 |
"learning_rate": 0.0009985168747689707,
|
| 155 |
-
"loss": 0.
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.12867647058823528,
|
| 160 |
-
"grad_norm": 0.
|
| 161 |
"learning_rate": 0.0009975792354368017,
|
| 162 |
"loss": 0.9934,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.13480392156862744,
|
| 167 |
-
"grad_norm": 0.
|
| 168 |
"learning_rate": 0.0009964137228749407,
|
| 169 |
"loss": 0.9961,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.1409313725490196,
|
| 174 |
-
"grad_norm": 0.
|
| 175 |
"learning_rate": 0.000995020870845837,
|
| 176 |
-
"loss": 0.
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.14705882352941177,
|
| 181 |
-
"grad_norm": 0.
|
| 182 |
"learning_rate": 0.0009934013172251653,
|
| 183 |
"loss": 0.9824,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.15318627450980393,
|
| 188 |
-
"grad_norm": 0.
|
| 189 |
"learning_rate": 0.0009915558037097002,
|
| 190 |
"loss": 0.977,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.15931372549019607,
|
| 195 |
-
"grad_norm": 0.
|
| 196 |
"learning_rate": 0.0009894851754776472,
|
| 197 |
"loss": 0.9712,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.16544117647058823,
|
| 202 |
-
"grad_norm": 0.
|
| 203 |
"learning_rate": 0.0009871903808015812,
|
| 204 |
"loss": 0.9807,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.1715686274509804,
|
| 209 |
-
"grad_norm": 0.
|
| 210 |
"learning_rate": 0.0009846724706141716,
|
| 211 |
"loss": 0.977,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.17769607843137256,
|
| 216 |
-
"grad_norm": 0.
|
| 217 |
"learning_rate": 0.0009819325980268945,
|
| 218 |
"loss": 0.9743,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.18382352941176472,
|
| 223 |
-
"grad_norm": 0.
|
| 224 |
"learning_rate": 0.0009789720178019483,
|
| 225 |
-
"loss": 0.
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.18995098039215685,
|
| 230 |
-
"grad_norm": 0.
|
| 231 |
"learning_rate": 0.0009757920857776188,
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.19607843137254902,
|
| 237 |
-
"grad_norm": 0.
|
| 238 |
"learning_rate": 0.0009723942582473544,
|
| 239 |
"loss": 0.9544,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.20220588235294118,
|
| 244 |
-
"grad_norm": 0.
|
| 245 |
"learning_rate": 0.0009687800912928362,
|
| 246 |
-
"loss": 0.
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.20833333333333334,
|
| 251 |
-
"grad_norm": 0.
|
| 252 |
"learning_rate": 0.0009649512400713498,
|
| 253 |
"loss": 0.963,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.21446078431372548,
|
| 258 |
-
"grad_norm": 0.
|
| 259 |
"learning_rate": 0.0009609094580577824,
|
| 260 |
-
"loss": 0.
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.22058823529411764,
|
| 265 |
-
"grad_norm": 0.
|
| 266 |
"learning_rate": 0.0009566565962415959,
|
| 267 |
"loss": 0.9578,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 0.2267156862745098,
|
| 272 |
-
"grad_norm": 0.
|
| 273 |
"learning_rate": 0.0009521946022791401,
|
| 274 |
"loss": 0.9555,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.23284313725490197,
|
| 279 |
-
"grad_norm": 0.
|
| 280 |
"learning_rate": 0.0009475255196016972,
|
| 281 |
"loss": 0.9579,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.23897058823529413,
|
| 286 |
-
"grad_norm": 0.
|
| 287 |
"learning_rate": 0.0009426514864796647,
|
| 288 |
"loss": 0.9494,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.24509803921568626,
|
| 293 |
-
"grad_norm": 0.
|
| 294 |
"learning_rate": 0.0009375747350433044,
|
| 295 |
-
"loss": 0.
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.2512254901960784,
|
| 300 |
-
"grad_norm": 0.
|
| 301 |
"learning_rate": 0.0009322975902605082,
|
| 302 |
-
"loss": 0.
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.25735294117647056,
|
| 307 |
-
"grad_norm": 0.
|
| 308 |
"learning_rate": 0.0009268224688720474,
|
| 309 |
-
"loss": 0.
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.26348039215686275,
|
| 314 |
-
"grad_norm": 0.
|
| 315 |
"learning_rate": 0.0009211518782847931,
|
| 316 |
-
"loss": 0.
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.2696078431372549,
|
| 321 |
-
"grad_norm": 0.
|
| 322 |
"learning_rate": 0.0009152884154234145,
|
| 323 |
"loss": 0.9451,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.2757352941176471,
|
| 328 |
-
"grad_norm": 0.
|
| 329 |
"learning_rate": 0.0009092347655410818,
|
| 330 |
-
"loss": 0.
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.2818627450980392,
|
| 335 |
-
"grad_norm": 0.
|
| 336 |
"learning_rate": 0.0009029937009897176,
|
| 337 |
"loss": 0.9349,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.28799019607843135,
|
| 342 |
-
"grad_norm": 0.
|
| 343 |
"learning_rate": 0.0008965680799503608,
|
| 344 |
"loss": 0.9329,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.29411764705882354,
|
| 349 |
-
"grad_norm": 0.
|
| 350 |
"learning_rate": 0.0008899608451242233,
|
| 351 |
-
"loss": 0.
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 0.3002450980392157,
|
| 356 |
-
"grad_norm": 0.
|
| 357 |
"learning_rate": 0.0008831750223850389,
|
| 358 |
-
"loss": 0.
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.30637254901960786,
|
| 363 |
-
"grad_norm": 0.
|
| 364 |
"learning_rate": 0.0008762137193933241,
|
| 365 |
"loss": 0.9296,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 0.3125,
|
| 370 |
-
"grad_norm": 0.
|
| 371 |
"learning_rate": 0.0008690801241731818,
|
| 372 |
-
"loss": 0.
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.31862745098039214,
|
| 377 |
-
"grad_norm": 1.
|
| 378 |
"learning_rate": 0.0008617775036523015,
|
| 379 |
-
"loss": 0.
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 0.3247549019607843,
|
| 384 |
-
"grad_norm": 0.
|
| 385 |
"learning_rate": 0.0008543092021658259,
|
| 386 |
"loss": 0.9367,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.33088235294117646,
|
| 391 |
-
"grad_norm": 0.
|
| 392 |
"learning_rate": 0.0008466786399247663,
|
| 393 |
-
"loss": 0.
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
"epoch": 0.33700980392156865,
|
| 398 |
-
"grad_norm": 0.
|
| 399 |
"learning_rate": 0.0008388893114496705,
|
| 400 |
-
"loss": 0.
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.3431372549019608,
|
| 405 |
-
"grad_norm": 0.
|
| 406 |
"learning_rate": 0.0008309447839702582,
|
| 407 |
-
"loss": 0.
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 0.3492647058823529,
|
| 412 |
-
"grad_norm": 0.
|
| 413 |
"learning_rate": 0.0008228486957917607,
|
| 414 |
-
"loss": 0.
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 0.3553921568627451,
|
| 419 |
-
"grad_norm": 0.
|
| 420 |
"learning_rate": 0.0008146047546287076,
|
| 421 |
-
"loss": 0.
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
"epoch": 0.36151960784313725,
|
| 426 |
-
"grad_norm": 0.
|
| 427 |
"learning_rate": 0.0008062167359069301,
|
| 428 |
-
"loss": 0.
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 0.36764705882352944,
|
| 433 |
-
"grad_norm": 0.
|
| 434 |
"learning_rate": 0.000797688481034551,
|
| 435 |
-
"loss": 0.
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
"epoch": 0.3737745098039216,
|
| 440 |
-
"grad_norm": 0.
|
| 441 |
"learning_rate": 0.00078902389564276,
|
| 442 |
"loss": 0.9239,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 0.3799019607843137,
|
| 447 |
-
"grad_norm": 0.
|
| 448 |
"learning_rate": 0.0007802269477971771,
|
| 449 |
-
"loss": 0.
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
"epoch": 0.3860294117647059,
|
| 454 |
-
"grad_norm": 0.
|
| 455 |
"learning_rate": 0.0007713016661806211,
|
| 456 |
-
"loss": 0.
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 0.39215686274509803,
|
| 461 |
-
"grad_norm": 0.
|
| 462 |
"learning_rate": 0.0007622521382481208,
|
| 463 |
-
"loss": 0.
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
"epoch": 0.39828431372549017,
|
| 468 |
-
"grad_norm": 0.
|
| 469 |
"learning_rate": 0.0007530825083550073,
|
| 470 |
-
"loss": 0.
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 0.40441176470588236,
|
| 475 |
-
"grad_norm": 0.
|
| 476 |
"learning_rate": 0.0007437969758589507,
|
| 477 |
-
"loss": 0.
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
"epoch": 0.4105392156862745,
|
| 482 |
-
"grad_norm": 0.
|
| 483 |
"learning_rate": 0.0007343997931968067,
|
| 484 |
-
"loss": 0.
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"epoch": 0.4166666666666667,
|
| 489 |
-
"grad_norm": 0.
|
| 490 |
"learning_rate": 0.0007248952639371542,
|
| 491 |
-
"loss": 0.
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
"epoch": 0.4227941176470588,
|
| 496 |
-
"grad_norm": 0.
|
| 497 |
"learning_rate": 0.0007152877408094178,
|
| 498 |
-
"loss": 0.
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 0.42892156862745096,
|
| 503 |
-
"grad_norm": 0.
|
| 504 |
"learning_rate": 0.0007055816237104753,
|
| 505 |
-
"loss": 0.
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
"epoch": 0.43504901960784315,
|
| 510 |
-
"grad_norm": 0.
|
| 511 |
"learning_rate": 0.0006957813576896647,
|
| 512 |
-
"loss": 0.
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
"epoch": 0.4411764705882353,
|
| 517 |
-
"grad_norm": 0.
|
| 518 |
"learning_rate": 0.000685891430913113,
|
| 519 |
-
"loss": 0.
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
"epoch": 0.44730392156862747,
|
| 524 |
-
"grad_norm": 0.
|
| 525 |
"learning_rate": 0.0006759163726083191,
|
| 526 |
-
"loss": 0.
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 0.4534313725490196,
|
| 531 |
-
"grad_norm": 0.
|
| 532 |
"learning_rate": 0.0006658607509899319,
|
| 533 |
-
"loss": 0.
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 0.45955882352941174,
|
| 538 |
-
"grad_norm": 0.
|
| 539 |
"learning_rate": 0.0006557291711676738,
|
| 540 |
-
"loss": 0.
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
"epoch": 0.46568627450980393,
|
| 545 |
-
"grad_norm": 0.
|
| 546 |
"learning_rate": 0.0006455262730373672,
|
| 547 |
-
"loss": 0.
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
"epoch": 0.47181372549019607,
|
| 552 |
-
"grad_norm": 0.
|
| 553 |
"learning_rate": 0.0006352567291560318,
|
| 554 |
-
"loss": 0.
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 0.47794117647058826,
|
| 559 |
-
"grad_norm": 0.
|
| 560 |
"learning_rate": 0.0006249252426020216,
|
| 561 |
-
"loss": 0.
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 0.4840686274509804,
|
| 566 |
-
"grad_norm": 0.
|
| 567 |
"learning_rate": 0.0006145365448211866,
|
| 568 |
-
"loss": 0.
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 0.49019607843137253,
|
| 573 |
-
"grad_norm": 0.
|
| 574 |
"learning_rate": 0.0006040953934600423,
|
| 575 |
-
"loss": 0.
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 0.4963235294117647,
|
| 580 |
-
"grad_norm": 0.
|
| 581 |
"learning_rate": 0.0005936065701869403,
|
| 582 |
-
"loss": 0.
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 0.5024509803921569,
|
| 587 |
-
"grad_norm": 0.
|
| 588 |
"learning_rate": 0.0005830748785022368,
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 410
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 0.508578431372549,
|
| 594 |
-
"grad_norm": 0.
|
| 595 |
"learning_rate": 0.0005725051415384657,
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 415
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 0.5147058823529411,
|
| 601 |
-
"grad_norm":
|
| 602 |
"learning_rate": 0.0005619021998515165,
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 420
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 0.5208333333333334,
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
"learning_rate": 0.000551270909203838,
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 425
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 0.5269607843137255,
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
"learning_rate": 0.0005406161383406731,
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 430
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 0.5330882352941176,
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
"learning_rate": 0.0005299427667603515,
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 435
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 0.5392156862745098,
|
| 629 |
-
"grad_norm": 0.
|
| 630 |
"learning_rate": 0.0005192556824796568,
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 440
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 0.5453431372549019,
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
"learning_rate": 0.0005085597797952905,
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 445
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 0.5514705882352942,
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
"learning_rate": 0.0004978599570424639,
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 450
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 0.5575980392156863,
|
| 650 |
-
"grad_norm": 0.
|
| 651 |
"learning_rate": 0.0004871611143516367,
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 455
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 0.5637254901960784,
|
| 657 |
-
"grad_norm": 0.
|
| 658 |
"learning_rate": 0.0004764681514044362,
|
| 659 |
-
"loss": 0.
|
| 660 |
"step": 460
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 0.5698529411764706,
|
| 664 |
-
"grad_norm": 0.
|
| 665 |
"learning_rate": 0.0004657859651897806,
|
| 666 |
-
"loss": 0.
|
| 667 |
"step": 465
|
| 668 |
},
|
| 669 |
{
|
| 670 |
"epoch": 0.5759803921568627,
|
| 671 |
-
"grad_norm": 0.
|
| 672 |
"learning_rate": 0.00045511944776123513,
|
| 673 |
-
"loss": 0.
|
| 674 |
"step": 470
|
| 675 |
},
|
| 676 |
{
|
| 677 |
"epoch": 0.5821078431372549,
|
| 678 |
-
"grad_norm": 0.
|
| 679 |
"learning_rate": 0.00044447348399663056,
|
| 680 |
-
"loss": 0.
|
| 681 |
"step": 475
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 0.5882352941176471,
|
| 685 |
-
"grad_norm": 0.
|
| 686 |
"learning_rate": 0.0004338529493609647,
|
| 687 |
-
"loss": 0.
|
| 688 |
"step": 480
|
| 689 |
},
|
| 690 |
{
|
| 691 |
"epoch": 0.5943627450980392,
|
| 692 |
-
"grad_norm": 0.
|
| 693 |
"learning_rate": 0.00042326270767361815,
|
| 694 |
-
"loss": 0.
|
| 695 |
"step": 485
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 0.6004901960784313,
|
| 699 |
-
"grad_norm": 0.
|
| 700 |
"learning_rate": 0.00041270760888089997,
|
| 701 |
-
"loss": 0.
|
| 702 |
"step": 490
|
| 703 |
},
|
| 704 |
{
|
| 705 |
"epoch": 0.6066176470588235,
|
| 706 |
-
"grad_norm": 0.
|
| 707 |
"learning_rate": 0.00040219248683494925,
|
| 708 |
-
"loss": 0.
|
| 709 |
"step": 495
|
| 710 |
},
|
| 711 |
{
|
| 712 |
"epoch": 0.6127450980392157,
|
| 713 |
-
"grad_norm": 0.
|
| 714 |
"learning_rate": 0.0003917221570800065,
|
| 715 |
-
"loss": 0.
|
| 716 |
"step": 500
|
| 717 |
},
|
| 718 |
{
|
| 719 |
"epoch": 0.6188725490196079,
|
| 720 |
-
"grad_norm": 0.
|
| 721 |
"learning_rate": 0.000381301414647068,
|
| 722 |
-
"loss": 0.
|
| 723 |
"step": 505
|
| 724 |
},
|
| 725 |
{
|
| 726 |
"epoch": 0.625,
|
| 727 |
-
"grad_norm": 0.
|
| 728 |
"learning_rate": 0.0003709350318579371,
|
| 729 |
-
"loss": 0.
|
| 730 |
"step": 510
|
| 731 |
},
|
| 732 |
{
|
| 733 |
"epoch": 0.6311274509803921,
|
| 734 |
-
"grad_norm": 0.
|
| 735 |
"learning_rate": 0.0003606277561396726,
|
| 736 |
-
"loss": 0.
|
| 737 |
"step": 515
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 0.6372549019607843,
|
| 741 |
-
"grad_norm": 0.
|
| 742 |
"learning_rate": 0.00035038430785044053,
|
| 743 |
-
"loss": 0.
|
| 744 |
"step": 520
|
| 745 |
},
|
| 746 |
{
|
| 747 |
"epoch": 0.6433823529411765,
|
| 748 |
-
"grad_norm": 0.
|
| 749 |
"learning_rate": 0.00034020937811776156,
|
| 750 |
-
"loss": 0.
|
| 751 |
"step": 525
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 0.6495098039215687,
|
| 755 |
-
"grad_norm": 0.
|
| 756 |
"learning_rate": 0.00033010762669014347,
|
| 757 |
-
"loss": 0.
|
| 758 |
"step": 530
|
| 759 |
},
|
| 760 |
{
|
| 761 |
"epoch": 0.6556372549019608,
|
| 762 |
-
"grad_norm": 0.
|
| 763 |
"learning_rate": 0.00032008367980308734,
|
| 764 |
-
"loss": 0.
|
| 765 |
"step": 535
|
| 766 |
},
|
| 767 |
{
|
| 768 |
"epoch": 0.6617647058823529,
|
| 769 |
-
"grad_norm": 0.
|
| 770 |
"learning_rate": 0.0003101421280604379,
|
| 771 |
-
"loss": 0.
|
| 772 |
"step": 540
|
| 773 |
},
|
| 774 |
{
|
| 775 |
"epoch": 0.6678921568627451,
|
| 776 |
-
"grad_norm": 0.
|
| 777 |
"learning_rate": 0.00030028752433205476,
|
| 778 |
-
"loss": 0.
|
| 779 |
"step": 545
|
| 780 |
},
|
| 781 |
{
|
| 782 |
"epoch": 0.6740196078431373,
|
| 783 |
-
"grad_norm": 0.
|
| 784 |
"learning_rate": 0.00029052438166876307,
|
| 785 |
-
"loss": 0.
|
| 786 |
"step": 550
|
| 787 |
},
|
| 788 |
{
|
| 789 |
"epoch": 0.6801470588235294,
|
| 790 |
-
"grad_norm": 0.
|
| 791 |
"learning_rate": 0.0002808571712355389,
|
| 792 |
-
"loss": 0.
|
| 793 |
"step": 555
|
| 794 |
},
|
| 795 |
{
|
| 796 |
"epoch": 0.6862745098039216,
|
| 797 |
-
"grad_norm": 0.
|
| 798 |
"learning_rate": 0.00027129032026388045,
|
| 799 |
-
"loss": 0.
|
| 800 |
"step": 560
|
| 801 |
},
|
| 802 |
{
|
| 803 |
"epoch": 0.6924019607843137,
|
| 804 |
-
"grad_norm": 0.
|
| 805 |
"learning_rate": 0.00026182821002429345,
|
| 806 |
-
"loss": 0.
|
| 807 |
"step": 565
|
| 808 |
},
|
| 809 |
{
|
| 810 |
"epoch": 0.6985294117647058,
|
| 811 |
-
"grad_norm": 0.
|
| 812 |
"learning_rate": 0.00025247517381983136,
|
| 813 |
-
"loss": 0.
|
| 814 |
"step": 570
|
| 815 |
},
|
| 816 |
{
|
| 817 |
"epoch": 0.7046568627450981,
|
| 818 |
-
"grad_norm": 0.
|
| 819 |
"learning_rate": 0.00024323549500159802,
|
| 820 |
-
"loss": 0.
|
| 821 |
"step": 575
|
| 822 |
},
|
| 823 |
{
|
| 824 |
"epoch": 0.7107843137254902,
|
| 825 |
-
"grad_norm": 0.
|
| 826 |
"learning_rate": 0.0002341134050071283,
|
| 827 |
-
"loss": 0.
|
| 828 |
"step": 580
|
| 829 |
},
|
| 830 |
{
|
| 831 |
"epoch": 0.7169117647058824,
|
| 832 |
-
"grad_norm": 0.
|
| 833 |
"learning_rate": 0.00022511308142254488,
|
| 834 |
-
"loss": 0.
|
| 835 |
"step": 585
|
| 836 |
},
|
| 837 |
{
|
| 838 |
"epoch": 0.7230392156862745,
|
| 839 |
-
"grad_norm": 0.
|
| 840 |
"learning_rate": 0.000216238646069373,
|
| 841 |
-
"loss": 0.
|
| 842 |
"step": 590
|
| 843 |
},
|
| 844 |
{
|
| 845 |
"epoch": 0.7291666666666666,
|
| 846 |
-
"grad_norm": 0.
|
| 847 |
"learning_rate": 0.00020749416311689845,
|
| 848 |
-
"loss": 0.
|
| 849 |
"step": 595
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 0.7352941176470589,
|
| 853 |
-
"grad_norm": 0.
|
| 854 |
"learning_rate": 0.00019888363722092372,
|
| 855 |
-
"loss": 0.
|
| 856 |
"step": 600
|
| 857 |
},
|
| 858 |
{
|
| 859 |
"epoch": 0.741421568627451,
|
| 860 |
-
"grad_norm": 0.
|
| 861 |
"learning_rate": 0.00019041101168978093,
|
| 862 |
-
"loss": 0.
|
| 863 |
"step": 605
|
| 864 |
},
|
| 865 |
{
|
| 866 |
"epoch": 0.7475490196078431,
|
| 867 |
-
"grad_norm": 0.
|
| 868 |
"learning_rate": 0.00018208016667844152,
|
| 869 |
-
"loss": 0.
|
| 870 |
"step": 610
|
| 871 |
},
|
| 872 |
{
|
| 873 |
"epoch": 0.7536764705882353,
|
| 874 |
-
"grad_norm": 0.
|
| 875 |
"learning_rate": 0.00017389491741154372,
|
| 876 |
-
"loss": 0.
|
| 877 |
"step": 615
|
| 878 |
},
|
| 879 |
{
|
| 880 |
"epoch": 0.7598039215686274,
|
| 881 |
-
"grad_norm": 0.
|
| 882 |
"learning_rate": 0.00016585901243616042,
|
| 883 |
-
"loss": 0.
|
| 884 |
"step": 620
|
| 885 |
},
|
| 886 |
{
|
| 887 |
"epoch": 0.7659313725490197,
|
| 888 |
-
"grad_norm": 0.
|
| 889 |
"learning_rate": 0.0001579761319050991,
|
| 890 |
-
"loss": 0.
|
| 891 |
"step": 625
|
| 892 |
},
|
| 893 |
{
|
| 894 |
"epoch": 0.7720588235294118,
|
| 895 |
-
"grad_norm": 0.
|
| 896 |
"learning_rate": 0.00015024988589152537,
|
| 897 |
-
"loss": 0.
|
| 898 |
"step": 630
|
| 899 |
},
|
| 900 |
{
|
| 901 |
"epoch": 0.7781862745098039,
|
| 902 |
-
"grad_norm": 0.
|
| 903 |
"learning_rate": 0.0001426838127356823,
|
| 904 |
-
"loss": 0.
|
| 905 |
"step": 635
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 0.7843137254901961,
|
| 909 |
-
"grad_norm": 0.
|
| 910 |
"learning_rate": 0.0001352813774244565,
|
| 911 |
-
"loss": 0.
|
| 912 |
"step": 640
|
| 913 |
},
|
| 914 |
{
|
| 915 |
"epoch": 0.7904411764705882,
|
| 916 |
-
"grad_norm": 0.
|
| 917 |
"learning_rate": 0.00012804597000454215,
|
| 918 |
-
"loss": 0.
|
| 919 |
"step": 645
|
| 920 |
},
|
| 921 |
{
|
| 922 |
"epoch": 0.7965686274509803,
|
| 923 |
-
"grad_norm": 0.
|
| 924 |
"learning_rate": 0.00012098090402992085,
|
| 925 |
-
"loss": 0.
|
| 926 |
"step": 650
|
| 927 |
},
|
| 928 |
{
|
| 929 |
"epoch": 0.8026960784313726,
|
| 930 |
-
"grad_norm": 0.
|
| 931 |
"learning_rate": 0.00011408941504437532,
|
| 932 |
-
"loss": 0.
|
| 933 |
"step": 655
|
| 934 |
},
|
| 935 |
{
|
| 936 |
"epoch": 0.8088235294117647,
|
| 937 |
-
"grad_norm": 0.
|
| 938 |
"learning_rate": 0.00010737465909972776,
|
| 939 |
-
"loss": 0.
|
| 940 |
"step": 660
|
| 941 |
},
|
| 942 |
{
|
| 943 |
"epoch": 0.8149509803921569,
|
| 944 |
-
"grad_norm": 0.
|
| 945 |
"learning_rate": 0.00010083971131048159,
|
| 946 |
-
"loss": 0.
|
| 947 |
"step": 665
|
| 948 |
},
|
| 949 |
{
|
| 950 |
"epoch": 0.821078431372549,
|
| 951 |
-
"grad_norm": 0.
|
| 952 |
"learning_rate": 9.448756444553224e-05,
|
| 953 |
-
"loss": 0.
|
| 954 |
"step": 670
|
| 955 |
},
|
| 956 |
{
|
| 957 |
"epoch": 0.8272058823529411,
|
| 958 |
-
"grad_norm": 0.
|
| 959 |
"learning_rate": 8.832112755758598e-05,
|
| 960 |
-
"loss": 0.
|
| 961 |
"step": 675
|
| 962 |
},
|
| 963 |
{
|
| 964 |
"epoch": 0.8333333333333334,
|
| 965 |
-
"grad_norm": 0.
|
| 966 |
"learning_rate": 8.234322465092047e-05,
|
| 967 |
-
"loss": 0.
|
| 968 |
"step": 680
|
| 969 |
},
|
| 970 |
{
|
| 971 |
"epoch": 0.8394607843137255,
|
| 972 |
-
"grad_norm": 0.
|
| 973 |
"learning_rate": 7.655659338809329e-05,
|
| 974 |
-
"loss": 0.
|
| 975 |
"step": 685
|
| 976 |
},
|
| 977 |
{
|
| 978 |
"epoch": 0.8455882352941176,
|
| 979 |
-
"grad_norm": 0.
|
| 980 |
"learning_rate": 7.096388383619079e-05,
|
| 981 |
-
"loss": 0.
|
| 982 |
"step": 690
|
| 983 |
},
|
| 984 |
{
|
| 985 |
"epoch": 0.8517156862745098,
|
| 986 |
-
"grad_norm": 0.
|
| 987 |
"learning_rate": 6.556765725319525e-05,
|
| 988 |
-
"loss": 0.
|
| 989 |
"step": 695
|
| 990 |
},
|
| 991 |
{
|
| 992 |
"epoch": 0.8578431372549019,
|
| 993 |
-
"grad_norm": 0.
|
| 994 |
"learning_rate": 6.037038491501978e-05,
|
| 995 |
-
"loss": 0.
|
| 996 |
"step": 700
|
| 997 |
},
|
| 998 |
{
|
| 999 |
"epoch": 0.8639705882352942,
|
| 1000 |
-
"grad_norm": 0.
|
| 1001 |
"learning_rate": 5.53744469837551e-05,
|
| 1002 |
-
"loss": 0.
|
| 1003 |
"step": 705
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
"epoch": 0.8700980392156863,
|
| 1007 |
-
"grad_norm": 0.
|
| 1008 |
"learning_rate": 5.058213141764151e-05,
|
| 1009 |
-
"loss": 0.
|
| 1010 |
"step": 710
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
"epoch": 0.8762254901960784,
|
| 1014 |
-
"grad_norm": 0.
|
| 1015 |
"learning_rate": 4.599563292326592e-05,
|
| 1016 |
-
"loss": 0.
|
| 1017 |
"step": 715
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
"epoch": 0.8823529411764706,
|
| 1021 |
-
"grad_norm": 0.
|
| 1022 |
"learning_rate": 4.161705195046761e-05,
|
| 1023 |
-
"loss": 0.
|
| 1024 |
"step": 720
|
| 1025 |
},
|
| 1026 |
{
|
| 1027 |
"epoch": 0.8884803921568627,
|
| 1028 |
-
"grad_norm": 0.
|
| 1029 |
"learning_rate": 3.744839373040682e-05,
|
| 1030 |
-
"loss": 0.
|
| 1031 |
"step": 725
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
"epoch": 0.8946078431372549,
|
| 1035 |
-
"grad_norm": 0.
|
| 1036 |
"learning_rate": 3.349156735724274e-05,
|
| 1037 |
-
"loss": 0.
|
| 1038 |
"step": 730
|
| 1039 |
},
|
| 1040 |
{
|
| 1041 |
"epoch": 0.9007352941176471,
|
| 1042 |
-
"grad_norm": 0.
|
| 1043 |
"learning_rate": 2.9748384913837522e-05,
|
| 1044 |
-
"loss": 0.
|
| 1045 |
"step": 735
|
| 1046 |
},
|
| 1047 |
{
|
| 1048 |
"epoch": 0.9068627450980392,
|
| 1049 |
-
"grad_norm": 0.
|
| 1050 |
"learning_rate": 2.622056064188738e-05,
|
| 1051 |
-
"loss": 0.
|
| 1052 |
"step": 740
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
"epoch": 0.9129901960784313,
|
| 1056 |
-
"grad_norm": 0.
|
| 1057 |
"learning_rate": 2.2909710156863274e-05,
|
| 1058 |
-
"loss": 0.
|
| 1059 |
"step": 745
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
"epoch": 0.9191176470588235,
|
| 1063 |
-
"grad_norm": 0.
|
| 1064 |
"learning_rate": 1.981734970811644e-05,
|
| 1065 |
-
"loss": 0.
|
| 1066 |
"step": 750
|
| 1067 |
},
|
| 1068 |
{
|
| 1069 |
"epoch": 0.9252450980392157,
|
| 1070 |
-
"grad_norm": 0.
|
| 1071 |
"learning_rate": 1.6944895484492072e-05,
|
| 1072 |
-
"loss": 0.
|
| 1073 |
"step": 755
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
"epoch": 0.9313725490196079,
|
| 1077 |
-
"grad_norm": 0.
|
| 1078 |
"learning_rate": 1.429366296576623e-05,
|
| 1079 |
-
"loss": 0.
|
| 1080 |
"step": 760
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
"epoch": 0.9375,
|
| 1084 |
-
"grad_norm": 0.
|
| 1085 |
"learning_rate": 1.1864866320203115e-05,
|
| 1086 |
-
"loss": 0.
|
| 1087 |
"step": 765
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
"epoch": 0.9436274509803921,
|
| 1091 |
-
"grad_norm": 0.
|
| 1092 |
"learning_rate": 9.659617848510882e-06,
|
| 1093 |
-
"loss": 0.
|
| 1094 |
"step": 770
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
"epoch": 0.9497549019607843,
|
| 1098 |
-
"grad_norm": 0.
|
| 1099 |
"learning_rate": 7.678927474447817e-06,
|
| 1100 |
-
"loss": 0.
|
| 1101 |
"step": 775
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
"epoch": 0.9558823529411765,
|
| 1105 |
-
"grad_norm": 0.
|
| 1106 |
"learning_rate": 5.923702282314092e-06,
|
| 1107 |
-
"loss": 0.
|
| 1108 |
"step": 780
|
| 1109 |
},
|
| 1110 |
{
|
| 1111 |
"epoch": 0.9620098039215687,
|
| 1112 |
-
"grad_norm": 0.
|
| 1113 |
"learning_rate": 4.394746101540115e-06,
|
| 1114 |
-
"loss": 0.
|
| 1115 |
"step": 785
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
"epoch": 0.9681372549019608,
|
| 1119 |
-
"grad_norm": 0.
|
| 1120 |
"learning_rate": 3.092759138561607e-06,
|
| 1121 |
-
"loss": 0.
|
| 1122 |
"step": 790
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
"epoch": 0.9742647058823529,
|
| 1126 |
-
"grad_norm": 0.
|
| 1127 |
"learning_rate": 2.018337656150726e-06,
|
| 1128 |
-
"loss": 0.
|
| 1129 |
"step": 795
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 0.9803921568627451,
|
| 1133 |
-
"grad_norm": 0.
|
| 1134 |
"learning_rate": 1.1719737003492159e-06,
|
| 1135 |
-
"loss": 0.
|
| 1136 |
"step": 800
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
"epoch": 0.9865196078431373,
|
| 1140 |
-
"grad_norm": 0.
|
| 1141 |
"learning_rate": 5.540548751292173e-07,
|
| 1142 |
-
"loss": 0.
|
| 1143 |
"step": 805
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 0.9926470588235294,
|
| 1147 |
-
"grad_norm": 0.
|
| 1148 |
"learning_rate": 1.6486416488459277e-07,
|
| 1149 |
-
"loss": 0.
|
| 1150 |
"step": 810
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
"epoch": 0.9987745098039216,
|
| 1154 |
-
"grad_norm": 0.
|
| 1155 |
"learning_rate": 4.579804834703438e-09,
|
| 1156 |
-
"loss": 0.
|
| 1157 |
"step": 815
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 1.0,
|
| 1161 |
-
"eval_loss": 1.
|
| 1162 |
-
"eval_runtime":
|
| 1163 |
-
"eval_samples_per_second":
|
| 1164 |
-
"eval_steps_per_second": 5.
|
| 1165 |
"step": 816
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
"epoch": 1.0,
|
| 1169 |
"step": 816,
|
| 1170 |
"total_flos": 80063181619200.0,
|
| 1171 |
-
"train_loss": 0.
|
| 1172 |
-
"train_runtime":
|
| 1173 |
-
"train_samples_per_second":
|
| 1174 |
-
"train_steps_per_second": 0.
|
| 1175 |
}
|
| 1176 |
],
|
| 1177 |
"logging_steps": 5,
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0012254901960784314,
|
| 13 |
+
"grad_norm": 1.603988652495189,
|
| 14 |
"learning_rate": 1.2195121951219513e-05,
|
| 15 |
"loss": 1.3541,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.006127450980392157,
|
| 20 |
+
"grad_norm": 1.4587743423854005,
|
| 21 |
"learning_rate": 6.097560975609756e-05,
|
| 22 |
"loss": 1.355,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.012254901960784314,
|
| 27 |
+
"grad_norm": 1.538685925430638,
|
| 28 |
"learning_rate": 0.00012195121951219512,
|
| 29 |
"loss": 1.3083,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.01838235294117647,
|
| 34 |
+
"grad_norm": 0.5820643527511544,
|
| 35 |
"learning_rate": 0.00018292682926829268,
|
| 36 |
"loss": 1.2226,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.024509803921568627,
|
| 41 |
+
"grad_norm": 0.37609725156071605,
|
| 42 |
"learning_rate": 0.00024390243902439024,
|
| 43 |
+
"loss": 1.1399,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.030637254901960783,
|
| 48 |
+
"grad_norm": 0.2477379120591674,
|
| 49 |
"learning_rate": 0.0003048780487804878,
|
| 50 |
"loss": 1.0942,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.03676470588235294,
|
| 55 |
+
"grad_norm": 0.18934222364015724,
|
| 56 |
"learning_rate": 0.00036585365853658537,
|
| 57 |
"loss": 1.0785,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.0428921568627451,
|
| 62 |
+
"grad_norm": 0.14574615338739755,
|
| 63 |
"learning_rate": 0.0004268292682926829,
|
| 64 |
"loss": 1.0549,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.049019607843137254,
|
| 69 |
+
"grad_norm": 0.12815166481708085,
|
| 70 |
"learning_rate": 0.0004878048780487805,
|
| 71 |
"loss": 1.0493,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.05514705882352941,
|
| 76 |
+
"grad_norm": 0.15010519509218812,
|
| 77 |
"learning_rate": 0.0005487804878048781,
|
| 78 |
"loss": 1.0306,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.061274509803921566,
|
| 83 |
+
"grad_norm": 0.13010925959434533,
|
| 84 |
"learning_rate": 0.0006097560975609756,
|
| 85 |
"loss": 1.0204,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.06740196078431372,
|
| 90 |
+
"grad_norm": 0.11891525726508857,
|
| 91 |
"learning_rate": 0.0006707317073170732,
|
| 92 |
"loss": 1.0281,
|
| 93 |
"step": 55
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.07352941176470588,
|
| 97 |
+
"grad_norm": 0.12117005404429922,
|
| 98 |
"learning_rate": 0.0007317073170731707,
|
| 99 |
+
"loss": 1.0187,
|
| 100 |
"step": 60
|
| 101 |
},
|
| 102 |
{
|
| 103 |
"epoch": 0.07965686274509803,
|
| 104 |
+
"grad_norm": 0.11923924460163615,
|
| 105 |
"learning_rate": 0.0007926829268292683,
|
| 106 |
"loss": 1.0019,
|
| 107 |
"step": 65
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 0.0857843137254902,
|
| 111 |
+
"grad_norm": 0.13523477315023974,
|
| 112 |
"learning_rate": 0.0008536585365853659,
|
| 113 |
+
"loss": 1.0044,
|
| 114 |
"step": 70
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 0.09191176470588236,
|
| 118 |
+
"grad_norm": 0.11307823129618054,
|
| 119 |
"learning_rate": 0.0009146341463414635,
|
| 120 |
"loss": 1.0071,
|
| 121 |
"step": 75
|
| 122 |
},
|
| 123 |
{
|
| 124 |
"epoch": 0.09803921568627451,
|
| 125 |
+
"grad_norm": 0.1206648748330027,
|
| 126 |
"learning_rate": 0.000975609756097561,
|
| 127 |
"loss": 0.9962,
|
| 128 |
"step": 80
|
| 129 |
},
|
| 130 |
{
|
| 131 |
"epoch": 0.10416666666666667,
|
| 132 |
+
"grad_norm": 0.15935283992889565,
|
| 133 |
"learning_rate": 0.000999958782259877,
|
| 134 |
+
"loss": 0.9998,
|
| 135 |
"step": 85
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 0.11029411764705882,
|
| 139 |
+
"grad_norm": 0.13630538643217202,
|
| 140 |
"learning_rate": 0.0009997069206794246,
|
| 141 |
+
"loss": 1.0101,
|
| 142 |
"step": 90
|
| 143 |
},
|
| 144 |
{
|
| 145 |
"epoch": 0.11642156862745098,
|
| 146 |
+
"grad_norm": 0.13281831595913912,
|
| 147 |
"learning_rate": 0.0009992262114666653,
|
| 148 |
"loss": 0.9904,
|
| 149 |
"step": 95
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 0.12254901960784313,
|
| 153 |
+
"grad_norm": 0.13570583268291556,
|
| 154 |
"learning_rate": 0.0009985168747689707,
|
| 155 |
+
"loss": 0.986,
|
| 156 |
"step": 100
|
| 157 |
},
|
| 158 |
{
|
| 159 |
"epoch": 0.12867647058823528,
|
| 160 |
+
"grad_norm": 0.14302939089927838,
|
| 161 |
"learning_rate": 0.0009975792354368017,
|
| 162 |
"loss": 0.9934,
|
| 163 |
"step": 105
|
| 164 |
},
|
| 165 |
{
|
| 166 |
"epoch": 0.13480392156862744,
|
| 167 |
+
"grad_norm": 0.14349254312543258,
|
| 168 |
"learning_rate": 0.0009964137228749407,
|
| 169 |
"loss": 0.9961,
|
| 170 |
"step": 110
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 0.1409313725490196,
|
| 174 |
+
"grad_norm": 0.13092834458351188,
|
| 175 |
"learning_rate": 0.000995020870845837,
|
| 176 |
+
"loss": 0.9949,
|
| 177 |
"step": 115
|
| 178 |
},
|
| 179 |
{
|
| 180 |
"epoch": 0.14705882352941177,
|
| 181 |
+
"grad_norm": 0.1251118799739706,
|
| 182 |
"learning_rate": 0.0009934013172251653,
|
| 183 |
"loss": 0.9824,
|
| 184 |
"step": 120
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"epoch": 0.15318627450980393,
|
| 188 |
+
"grad_norm": 0.1323892618605143,
|
| 189 |
"learning_rate": 0.0009915558037097002,
|
| 190 |
"loss": 0.977,
|
| 191 |
"step": 125
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.15931372549019607,
|
| 195 |
+
"grad_norm": 0.13068519381648078,
|
| 196 |
"learning_rate": 0.0009894851754776472,
|
| 197 |
"loss": 0.9712,
|
| 198 |
"step": 130
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 0.16544117647058823,
|
| 202 |
+
"grad_norm": 0.1310198992984819,
|
| 203 |
"learning_rate": 0.0009871903808015812,
|
| 204 |
"loss": 0.9807,
|
| 205 |
"step": 135
|
| 206 |
},
|
| 207 |
{
|
| 208 |
"epoch": 0.1715686274509804,
|
| 209 |
+
"grad_norm": 0.10811315309592277,
|
| 210 |
"learning_rate": 0.0009846724706141716,
|
| 211 |
"loss": 0.977,
|
| 212 |
"step": 140
|
| 213 |
},
|
| 214 |
{
|
| 215 |
"epoch": 0.17769607843137256,
|
| 216 |
+
"grad_norm": 0.11603574555194691,
|
| 217 |
"learning_rate": 0.0009819325980268945,
|
| 218 |
"loss": 0.9743,
|
| 219 |
"step": 145
|
| 220 |
},
|
| 221 |
{
|
| 222 |
"epoch": 0.18382352941176472,
|
| 223 |
+
"grad_norm": 0.11664960595520962,
|
| 224 |
"learning_rate": 0.0009789720178019483,
|
| 225 |
+
"loss": 0.9742,
|
| 226 |
"step": 150
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"epoch": 0.18995098039215685,
|
| 230 |
+
"grad_norm": 0.11920852297334043,
|
| 231 |
"learning_rate": 0.0009757920857776188,
|
| 232 |
+
"loss": 0.9635,
|
| 233 |
"step": 155
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 0.19607843137254902,
|
| 237 |
+
"grad_norm": 0.13745202686899544,
|
| 238 |
"learning_rate": 0.0009723942582473544,
|
| 239 |
"loss": 0.9544,
|
| 240 |
"step": 160
|
| 241 |
},
|
| 242 |
{
|
| 243 |
"epoch": 0.20220588235294118,
|
| 244 |
+
"grad_norm": 0.15444220703514816,
|
| 245 |
"learning_rate": 0.0009687800912928362,
|
| 246 |
+
"loss": 0.9697,
|
| 247 |
"step": 165
|
| 248 |
},
|
| 249 |
{
|
| 250 |
"epoch": 0.20833333333333334,
|
| 251 |
+
"grad_norm": 0.11511217870343073,
|
| 252 |
"learning_rate": 0.0009649512400713498,
|
| 253 |
"loss": 0.963,
|
| 254 |
"step": 170
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 0.21446078431372548,
|
| 258 |
+
"grad_norm": 0.163845433820889,
|
| 259 |
"learning_rate": 0.0009609094580577824,
|
| 260 |
+
"loss": 0.9601,
|
| 261 |
"step": 175
|
| 262 |
},
|
| 263 |
{
|
| 264 |
"epoch": 0.22058823529411764,
|
| 265 |
+
"grad_norm": 0.12370218334013189,
|
| 266 |
"learning_rate": 0.0009566565962415959,
|
| 267 |
"loss": 0.9578,
|
| 268 |
"step": 180
|
| 269 |
},
|
| 270 |
{
|
| 271 |
"epoch": 0.2267156862745098,
|
| 272 |
+
"grad_norm": 0.138057520129555,
|
| 273 |
"learning_rate": 0.0009521946022791401,
|
| 274 |
"loss": 0.9555,
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.23284313725490197,
|
| 279 |
+
"grad_norm": 0.161151229045878,
|
| 280 |
"learning_rate": 0.0009475255196016972,
|
| 281 |
"loss": 0.9579,
|
| 282 |
"step": 190
|
| 283 |
},
|
| 284 |
{
|
| 285 |
"epoch": 0.23897058823529413,
|
| 286 |
+
"grad_norm": 0.13900328482304902,
|
| 287 |
"learning_rate": 0.0009426514864796647,
|
| 288 |
"loss": 0.9494,
|
| 289 |
"step": 195
|
| 290 |
},
|
| 291 |
{
|
| 292 |
"epoch": 0.24509803921568626,
|
| 293 |
+
"grad_norm": 0.14057545846182565,
|
| 294 |
"learning_rate": 0.0009375747350433044,
|
| 295 |
+
"loss": 0.9478,
|
| 296 |
"step": 200
|
| 297 |
},
|
| 298 |
{
|
| 299 |
"epoch": 0.2512254901960784,
|
| 300 |
+
"grad_norm": 0.15616371521107208,
|
| 301 |
"learning_rate": 0.0009322975902605082,
|
| 302 |
+
"loss": 0.9654,
|
| 303 |
"step": 205
|
| 304 |
},
|
| 305 |
{
|
| 306 |
"epoch": 0.25735294117647056,
|
| 307 |
+
"grad_norm": 0.11827026404580182,
|
| 308 |
"learning_rate": 0.0009268224688720474,
|
| 309 |
+
"loss": 0.9445,
|
| 310 |
"step": 210
|
| 311 |
},
|
| 312 |
{
|
| 313 |
"epoch": 0.26348039215686275,
|
| 314 |
+
"grad_norm": 0.11103911913637518,
|
| 315 |
"learning_rate": 0.0009211518782847931,
|
| 316 |
+
"loss": 0.9424,
|
| 317 |
"step": 215
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 0.2696078431372549,
|
| 321 |
+
"grad_norm": 0.11604427070566481,
|
| 322 |
"learning_rate": 0.0009152884154234145,
|
| 323 |
"loss": 0.9451,
|
| 324 |
"step": 220
|
| 325 |
},
|
| 326 |
{
|
| 327 |
"epoch": 0.2757352941176471,
|
| 328 |
+
"grad_norm": 0.1099562215414043,
|
| 329 |
"learning_rate": 0.0009092347655410818,
|
| 330 |
+
"loss": 0.9402,
|
| 331 |
"step": 225
|
| 332 |
},
|
| 333 |
{
|
| 334 |
"epoch": 0.2818627450980392,
|
| 335 |
+
"grad_norm": 0.14837520991789005,
|
| 336 |
"learning_rate": 0.0009029937009897176,
|
| 337 |
"loss": 0.9349,
|
| 338 |
"step": 230
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 0.28799019607843135,
|
| 342 |
+
"grad_norm": 0.10928552841333679,
|
| 343 |
"learning_rate": 0.0008965680799503608,
|
| 344 |
"loss": 0.9329,
|
| 345 |
"step": 235
|
| 346 |
},
|
| 347 |
{
|
| 348 |
"epoch": 0.29411764705882354,
|
| 349 |
+
"grad_norm": 0.11407153214331639,
|
| 350 |
"learning_rate": 0.0008899608451242233,
|
| 351 |
+
"loss": 0.9379,
|
| 352 |
"step": 240
|
| 353 |
},
|
| 354 |
{
|
| 355 |
"epoch": 0.3002450980392157,
|
| 356 |
+
"grad_norm": 0.11231916470556697,
|
| 357 |
"learning_rate": 0.0008831750223850389,
|
| 358 |
+
"loss": 0.9229,
|
| 359 |
"step": 245
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 0.30637254901960786,
|
| 363 |
+
"grad_norm": 0.11185052745256109,
|
| 364 |
"learning_rate": 0.0008762137193933241,
|
| 365 |
"loss": 0.9296,
|
| 366 |
"step": 250
|
| 367 |
},
|
| 368 |
{
|
| 369 |
"epoch": 0.3125,
|
| 370 |
+
"grad_norm": 0.11855657350077958,
|
| 371 |
"learning_rate": 0.0008690801241731818,
|
| 372 |
+
"loss": 0.9207,
|
| 373 |
"step": 255
|
| 374 |
},
|
| 375 |
{
|
| 376 |
"epoch": 0.31862745098039214,
|
| 377 |
+
"grad_norm": 1.8537407128611012,
|
| 378 |
"learning_rate": 0.0008617775036523015,
|
| 379 |
+
"loss": 0.9387,
|
| 380 |
"step": 260
|
| 381 |
},
|
| 382 |
{
|
| 383 |
"epoch": 0.3247549019607843,
|
| 384 |
+
"grad_norm": 0.11676606107692747,
|
| 385 |
"learning_rate": 0.0008543092021658259,
|
| 386 |
"loss": 0.9367,
|
| 387 |
"step": 265
|
| 388 |
},
|
| 389 |
{
|
| 390 |
"epoch": 0.33088235294117646,
|
| 391 |
+
"grad_norm": 0.10492580984162286,
|
| 392 |
"learning_rate": 0.0008466786399247663,
|
| 393 |
+
"loss": 0.928,
|
| 394 |
"step": 270
|
| 395 |
},
|
| 396 |
{
|
| 397 |
"epoch": 0.33700980392156865,
|
| 398 |
+
"grad_norm": 0.10281631398110604,
|
| 399 |
"learning_rate": 0.0008388893114496705,
|
| 400 |
+
"loss": 0.935,
|
| 401 |
"step": 275
|
| 402 |
},
|
| 403 |
{
|
| 404 |
"epoch": 0.3431372549019608,
|
| 405 |
+
"grad_norm": 0.11217225067437296,
|
| 406 |
"learning_rate": 0.0008309447839702582,
|
| 407 |
+
"loss": 0.9298,
|
| 408 |
"step": 280
|
| 409 |
},
|
| 410 |
{
|
| 411 |
"epoch": 0.3492647058823529,
|
| 412 |
+
"grad_norm": 0.11327220268180357,
|
| 413 |
"learning_rate": 0.0008228486957917607,
|
| 414 |
+
"loss": 0.9219,
|
| 415 |
"step": 285
|
| 416 |
},
|
| 417 |
{
|
| 418 |
"epoch": 0.3553921568627451,
|
| 419 |
+
"grad_norm": 0.11554152008646122,
|
| 420 |
"learning_rate": 0.0008146047546287076,
|
| 421 |
+
"loss": 0.934,
|
| 422 |
"step": 290
|
| 423 |
},
|
| 424 |
{
|
| 425 |
"epoch": 0.36151960784313725,
|
| 426 |
+
"grad_norm": 0.13610027478132888,
|
| 427 |
"learning_rate": 0.0008062167359069301,
|
| 428 |
+
"loss": 0.9276,
|
| 429 |
"step": 295
|
| 430 |
},
|
| 431 |
{
|
| 432 |
"epoch": 0.36764705882352944,
|
| 433 |
+
"grad_norm": 0.12248610966496465,
|
| 434 |
"learning_rate": 0.000797688481034551,
|
| 435 |
+
"loss": 0.9175,
|
| 436 |
"step": 300
|
| 437 |
},
|
| 438 |
{
|
| 439 |
"epoch": 0.3737745098039216,
|
| 440 |
+
"grad_norm": 0.10512495641494239,
|
| 441 |
"learning_rate": 0.00078902389564276,
|
| 442 |
"loss": 0.9239,
|
| 443 |
"step": 305
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 0.3799019607843137,
|
| 447 |
+
"grad_norm": 0.12079056888085157,
|
| 448 |
"learning_rate": 0.0007802269477971771,
|
| 449 |
+
"loss": 0.9167,
|
| 450 |
"step": 310
|
| 451 |
},
|
| 452 |
{
|
| 453 |
"epoch": 0.3860294117647059,
|
| 454 |
+
"grad_norm": 0.1311550506036977,
|
| 455 |
"learning_rate": 0.0007713016661806211,
|
| 456 |
+
"loss": 0.9165,
|
| 457 |
"step": 315
|
| 458 |
},
|
| 459 |
{
|
| 460 |
"epoch": 0.39215686274509803,
|
| 461 |
+
"grad_norm": 0.12748855363301959,
|
| 462 |
"learning_rate": 0.0007622521382481208,
|
| 463 |
+
"loss": 0.9099,
|
| 464 |
"step": 320
|
| 465 |
},
|
| 466 |
{
|
| 467 |
"epoch": 0.39828431372549017,
|
| 468 |
+
"grad_norm": 0.11389138878908127,
|
| 469 |
"learning_rate": 0.0007530825083550073,
|
| 470 |
+
"loss": 0.9034,
|
| 471 |
"step": 325
|
| 472 |
},
|
| 473 |
{
|
| 474 |
"epoch": 0.40441176470588236,
|
| 475 |
+
"grad_norm": 0.10172199627242663,
|
| 476 |
"learning_rate": 0.0007437969758589507,
|
| 477 |
+
"loss": 0.9147,
|
| 478 |
"step": 330
|
| 479 |
},
|
| 480 |
{
|
| 481 |
"epoch": 0.4105392156862745,
|
| 482 |
+
"grad_norm": 0.1136698134249708,
|
| 483 |
"learning_rate": 0.0007343997931968067,
|
| 484 |
+
"loss": 0.9076,
|
| 485 |
"step": 335
|
| 486 |
},
|
| 487 |
{
|
| 488 |
"epoch": 0.4166666666666667,
|
| 489 |
+
"grad_norm": 0.1110896296260987,
|
| 490 |
"learning_rate": 0.0007248952639371542,
|
| 491 |
+
"loss": 0.9075,
|
| 492 |
"step": 340
|
| 493 |
},
|
| 494 |
{
|
| 495 |
"epoch": 0.4227941176470588,
|
| 496 |
+
"grad_norm": 0.10357314484765201,
|
| 497 |
"learning_rate": 0.0007152877408094178,
|
| 498 |
+
"loss": 0.8998,
|
| 499 |
"step": 345
|
| 500 |
},
|
| 501 |
{
|
| 502 |
"epoch": 0.42892156862745096,
|
| 503 |
+
"grad_norm": 0.11773981651015025,
|
| 504 |
"learning_rate": 0.0007055816237104753,
|
| 505 |
+
"loss": 0.9094,
|
| 506 |
"step": 350
|
| 507 |
},
|
| 508 |
{
|
| 509 |
"epoch": 0.43504901960784315,
|
| 510 |
+
"grad_norm": 0.1283630128752841,
|
| 511 |
"learning_rate": 0.0006957813576896647,
|
| 512 |
+
"loss": 0.899,
|
| 513 |
"step": 355
|
| 514 |
},
|
| 515 |
{
|
| 516 |
"epoch": 0.4411764705882353,
|
| 517 |
+
"grad_norm": 0.1326640375854421,
|
| 518 |
"learning_rate": 0.000685891430913113,
|
| 519 |
+
"loss": 0.9091,
|
| 520 |
"step": 360
|
| 521 |
},
|
| 522 |
{
|
| 523 |
"epoch": 0.44730392156862747,
|
| 524 |
+
"grad_norm": 0.12057333477888295,
|
| 525 |
"learning_rate": 0.0006759163726083191,
|
| 526 |
+
"loss": 0.9005,
|
| 527 |
"step": 365
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 0.4534313725490196,
|
| 531 |
+
"grad_norm": 0.10157867473834796,
|
| 532 |
"learning_rate": 0.0006658607509899319,
|
| 533 |
+
"loss": 0.8995,
|
| 534 |
"step": 370
|
| 535 |
},
|
| 536 |
{
|
| 537 |
"epoch": 0.45955882352941174,
|
| 538 |
+
"grad_norm": 0.13679116304924,
|
| 539 |
"learning_rate": 0.0006557291711676738,
|
| 540 |
+
"loss": 0.9064,
|
| 541 |
"step": 375
|
| 542 |
},
|
| 543 |
{
|
| 544 |
"epoch": 0.46568627450980393,
|
| 545 |
+
"grad_norm": 0.10228308226469025,
|
| 546 |
"learning_rate": 0.0006455262730373672,
|
| 547 |
+
"loss": 0.8902,
|
| 548 |
"step": 380
|
| 549 |
},
|
| 550 |
{
|
| 551 |
"epoch": 0.47181372549019607,
|
| 552 |
+
"grad_norm": 0.11810749832493427,
|
| 553 |
"learning_rate": 0.0006352567291560318,
|
| 554 |
+
"loss": 0.8947,
|
| 555 |
"step": 385
|
| 556 |
},
|
| 557 |
{
|
| 558 |
"epoch": 0.47794117647058826,
|
| 559 |
+
"grad_norm": 0.11253919001414733,
|
| 560 |
"learning_rate": 0.0006249252426020216,
|
| 561 |
+
"loss": 0.8984,
|
| 562 |
"step": 390
|
| 563 |
},
|
| 564 |
{
|
| 565 |
"epoch": 0.4840686274509804,
|
| 566 |
+
"grad_norm": 0.10889918340035115,
|
| 567 |
"learning_rate": 0.0006145365448211866,
|
| 568 |
+
"loss": 0.9001,
|
| 569 |
"step": 395
|
| 570 |
},
|
| 571 |
{
|
| 572 |
"epoch": 0.49019607843137253,
|
| 573 |
+
"grad_norm": 0.10602494662106901,
|
| 574 |
"learning_rate": 0.0006040953934600423,
|
| 575 |
+
"loss": 0.8924,
|
| 576 |
"step": 400
|
| 577 |
},
|
| 578 |
{
|
| 579 |
"epoch": 0.4963235294117647,
|
| 580 |
+
"grad_norm": 0.09537450461248778,
|
| 581 |
"learning_rate": 0.0005936065701869403,
|
| 582 |
+
"loss": 0.8971,
|
| 583 |
"step": 405
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 0.5024509803921569,
|
| 587 |
+
"grad_norm": 0.1135732875240647,
|
| 588 |
"learning_rate": 0.0005830748785022368,
|
| 589 |
+
"loss": 0.8956,
|
| 590 |
"step": 410
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 0.508578431372549,
|
| 594 |
+
"grad_norm": 0.11824825784313651,
|
| 595 |
"learning_rate": 0.0005725051415384657,
|
| 596 |
+
"loss": 0.9014,
|
| 597 |
"step": 415
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 0.5147058823529411,
|
| 601 |
+
"grad_norm": 2.3957029087137602,
|
| 602 |
"learning_rate": 0.0005619021998515165,
|
| 603 |
+
"loss": 0.8937,
|
| 604 |
"step": 420
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 0.5208333333333334,
|
| 608 |
+
"grad_norm": 0.1305239745293032,
|
| 609 |
"learning_rate": 0.000551270909203838,
|
| 610 |
+
"loss": 0.889,
|
| 611 |
"step": 425
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 0.5269607843137255,
|
| 615 |
+
"grad_norm": 0.10923687170047386,
|
| 616 |
"learning_rate": 0.0005406161383406731,
|
| 617 |
+
"loss": 0.9009,
|
| 618 |
"step": 430
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 0.5330882352941176,
|
| 622 |
+
"grad_norm": 0.11720531307848668,
|
| 623 |
"learning_rate": 0.0005299427667603515,
|
| 624 |
+
"loss": 0.9035,
|
| 625 |
"step": 435
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 0.5392156862745098,
|
| 629 |
+
"grad_norm": 0.1043777454103823,
|
| 630 |
"learning_rate": 0.0005192556824796568,
|
| 631 |
+
"loss": 0.887,
|
| 632 |
"step": 440
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 0.5453431372549019,
|
| 636 |
+
"grad_norm": 0.12019301588246883,
|
| 637 |
"learning_rate": 0.0005085597797952905,
|
| 638 |
+
"loss": 0.8852,
|
| 639 |
"step": 445
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 0.5514705882352942,
|
| 643 |
+
"grad_norm": 0.09829925409523375,
|
| 644 |
"learning_rate": 0.0004978599570424639,
|
| 645 |
+
"loss": 0.8841,
|
| 646 |
"step": 450
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 0.5575980392156863,
|
| 650 |
+
"grad_norm": 0.110813034496191,
|
| 651 |
"learning_rate": 0.0004871611143516367,
|
| 652 |
+
"loss": 0.8888,
|
| 653 |
"step": 455
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 0.5637254901960784,
|
| 657 |
+
"grad_norm": 0.14013694091933743,
|
| 658 |
"learning_rate": 0.0004764681514044362,
|
| 659 |
+
"loss": 0.8863,
|
| 660 |
"step": 460
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 0.5698529411764706,
|
| 664 |
+
"grad_norm": 0.10955250297933698,
|
| 665 |
"learning_rate": 0.0004657859651897806,
|
| 666 |
+
"loss": 0.8904,
|
| 667 |
"step": 465
|
| 668 |
},
|
| 669 |
{
|
| 670 |
"epoch": 0.5759803921568627,
|
| 671 |
+
"grad_norm": 0.13711186271821346,
|
| 672 |
"learning_rate": 0.00045511944776123513,
|
| 673 |
+
"loss": 0.8789,
|
| 674 |
"step": 470
|
| 675 |
},
|
| 676 |
{
|
| 677 |
"epoch": 0.5821078431372549,
|
| 678 |
+
"grad_norm": 0.09396380277187082,
|
| 679 |
"learning_rate": 0.00044447348399663056,
|
| 680 |
+
"loss": 0.8847,
|
| 681 |
"step": 475
|
| 682 |
},
|
| 683 |
{
|
| 684 |
"epoch": 0.5882352941176471,
|
| 685 |
+
"grad_norm": 0.21392349020058346,
|
| 686 |
"learning_rate": 0.0004338529493609647,
|
| 687 |
+
"loss": 0.8824,
|
| 688 |
"step": 480
|
| 689 |
},
|
| 690 |
{
|
| 691 |
"epoch": 0.5943627450980392,
|
| 692 |
+
"grad_norm": 0.12755805564480172,
|
| 693 |
"learning_rate": 0.00042326270767361815,
|
| 694 |
+
"loss": 0.8884,
|
| 695 |
"step": 485
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 0.6004901960784313,
|
| 699 |
+
"grad_norm": 0.09157375745294742,
|
| 700 |
"learning_rate": 0.00041270760888089997,
|
| 701 |
+
"loss": 0.8825,
|
| 702 |
"step": 490
|
| 703 |
},
|
| 704 |
{
|
| 705 |
"epoch": 0.6066176470588235,
|
| 706 |
+
"grad_norm": 0.10173653886247282,
|
| 707 |
"learning_rate": 0.00040219248683494925,
|
| 708 |
+
"loss": 0.8637,
|
| 709 |
"step": 495
|
| 710 |
},
|
| 711 |
{
|
| 712 |
"epoch": 0.6127450980392157,
|
| 713 |
+
"grad_norm": 0.12386704656315299,
|
| 714 |
"learning_rate": 0.0003917221570800065,
|
| 715 |
+
"loss": 0.8719,
|
| 716 |
"step": 500
|
| 717 |
},
|
| 718 |
{
|
| 719 |
"epoch": 0.6188725490196079,
|
| 720 |
+
"grad_norm": 0.10921071757131698,
|
| 721 |
"learning_rate": 0.000381301414647068,
|
| 722 |
+
"loss": 0.8707,
|
| 723 |
"step": 505
|
| 724 |
},
|
| 725 |
{
|
| 726 |
"epoch": 0.625,
|
| 727 |
+
"grad_norm": 0.10860919138034633,
|
| 728 |
"learning_rate": 0.0003709350318579371,
|
| 729 |
+
"loss": 0.8934,
|
| 730 |
"step": 510
|
| 731 |
},
|
| 732 |
{
|
| 733 |
"epoch": 0.6311274509803921,
|
| 734 |
+
"grad_norm": 0.08765926558701954,
|
| 735 |
"learning_rate": 0.0003606277561396726,
|
| 736 |
+
"loss": 0.8595,
|
| 737 |
"step": 515
|
| 738 |
},
|
| 739 |
{
|
| 740 |
"epoch": 0.6372549019607843,
|
| 741 |
+
"grad_norm": 0.08795902636008367,
|
| 742 |
"learning_rate": 0.00035038430785044053,
|
| 743 |
+
"loss": 0.8629,
|
| 744 |
"step": 520
|
| 745 |
},
|
| 746 |
{
|
| 747 |
"epoch": 0.6433823529411765,
|
| 748 |
+
"grad_norm": 0.10125788693590333,
|
| 749 |
"learning_rate": 0.00034020937811776156,
|
| 750 |
+
"loss": 0.8597,
|
| 751 |
"step": 525
|
| 752 |
},
|
| 753 |
{
|
| 754 |
"epoch": 0.6495098039215687,
|
| 755 |
+
"grad_norm": 0.09640732281156021,
|
| 756 |
"learning_rate": 0.00033010762669014347,
|
| 757 |
+
"loss": 0.8672,
|
| 758 |
"step": 530
|
| 759 |
},
|
| 760 |
{
|
| 761 |
"epoch": 0.6556372549019608,
|
| 762 |
+
"grad_norm": 0.09206201588796137,
|
| 763 |
"learning_rate": 0.00032008367980308734,
|
| 764 |
+
"loss": 0.8723,
|
| 765 |
"step": 535
|
| 766 |
},
|
| 767 |
{
|
| 768 |
"epoch": 0.6617647058823529,
|
| 769 |
+
"grad_norm": 0.089094237721721,
|
| 770 |
"learning_rate": 0.0003101421280604379,
|
| 771 |
+
"loss": 0.884,
|
| 772 |
"step": 540
|
| 773 |
},
|
| 774 |
{
|
| 775 |
"epoch": 0.6678921568627451,
|
| 776 |
+
"grad_norm": 0.10047930336023028,
|
| 777 |
"learning_rate": 0.00030028752433205476,
|
| 778 |
+
"loss": 0.8612,
|
| 779 |
"step": 545
|
| 780 |
},
|
| 781 |
{
|
| 782 |
"epoch": 0.6740196078431373,
|
| 783 |
+
"grad_norm": 0.09796290633516842,
|
| 784 |
"learning_rate": 0.00029052438166876307,
|
| 785 |
+
"loss": 0.8527,
|
| 786 |
"step": 550
|
| 787 |
},
|
| 788 |
{
|
| 789 |
"epoch": 0.6801470588235294,
|
| 790 |
+
"grad_norm": 0.08908481799962162,
|
| 791 |
"learning_rate": 0.0002808571712355389,
|
| 792 |
+
"loss": 0.8636,
|
| 793 |
"step": 555
|
| 794 |
},
|
| 795 |
{
|
| 796 |
"epoch": 0.6862745098039216,
|
| 797 |
+
"grad_norm": 0.09854862986040251,
|
| 798 |
"learning_rate": 0.00027129032026388045,
|
| 799 |
+
"loss": 0.8581,
|
| 800 |
"step": 560
|
| 801 |
},
|
| 802 |
{
|
| 803 |
"epoch": 0.6924019607843137,
|
| 804 |
+
"grad_norm": 0.096989721310236,
|
| 805 |
"learning_rate": 0.00026182821002429345,
|
| 806 |
+
"loss": 0.8617,
|
| 807 |
"step": 565
|
| 808 |
},
|
| 809 |
{
|
| 810 |
"epoch": 0.6985294117647058,
|
| 811 |
+
"grad_norm": 0.09027729876751488,
|
| 812 |
"learning_rate": 0.00025247517381983136,
|
| 813 |
+
"loss": 0.8654,
|
| 814 |
"step": 570
|
| 815 |
},
|
| 816 |
{
|
| 817 |
"epoch": 0.7046568627450981,
|
| 818 |
+
"grad_norm": 0.10227245851698821,
|
| 819 |
"learning_rate": 0.00024323549500159802,
|
| 820 |
+
"loss": 0.8618,
|
| 821 |
"step": 575
|
| 822 |
},
|
| 823 |
{
|
| 824 |
"epoch": 0.7107843137254902,
|
| 825 |
+
"grad_norm": 0.09927553647728089,
|
| 826 |
"learning_rate": 0.0002341134050071283,
|
| 827 |
+
"loss": 0.855,
|
| 828 |
"step": 580
|
| 829 |
},
|
| 830 |
{
|
| 831 |
"epoch": 0.7169117647058824,
|
| 832 |
+
"grad_norm": 0.09142338818988954,
|
| 833 |
"learning_rate": 0.00022511308142254488,
|
| 834 |
+
"loss": 0.8577,
|
| 835 |
"step": 585
|
| 836 |
},
|
| 837 |
{
|
| 838 |
"epoch": 0.7230392156862745,
|
| 839 |
+
"grad_norm": 0.10507626286878373,
|
| 840 |
"learning_rate": 0.000216238646069373,
|
| 841 |
+
"loss": 0.8605,
|
| 842 |
"step": 590
|
| 843 |
},
|
| 844 |
{
|
| 845 |
"epoch": 0.7291666666666666,
|
| 846 |
+
"grad_norm": 0.09773601600409339,
|
| 847 |
"learning_rate": 0.00020749416311689845,
|
| 848 |
+
"loss": 0.8605,
|
| 849 |
"step": 595
|
| 850 |
},
|
| 851 |
{
|
| 852 |
"epoch": 0.7352941176470589,
|
| 853 |
+
"grad_norm": 0.1053760063340528,
|
| 854 |
"learning_rate": 0.00019888363722092372,
|
| 855 |
+
"loss": 0.8631,
|
| 856 |
"step": 600
|
| 857 |
},
|
| 858 |
{
|
| 859 |
"epoch": 0.741421568627451,
|
| 860 |
+
"grad_norm": 0.09919853848427344,
|
| 861 |
"learning_rate": 0.00019041101168978093,
|
| 862 |
+
"loss": 0.8589,
|
| 863 |
"step": 605
|
| 864 |
},
|
| 865 |
{
|
| 866 |
"epoch": 0.7475490196078431,
|
| 867 |
+
"grad_norm": 0.09240852582600491,
|
| 868 |
"learning_rate": 0.00018208016667844152,
|
| 869 |
+
"loss": 0.8616,
|
| 870 |
"step": 610
|
| 871 |
},
|
| 872 |
{
|
| 873 |
"epoch": 0.7536764705882353,
|
| 874 |
+
"grad_norm": 0.09385869340911827,
|
| 875 |
"learning_rate": 0.00017389491741154372,
|
| 876 |
+
"loss": 0.8543,
|
| 877 |
"step": 615
|
| 878 |
},
|
| 879 |
{
|
| 880 |
"epoch": 0.7598039215686274,
|
| 881 |
+
"grad_norm": 0.08602993504708097,
|
| 882 |
"learning_rate": 0.00016585901243616042,
|
| 883 |
+
"loss": 0.8566,
|
| 884 |
"step": 620
|
| 885 |
},
|
| 886 |
{
|
| 887 |
"epoch": 0.7659313725490197,
|
| 888 |
+
"grad_norm": 0.08661913403120794,
|
| 889 |
"learning_rate": 0.0001579761319050991,
|
| 890 |
+
"loss": 0.8546,
|
| 891 |
"step": 625
|
| 892 |
},
|
| 893 |
{
|
| 894 |
"epoch": 0.7720588235294118,
|
| 895 |
+
"grad_norm": 0.08756073235275695,
|
| 896 |
"learning_rate": 0.00015024988589152537,
|
| 897 |
+
"loss": 0.8582,
|
| 898 |
"step": 630
|
| 899 |
},
|
| 900 |
{
|
| 901 |
"epoch": 0.7781862745098039,
|
| 902 |
+
"grad_norm": 0.08339963011288148,
|
| 903 |
"learning_rate": 0.0001426838127356823,
|
| 904 |
+
"loss": 0.8541,
|
| 905 |
"step": 635
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 0.7843137254901961,
|
| 909 |
+
"grad_norm": 0.07897307103939846,
|
| 910 |
"learning_rate": 0.0001352813774244565,
|
| 911 |
+
"loss": 0.849,
|
| 912 |
"step": 640
|
| 913 |
},
|
| 914 |
{
|
| 915 |
"epoch": 0.7904411764705882,
|
| 916 |
+
"grad_norm": 0.08692536794832408,
|
| 917 |
"learning_rate": 0.00012804597000454215,
|
| 918 |
+
"loss": 0.8559,
|
| 919 |
"step": 645
|
| 920 |
},
|
| 921 |
{
|
| 922 |
"epoch": 0.7965686274509803,
|
| 923 |
+
"grad_norm": 0.08695762926336753,
|
| 924 |
"learning_rate": 0.00012098090402992085,
|
| 925 |
+
"loss": 0.8665,
|
| 926 |
"step": 650
|
| 927 |
},
|
| 928 |
{
|
| 929 |
"epoch": 0.8026960784313726,
|
| 930 |
+
"grad_norm": 0.08241315305272631,
|
| 931 |
"learning_rate": 0.00011408941504437532,
|
| 932 |
+
"loss": 0.8544,
|
| 933 |
"step": 655
|
| 934 |
},
|
| 935 |
{
|
| 936 |
"epoch": 0.8088235294117647,
|
| 937 |
+
"grad_norm": 0.07821925622204019,
|
| 938 |
"learning_rate": 0.00010737465909972776,
|
| 939 |
+
"loss": 0.8474,
|
| 940 |
"step": 660
|
| 941 |
},
|
| 942 |
{
|
| 943 |
"epoch": 0.8149509803921569,
|
| 944 |
+
"grad_norm": 0.08521471066806094,
|
| 945 |
"learning_rate": 0.00010083971131048159,
|
| 946 |
+
"loss": 0.8495,
|
| 947 |
"step": 665
|
| 948 |
},
|
| 949 |
{
|
| 950 |
"epoch": 0.821078431372549,
|
| 951 |
+
"grad_norm": 0.08381156457580924,
|
| 952 |
"learning_rate": 9.448756444553224e-05,
|
| 953 |
+
"loss": 0.8506,
|
| 954 |
"step": 670
|
| 955 |
},
|
| 956 |
{
|
| 957 |
"epoch": 0.8272058823529411,
|
| 958 |
+
"grad_norm": 0.080205577901611,
|
| 959 |
"learning_rate": 8.832112755758598e-05,
|
| 960 |
+
"loss": 0.8482,
|
| 961 |
"step": 675
|
| 962 |
},
|
| 963 |
{
|
| 964 |
"epoch": 0.8333333333333334,
|
| 965 |
+
"grad_norm": 0.08210122013268317,
|
| 966 |
"learning_rate": 8.234322465092047e-05,
|
| 967 |
+
"loss": 0.8491,
|
| 968 |
"step": 680
|
| 969 |
},
|
| 970 |
{
|
| 971 |
"epoch": 0.8394607843137255,
|
| 972 |
+
"grad_norm": 0.08195821975889148,
|
| 973 |
"learning_rate": 7.655659338809329e-05,
|
| 974 |
+
"loss": 0.8484,
|
| 975 |
"step": 685
|
| 976 |
},
|
| 977 |
{
|
| 978 |
"epoch": 0.8455882352941176,
|
| 979 |
+
"grad_norm": 0.08225068138923354,
|
| 980 |
"learning_rate": 7.096388383619079e-05,
|
| 981 |
+
"loss": 0.8436,
|
| 982 |
"step": 690
|
| 983 |
},
|
| 984 |
{
|
| 985 |
"epoch": 0.8517156862745098,
|
| 986 |
+
"grad_norm": 0.10816220803390626,
|
| 987 |
"learning_rate": 6.556765725319525e-05,
|
| 988 |
+
"loss": 0.8479,
|
| 989 |
"step": 695
|
| 990 |
},
|
| 991 |
{
|
| 992 |
"epoch": 0.8578431372549019,
|
| 993 |
+
"grad_norm": 0.08954785260614277,
|
| 994 |
"learning_rate": 6.037038491501978e-05,
|
| 995 |
+
"loss": 0.8524,
|
| 996 |
"step": 700
|
| 997 |
},
|
| 998 |
{
|
| 999 |
"epoch": 0.8639705882352942,
|
| 1000 |
+
"grad_norm": 0.07665058203914679,
|
| 1001 |
"learning_rate": 5.53744469837551e-05,
|
| 1002 |
+
"loss": 0.8431,
|
| 1003 |
"step": 705
|
| 1004 |
},
|
| 1005 |
{
|
| 1006 |
"epoch": 0.8700980392156863,
|
| 1007 |
+
"grad_norm": 0.0782138298232773,
|
| 1008 |
"learning_rate": 5.058213141764151e-05,
|
| 1009 |
+
"loss": 0.8438,
|
| 1010 |
"step": 710
|
| 1011 |
},
|
| 1012 |
{
|
| 1013 |
"epoch": 0.8762254901960784,
|
| 1014 |
+
"grad_norm": 0.08269251578264038,
|
| 1015 |
"learning_rate": 4.599563292326592e-05,
|
| 1016 |
+
"loss": 0.8485,
|
| 1017 |
"step": 715
|
| 1018 |
},
|
| 1019 |
{
|
| 1020 |
"epoch": 0.8823529411764706,
|
| 1021 |
+
"grad_norm": 0.0907687363220474,
|
| 1022 |
"learning_rate": 4.161705195046761e-05,
|
| 1023 |
+
"loss": 0.8443,
|
| 1024 |
"step": 720
|
| 1025 |
},
|
| 1026 |
{
|
| 1027 |
"epoch": 0.8884803921568627,
|
| 1028 |
+
"grad_norm": 0.08259230750361556,
|
| 1029 |
"learning_rate": 3.744839373040682e-05,
|
| 1030 |
+
"loss": 0.8467,
|
| 1031 |
"step": 725
|
| 1032 |
},
|
| 1033 |
{
|
| 1034 |
"epoch": 0.8946078431372549,
|
| 1035 |
+
"grad_norm": 0.07999560967778772,
|
| 1036 |
"learning_rate": 3.349156735724274e-05,
|
| 1037 |
+
"loss": 0.848,
|
| 1038 |
"step": 730
|
| 1039 |
},
|
| 1040 |
{
|
| 1041 |
"epoch": 0.9007352941176471,
|
| 1042 |
+
"grad_norm": 0.07524504711853225,
|
| 1043 |
"learning_rate": 2.9748384913837522e-05,
|
| 1044 |
+
"loss": 0.8348,
|
| 1045 |
"step": 735
|
| 1046 |
},
|
| 1047 |
{
|
| 1048 |
"epoch": 0.9068627450980392,
|
| 1049 |
+
"grad_norm": 0.08135847243984051,
|
| 1050 |
"learning_rate": 2.622056064188738e-05,
|
| 1051 |
+
"loss": 0.854,
|
| 1052 |
"step": 740
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
"epoch": 0.9129901960784313,
|
| 1056 |
+
"grad_norm": 0.07885634814452873,
|
| 1057 |
"learning_rate": 2.2909710156863274e-05,
|
| 1058 |
+
"loss": 0.8514,
|
| 1059 |
"step": 745
|
| 1060 |
},
|
| 1061 |
{
|
| 1062 |
"epoch": 0.9191176470588235,
|
| 1063 |
+
"grad_norm": 0.08736339560766254,
|
| 1064 |
"learning_rate": 1.981734970811644e-05,
|
| 1065 |
+
"loss": 0.8417,
|
| 1066 |
"step": 750
|
| 1067 |
},
|
| 1068 |
{
|
| 1069 |
"epoch": 0.9252450980392157,
|
| 1070 |
+
"grad_norm": 0.0767085793238129,
|
| 1071 |
"learning_rate": 1.6944895484492072e-05,
|
| 1072 |
+
"loss": 0.8523,
|
| 1073 |
"step": 755
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
"epoch": 0.9313725490196079,
|
| 1077 |
+
"grad_norm": 0.08318918651152993,
|
| 1078 |
"learning_rate": 1.429366296576623e-05,
|
| 1079 |
+
"loss": 0.8511,
|
| 1080 |
"step": 760
|
| 1081 |
},
|
| 1082 |
{
|
| 1083 |
"epoch": 0.9375,
|
| 1084 |
+
"grad_norm": 0.07933377923909153,
|
| 1085 |
"learning_rate": 1.1864866320203115e-05,
|
| 1086 |
+
"loss": 0.8479,
|
| 1087 |
"step": 765
|
| 1088 |
},
|
| 1089 |
{
|
| 1090 |
"epoch": 0.9436274509803921,
|
| 1091 |
+
"grad_norm": 0.09008515851237198,
|
| 1092 |
"learning_rate": 9.659617848510882e-06,
|
| 1093 |
+
"loss": 0.8449,
|
| 1094 |
"step": 770
|
| 1095 |
},
|
| 1096 |
{
|
| 1097 |
"epoch": 0.9497549019607843,
|
| 1098 |
+
"grad_norm": 0.07787795748629618,
|
| 1099 |
"learning_rate": 7.678927474447817e-06,
|
| 1100 |
+
"loss": 0.8446,
|
| 1101 |
"step": 775
|
| 1102 |
},
|
| 1103 |
{
|
| 1104 |
"epoch": 0.9558823529411765,
|
| 1105 |
+
"grad_norm": 0.08423127476840589,
|
| 1106 |
"learning_rate": 5.923702282314092e-06,
|
| 1107 |
+
"loss": 0.8466,
|
| 1108 |
"step": 780
|
| 1109 |
},
|
| 1110 |
{
|
| 1111 |
"epoch": 0.9620098039215687,
|
| 1112 |
+
"grad_norm": 0.07739503343274702,
|
| 1113 |
"learning_rate": 4.394746101540115e-06,
|
| 1114 |
+
"loss": 0.8423,
|
| 1115 |
"step": 785
|
| 1116 |
},
|
| 1117 |
{
|
| 1118 |
"epoch": 0.9681372549019608,
|
| 1119 |
+
"grad_norm": 0.07932764141883414,
|
| 1120 |
"learning_rate": 3.092759138561607e-06,
|
| 1121 |
+
"loss": 0.8405,
|
| 1122 |
"step": 790
|
| 1123 |
},
|
| 1124 |
{
|
| 1125 |
"epoch": 0.9742647058823529,
|
| 1126 |
+
"grad_norm": 0.08063542360073593,
|
| 1127 |
"learning_rate": 2.018337656150726e-06,
|
| 1128 |
+
"loss": 0.8461,
|
| 1129 |
"step": 795
|
| 1130 |
},
|
| 1131 |
{
|
| 1132 |
"epoch": 0.9803921568627451,
|
| 1133 |
+
"grad_norm": 0.08993712709783745,
|
| 1134 |
"learning_rate": 1.1719737003492159e-06,
|
| 1135 |
+
"loss": 0.8388,
|
| 1136 |
"step": 800
|
| 1137 |
},
|
| 1138 |
{
|
| 1139 |
"epoch": 0.9865196078431373,
|
| 1140 |
+
"grad_norm": 0.0792762901452283,
|
| 1141 |
"learning_rate": 5.540548751292173e-07,
|
| 1142 |
+
"loss": 0.8346,
|
| 1143 |
"step": 805
|
| 1144 |
},
|
| 1145 |
{
|
| 1146 |
"epoch": 0.9926470588235294,
|
| 1147 |
+
"grad_norm": 0.09250878108386706,
|
| 1148 |
"learning_rate": 1.6486416488459277e-07,
|
| 1149 |
+
"loss": 0.8478,
|
| 1150 |
"step": 810
|
| 1151 |
},
|
| 1152 |
{
|
| 1153 |
"epoch": 0.9987745098039216,
|
| 1154 |
+
"grad_norm": 0.07786954435284818,
|
| 1155 |
"learning_rate": 4.579804834703438e-09,
|
| 1156 |
+
"loss": 0.8401,
|
| 1157 |
"step": 815
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 1.0,
|
| 1161 |
+
"eval_loss": 1.202426552772522,
|
| 1162 |
+
"eval_runtime": 111.4035,
|
| 1163 |
+
"eval_samples_per_second": 187.974,
|
| 1164 |
+
"eval_steps_per_second": 5.88,
|
| 1165 |
"step": 816
|
| 1166 |
},
|
| 1167 |
{
|
| 1168 |
"epoch": 1.0,
|
| 1169 |
"step": 816,
|
| 1170 |
"total_flos": 80063181619200.0,
|
| 1171 |
+
"train_loss": 0.914715180794398,
|
| 1172 |
+
"train_runtime": 1874.4496,
|
| 1173 |
+
"train_samples_per_second": 55.714,
|
| 1174 |
+
"train_steps_per_second": 0.435
|
| 1175 |
}
|
| 1176 |
],
|
| 1177 |
"logging_steps": 5,
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 6456
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:850382808a98614a406a920bb7e54bab43a949c3aee2c03fb6a5de54b03f913e
|
| 3 |
size 6456
|