Model save
Browse files- README.md +13 -14
- adapter_model.safetensors +1 -1
- all_results.json +5 -10
- runs/Jun10_06-37-27_48ddfe8e991f/events.out.tfevents.1718001468.48ddfe8e991f.199825.0 +2 -2
- train_results.json +5 -5
- trainer_state.json +254 -254
README.md
CHANGED
|
@@ -2,13 +2,12 @@
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
| 8 |
- generated_from_trainer
|
| 9 |
base_model: google/gemma-2b
|
| 10 |
datasets:
|
| 11 |
-
-
|
| 12 |
model-index:
|
| 13 |
- name: gemma2b-summarize-gemini1_5flash-64k
|
| 14 |
results: []
|
|
@@ -19,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
|
| 20 |
# gemma2b-summarize-gemini1_5flash-64k
|
| 21 |
|
| 22 |
-
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
-
- Loss: 2.
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -58,16 +57,16 @@ The following hyperparameters were used during training:
|
|
| 58 |
|
| 59 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 60 |
|:-------------:|:------:|:----:|:---------------:|
|
| 61 |
-
| 1.
|
| 62 |
-
| 1.
|
| 63 |
-
| 1.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
|
| 72 |
|
| 73 |
### Framework versions
|
|
|
|
| 2 |
license: gemma
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
- generated_from_trainer
|
| 8 |
base_model: google/gemma-2b
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: gemma2b-summarize-gemini1_5flash-64k
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# gemma2b-summarize-gemini1_5flash-64k
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 2.5322
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 57 |
|
| 58 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 1.2288 | 0.9905 | 52 | 2.5091 |
|
| 61 |
+
| 1.0492 | 2.0 | 105 | 2.4525 |
|
| 62 |
+
| 1.0025 | 2.9905 | 157 | 2.4599 |
|
| 63 |
+
| 0.9597 | 4.0 | 210 | 2.4714 |
|
| 64 |
+
| 0.9394 | 4.9905 | 262 | 2.4863 |
|
| 65 |
+
| 0.912 | 6.0 | 315 | 2.5085 |
|
| 66 |
+
| 0.8901 | 6.9905 | 367 | 2.5172 |
|
| 67 |
+
| 0.8887 | 8.0 | 420 | 2.5294 |
|
| 68 |
+
| 0.8816 | 8.9905 | 472 | 2.5311 |
|
| 69 |
+
| 0.8797 | 9.9048 | 520 | 2.5322 |
|
| 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 78480320
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37862d1fcacaacad7227bafda722c1c245ca97683c7ced949e620f9d38111786
|
| 3 |
size 78480320
|
all_results.json
CHANGED
|
@@ -1,14 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 9.904761904761905,
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"eval_samples_per_second": 19.982,
|
| 7 |
-
"eval_steps_per_second": 1.998,
|
| 8 |
-
"total_flos": 8.18501087353897e+17,
|
| 9 |
-
"train_loss": 1.098491345002101,
|
| 10 |
-
"train_runtime": 2780.3493,
|
| 11 |
"train_samples": 63353,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 9.904761904761905,
|
| 3 |
+
"total_flos": 8.265211656850637e+17,
|
| 4 |
+
"train_loss": 1.0387523100926326,
|
| 5 |
+
"train_runtime": 3563.5293,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
"train_samples": 63353,
|
| 7 |
+
"train_samples_per_second": 18.703,
|
| 8 |
+
"train_steps_per_second": 0.146
|
| 9 |
}
|
runs/Jun10_06-37-27_48ddfe8e991f/events.out.tfevents.1718001468.48ddfe8e991f.199825.0
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:f6a5b092972114847ec0647c6a64142467c6db558679f538fd7a1081e9450b1c
|
| 3 |
+
size 30709
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 9.904761904761905,
|
| 3 |
-
"total_flos": 8.
|
| 4 |
-
"train_loss": 1.
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 63353,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 9.904761904761905,
|
| 3 |
+
"total_flos": 8.265211656850637e+17,
|
| 4 |
+
"train_loss": 1.0387523100926326,
|
| 5 |
+
"train_runtime": 3563.5293,
|
| 6 |
"train_samples": 63353,
|
| 7 |
+
"train_samples_per_second": 18.703,
|
| 8 |
+
"train_steps_per_second": 0.146
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -10,827 +10,827 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01904761904761905,
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 3.846153846153847e-06,
|
| 15 |
"loss": 3.0068,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09523809523809523,
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 1.923076923076923e-05,
|
| 22 |
-
"loss": 3.
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.19047619047619047,
|
| 27 |
-
"grad_norm": 2.
|
| 28 |
"learning_rate": 3.846153846153846e-05,
|
| 29 |
-
"loss": 2.
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2857142857142857,
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 5.769230769230769e-05,
|
| 36 |
-
"loss": 2.
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.38095238095238093,
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 7.692307692307693e-05,
|
| 43 |
-
"loss": 2.
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.47619047619047616,
|
| 48 |
-
"grad_norm":
|
| 49 |
"learning_rate": 9.615384615384617e-05,
|
| 50 |
-
"loss":
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.5714285714285714,
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 0.00011538461538461538,
|
| 57 |
-
"loss": 1.
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.6666666666666666,
|
| 62 |
-
"grad_norm": 0.
|
| 63 |
"learning_rate": 0.00013461538461538464,
|
| 64 |
-
"loss": 1.
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.7619047619047619,
|
| 69 |
-
"grad_norm": 0.
|
| 70 |
"learning_rate": 0.00015384615384615385,
|
| 71 |
-
"loss": 1.
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.8571428571428571,
|
| 76 |
-
"grad_norm": 0.
|
| 77 |
"learning_rate": 0.0001730769230769231,
|
| 78 |
-
"loss": 1.
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.9523809523809523,
|
| 83 |
-
"grad_norm": 1.
|
| 84 |
"learning_rate": 0.00019230769230769233,
|
| 85 |
-
"loss": 1.
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.9904761904761905,
|
| 90 |
-
"eval_loss": 2.
|
| 91 |
-
"eval_runtime": 0.
|
| 92 |
-
"eval_samples_per_second":
|
| 93 |
-
"eval_steps_per_second": 1.
|
| 94 |
"step": 52
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 1.0476190476190477,
|
| 98 |
-
"grad_norm":
|
| 99 |
"learning_rate": 0.00019997972289848503,
|
| 100 |
-
"loss": 1.
|
| 101 |
"step": 55
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 1.1428571428571428,
|
| 105 |
-
"grad_norm": 0.
|
| 106 |
"learning_rate": 0.00019985583705641418,
|
| 107 |
-
"loss": 1.
|
| 108 |
"step": 60
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 1.2380952380952381,
|
| 112 |
-
"grad_norm": 0.
|
| 113 |
"learning_rate": 0.00019961946980917456,
|
| 114 |
-
"loss": 1.
|
| 115 |
"step": 65
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 1.3333333333333333,
|
| 119 |
-
"grad_norm": 0.
|
| 120 |
"learning_rate": 0.0001992708874098054,
|
| 121 |
-
"loss": 1.
|
| 122 |
"step": 70
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 1.4285714285714286,
|
| 126 |
-
"grad_norm": 0.
|
| 127 |
"learning_rate": 0.0001988104825147528,
|
| 128 |
-
"loss": 1.
|
| 129 |
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 1.5238095238095237,
|
| 133 |
-
"grad_norm": 0.
|
| 134 |
"learning_rate": 0.00019823877374156647,
|
| 135 |
-
"loss": 1.
|
| 136 |
"step": 80
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 1.619047619047619,
|
| 140 |
-
"grad_norm": 0.
|
| 141 |
"learning_rate": 0.00019755640508470942,
|
| 142 |
-
"loss": 1.
|
| 143 |
"step": 85
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 1.7142857142857144,
|
| 147 |
-
"grad_norm": 0.
|
| 148 |
"learning_rate": 0.00019676414519013781,
|
| 149 |
-
"loss": 1.
|
| 150 |
"step": 90
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 1.8095238095238095,
|
| 154 |
-
"grad_norm": 0.
|
| 155 |
"learning_rate": 0.00019586288648946947,
|
| 156 |
-
"loss": 1.
|
| 157 |
"step": 95
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 1.9047619047619047,
|
| 161 |
-
"grad_norm": 0.
|
| 162 |
"learning_rate": 0.00019485364419471454,
|
| 163 |
-
"loss": 1.
|
| 164 |
"step": 100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 2.0,
|
| 168 |
-
"grad_norm": 0.
|
| 169 |
"learning_rate": 0.00019373755515470254,
|
| 170 |
-
"loss": 1.
|
| 171 |
"step": 105
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 2.0,
|
| 175 |
-
"eval_loss": 2.
|
| 176 |
-
"eval_runtime": 0.
|
| 177 |
-
"eval_samples_per_second": 20.
|
| 178 |
-
"eval_steps_per_second": 2.
|
| 179 |
"step": 105
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 2.0952380952380953,
|
| 183 |
-
"grad_norm": 0.
|
| 184 |
"learning_rate": 0.00019251587657449236,
|
| 185 |
-
"loss": 1.
|
| 186 |
"step": 110
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 2.1904761904761907,
|
| 190 |
-
"grad_norm": 0.
|
| 191 |
"learning_rate": 0.00019118998459920902,
|
| 192 |
-
"loss": 1.
|
| 193 |
"step": 115
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 2.2857142857142856,
|
| 197 |
-
"grad_norm": 0.
|
| 198 |
"learning_rate": 0.0001897613727639014,
|
| 199 |
-
"loss": 1.
|
| 200 |
"step": 120
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 2.380952380952381,
|
| 204 |
-
"grad_norm": 0.
|
| 205 |
"learning_rate": 0.0001882316503111678,
|
| 206 |
-
"loss": 1.
|
| 207 |
"step": 125
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 2.4761904761904763,
|
| 211 |
-
"grad_norm": 0.
|
| 212 |
"learning_rate": 0.00018660254037844388,
|
| 213 |
-
"loss": 1.
|
| 214 |
"step": 130
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 2.571428571428571,
|
| 218 |
-
"grad_norm": 0.
|
| 219 |
"learning_rate": 0.00018487587805699526,
|
| 220 |
-
"loss": 1.
|
| 221 |
"step": 135
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 2.6666666666666665,
|
| 225 |
-
"grad_norm": 0.
|
| 226 |
"learning_rate": 0.00018305360832480117,
|
| 227 |
-
"loss": 1.
|
| 228 |
"step": 140
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 2.761904761904762,
|
| 232 |
-
"grad_norm": 0.
|
| 233 |
"learning_rate": 0.00018113778385565733,
|
| 234 |
-
"loss": 1.
|
| 235 |
"step": 145
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 2.857142857142857,
|
| 239 |
-
"grad_norm": 0.
|
| 240 |
"learning_rate": 0.0001791305627069662,
|
| 241 |
-
"loss":
|
| 242 |
"step": 150
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 2.9523809523809526,
|
| 246 |
-
"grad_norm": 0.
|
| 247 |
"learning_rate": 0.00017703420588881946,
|
| 248 |
-
"loss": 1.
|
| 249 |
"step": 155
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 2.9904761904761905,
|
| 253 |
-
"eval_loss": 2.
|
| 254 |
-
"eval_runtime": 0.
|
| 255 |
-
"eval_samples_per_second":
|
| 256 |
-
"eval_steps_per_second":
|
| 257 |
"step": 157
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 3.0476190476190474,
|
| 261 |
-
"grad_norm": 0.
|
| 262 |
"learning_rate": 0.00017485107481711012,
|
| 263 |
-
"loss":
|
| 264 |
"step": 160
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 3.142857142857143,
|
| 268 |
-
"grad_norm": 0.
|
| 269 |
"learning_rate": 0.00017258362865354426,
|
| 270 |
-
"loss":
|
| 271 |
"step": 165
|
| 272 |
},
|
| 273 |
{
|
| 274 |
"epoch": 3.238095238095238,
|
| 275 |
-
"grad_norm": 0.
|
| 276 |
"learning_rate": 0.00017023442153554777,
|
| 277 |
-
"loss":
|
| 278 |
"step": 170
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 3.3333333333333335,
|
| 282 |
-
"grad_norm": 0.
|
| 283 |
"learning_rate": 0.0001678060996991891,
|
| 284 |
-
"loss":
|
| 285 |
"step": 175
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 3.4285714285714284,
|
| 289 |
-
"grad_norm": 0.
|
| 290 |
"learning_rate": 0.0001653013984983585,
|
| 291 |
-
"loss":
|
| 292 |
"step": 180
|
| 293 |
},
|
| 294 |
{
|
| 295 |
"epoch": 3.5238095238095237,
|
| 296 |
-
"grad_norm":
|
| 297 |
"learning_rate": 0.00016272313932356162,
|
| 298 |
-
"loss":
|
| 299 |
"step": 185
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 3.619047619047619,
|
| 303 |
-
"grad_norm": 0.
|
| 304 |
"learning_rate": 0.0001600742264237979,
|
| 305 |
-
"loss":
|
| 306 |
"step": 190
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 3.7142857142857144,
|
| 310 |
-
"grad_norm": 0.
|
| 311 |
"learning_rate": 0.0001573576436351046,
|
| 312 |
-
"loss":
|
| 313 |
"step": 195
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 3.8095238095238093,
|
| 317 |
-
"grad_norm": 0.
|
| 318 |
"learning_rate": 0.00015457645101945046,
|
| 319 |
-
"loss": 0.
|
| 320 |
"step": 200
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 3.9047619047619047,
|
| 324 |
-
"grad_norm": 0.
|
| 325 |
"learning_rate": 0.00015173378141776568,
|
| 326 |
-
"loss":
|
| 327 |
"step": 205
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 4.0,
|
| 331 |
-
"grad_norm": 0.
|
| 332 |
"learning_rate": 0.00014883283692099112,
|
| 333 |
-
"loss": 0.
|
| 334 |
"step": 210
|
| 335 |
},
|
| 336 |
{
|
| 337 |
"epoch": 4.0,
|
| 338 |
-
"eval_loss": 2.
|
| 339 |
-
"eval_runtime": 0.
|
| 340 |
-
"eval_samples_per_second":
|
| 341 |
-
"eval_steps_per_second": 2.
|
| 342 |
"step": 210
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 4.095238095238095,
|
| 346 |
-
"grad_norm": 0.
|
| 347 |
"learning_rate": 0.00014587688526312143,
|
| 348 |
-
"loss": 0.
|
| 349 |
"step": 215
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 4.190476190476191,
|
| 353 |
-
"grad_norm": 0.
|
| 354 |
"learning_rate": 0.00014286925614030542,
|
| 355 |
-
"loss": 0.
|
| 356 |
"step": 220
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 4.285714285714286,
|
| 360 |
-
"grad_norm": 0.
|
| 361 |
"learning_rate": 0.0001398133374601501,
|
| 362 |
-
"loss": 0.
|
| 363 |
"step": 225
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 4.380952380952381,
|
| 367 |
-
"grad_norm": 0.
|
| 368 |
"learning_rate": 0.00013671257152545277,
|
| 369 |
-
"loss": 0.
|
| 370 |
"step": 230
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 4.476190476190476,
|
| 374 |
-
"grad_norm": 0.
|
| 375 |
"learning_rate": 0.0001335704511566605,
|
| 376 |
-
"loss": 0.
|
| 377 |
"step": 235
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 4.571428571428571,
|
| 381 |
-
"grad_norm": 0.
|
| 382 |
"learning_rate": 0.0001303905157574247,
|
| 383 |
-
"loss": 0.
|
| 384 |
"step": 240
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 4.666666666666667,
|
| 388 |
-
"grad_norm": 0.
|
| 389 |
"learning_rate": 0.00012717634732768243,
|
| 390 |
-
"loss": 0.
|
| 391 |
"step": 245
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 4.761904761904762,
|
| 395 |
-
"grad_norm": 0.
|
| 396 |
"learning_rate": 0.0001239315664287558,
|
| 397 |
-
"loss": 0.
|
| 398 |
"step": 250
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 4.857142857142857,
|
| 402 |
-
"grad_norm": 0.
|
| 403 |
"learning_rate": 0.00012065982810501404,
|
| 404 |
-
"loss": 0.
|
| 405 |
"step": 255
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 4.9523809523809526,
|
| 409 |
-
"grad_norm": 0.
|
| 410 |
"learning_rate": 0.00011736481776669306,
|
| 411 |
-
"loss": 0.
|
| 412 |
"step": 260
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 4.9904761904761905,
|
| 416 |
-
"eval_loss": 2.
|
| 417 |
-
"eval_runtime": 0.
|
| 418 |
-
"eval_samples_per_second": 14.
|
| 419 |
-
"eval_steps_per_second": 1.
|
| 420 |
"step": 262
|
| 421 |
},
|
| 422 |
{
|
| 423 |
"epoch": 5.0476190476190474,
|
| 424 |
-
"grad_norm": 0.
|
| 425 |
"learning_rate": 0.00011405024703850929,
|
| 426 |
-
"loss": 0.
|
| 427 |
"step": 265
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 5.142857142857143,
|
| 431 |
-
"grad_norm": 0.
|
| 432 |
"learning_rate": 0.00011071984957874479,
|
| 433 |
-
"loss": 0.
|
| 434 |
"step": 270
|
| 435 |
},
|
| 436 |
{
|
| 437 |
"epoch": 5.238095238095238,
|
| 438 |
-
"grad_norm": 0.
|
| 439 |
"learning_rate": 0.00010737737687351284,
|
| 440 |
-
"loss": 0.
|
| 441 |
"step": 275
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 5.333333333333333,
|
| 445 |
-
"grad_norm": 0.
|
| 446 |
"learning_rate": 0.00010402659401094152,
|
| 447 |
-
"loss": 0.
|
| 448 |
"step": 280
|
| 449 |
},
|
| 450 |
{
|
| 451 |
"epoch": 5.428571428571429,
|
| 452 |
-
"grad_norm": 0.
|
| 453 |
"learning_rate": 0.00010067127544003563,
|
| 454 |
-
"loss": 0.
|
| 455 |
"step": 285
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 5.523809523809524,
|
| 459 |
-
"grad_norm": 0.
|
| 460 |
"learning_rate": 9.73152007189939e-05,
|
| 461 |
-
"loss": 0.
|
| 462 |
"step": 290
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 5.619047619047619,
|
| 466 |
-
"grad_norm": 0.
|
| 467 |
"learning_rate": 9.396215025777139e-05,
|
| 468 |
-
"loss": 0.
|
| 469 |
"step": 295
|
| 470 |
},
|
| 471 |
{
|
| 472 |
"epoch": 5.714285714285714,
|
| 473 |
-
"grad_norm": 0.
|
| 474 |
"learning_rate": 9.061590105968208e-05,
|
| 475 |
-
"loss": 0.
|
| 476 |
"step": 300
|
| 477 |
},
|
| 478 |
{
|
| 479 |
"epoch": 5.809523809523809,
|
| 480 |
-
"grad_norm": 0.
|
| 481 |
"learning_rate": 8.728022246683894e-05,
|
| 482 |
-
"loss": 0.
|
| 483 |
"step": 305
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 5.904761904761905,
|
| 487 |
-
"grad_norm": 0.
|
| 488 |
"learning_rate": 8.395887191422397e-05,
|
| 489 |
-
"loss": 0.
|
| 490 |
"step": 310
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 6.0,
|
| 494 |
-
"grad_norm": 0.
|
| 495 |
"learning_rate": 8.065559069717088e-05,
|
| 496 |
-
"loss": 0.
|
| 497 |
"step": 315
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 6.0,
|
| 501 |
-
"eval_loss": 2.
|
| 502 |
-
"eval_runtime": 0.
|
| 503 |
-
"eval_samples_per_second":
|
| 504 |
-
"eval_steps_per_second": 2.
|
| 505 |
"step": 315
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 6.095238095238095,
|
| 509 |
-
"grad_norm": 0.
|
| 510 |
"learning_rate": 7.73740997570278e-05,
|
| 511 |
-
"loss": 0.
|
| 512 |
"step": 320
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 6.190476190476191,
|
| 516 |
-
"grad_norm": 0.
|
| 517 |
"learning_rate": 7.411809548974792e-05,
|
| 518 |
-
"loss": 0.
|
| 519 |
"step": 325
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 6.285714285714286,
|
| 523 |
-
"grad_norm": 0.
|
| 524 |
"learning_rate": 7.089124558212871e-05,
|
| 525 |
-
"loss": 0.
|
| 526 |
"step": 330
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 6.380952380952381,
|
| 530 |
-
"grad_norm": 0.
|
| 531 |
"learning_rate": 6.769718488039023e-05,
|
| 532 |
-
"loss": 0.
|
| 533 |
"step": 335
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 6.476190476190476,
|
| 537 |
-
"grad_norm": 0.
|
| 538 |
"learning_rate": 6.453951129574644e-05,
|
| 539 |
-
"loss": 0.
|
| 540 |
"step": 340
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 6.571428571428571,
|
| 544 |
-
"grad_norm": 0.
|
| 545 |
"learning_rate": 6.142178175158149e-05,
|
| 546 |
-
"loss": 0.
|
| 547 |
"step": 345
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 6.666666666666667,
|
| 551 |
-
"grad_norm": 0.
|
| 552 |
"learning_rate": 5.834750817679606e-05,
|
| 553 |
-
"loss": 0.
|
| 554 |
"step": 350
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 6.761904761904762,
|
| 558 |
-
"grad_norm": 0.
|
| 559 |
"learning_rate": 5.5320153549837415e-05,
|
| 560 |
-
"loss": 0.
|
| 561 |
"step": 355
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 6.857142857142857,
|
| 565 |
-
"grad_norm": 0.
|
| 566 |
"learning_rate": 5.234312799786921e-05,
|
| 567 |
-
"loss": 0.
|
| 568 |
"step": 360
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 6.9523809523809526,
|
| 572 |
-
"grad_norm": 0.
|
| 573 |
"learning_rate": 4.9419784955474524e-05,
|
| 574 |
-
"loss": 0.
|
| 575 |
"step": 365
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 6.9904761904761905,
|
| 579 |
-
"eval_loss": 2.
|
| 580 |
-
"eval_runtime": 0.
|
| 581 |
-
"eval_samples_per_second":
|
| 582 |
-
"eval_steps_per_second": 1.
|
| 583 |
"step": 367
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 7.0476190476190474,
|
| 587 |
-
"grad_norm": 0.
|
| 588 |
"learning_rate": 4.6553417387219886e-05,
|
| 589 |
-
"loss": 0.
|
| 590 |
"step": 370
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 7.142857142857143,
|
| 594 |
-
"grad_norm": 0.
|
| 595 |
"learning_rate": 4.374725407833532e-05,
|
| 596 |
-
"loss": 0.
|
| 597 |
"step": 375
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 7.238095238095238,
|
| 601 |
-
"grad_norm": 0.
|
| 602 |
"learning_rate": 4.100445599768774e-05,
|
| 603 |
-
"loss": 0.
|
| 604 |
"step": 380
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 7.333333333333333,
|
| 608 |
-
"grad_norm": 0.
|
| 609 |
"learning_rate": 3.832811273714569e-05,
|
| 610 |
-
"loss": 0.
|
| 611 |
"step": 385
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 7.428571428571429,
|
| 615 |
-
"grad_norm": 0.
|
| 616 |
"learning_rate": 3.5721239031346066e-05,
|
| 617 |
-
"loss": 0.
|
| 618 |
"step": 390
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 7.523809523809524,
|
| 622 |
-
"grad_norm": 0.
|
| 623 |
"learning_rate": 3.318677136178228e-05,
|
| 624 |
-
"loss": 0.
|
| 625 |
"step": 395
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 7.619047619047619,
|
| 629 |
"grad_norm": 0.232421875,
|
| 630 |
"learning_rate": 3.072756464904006e-05,
|
| 631 |
-
"loss": 0.
|
| 632 |
"step": 400
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 7.714285714285714,
|
| 636 |
-
"grad_norm": 0.
|
| 637 |
"learning_rate": 2.8346389036906828e-05,
|
| 638 |
-
"loss": 0.
|
| 639 |
"step": 405
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 7.809523809523809,
|
| 643 |
-
"grad_norm": 0.
|
| 644 |
"learning_rate": 2.6045926771976303e-05,
|
| 645 |
-
"loss": 0.
|
| 646 |
"step": 410
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 7.904761904761905,
|
| 650 |
-
"grad_norm": 0.
|
| 651 |
"learning_rate": 2.382876918226409e-05,
|
| 652 |
-
"loss": 0.
|
| 653 |
"step": 415
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 8.0,
|
| 657 |
-
"grad_norm": 0.
|
| 658 |
"learning_rate": 2.1697413758237784e-05,
|
| 659 |
-
"loss": 0.
|
| 660 |
"step": 420
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 8.0,
|
| 664 |
-
"eval_loss": 2.
|
| 665 |
-
"eval_runtime": 0.
|
| 666 |
-
"eval_samples_per_second": 20.
|
| 667 |
-
"eval_steps_per_second": 2.
|
| 668 |
"step": 420
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 8.095238095238095,
|
| 672 |
-
"grad_norm": 0.
|
| 673 |
"learning_rate": 1.965426133954854e-05,
|
| 674 |
-
"loss": 0.
|
| 675 |
"step": 425
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 8.19047619047619,
|
| 679 |
-
"grad_norm": 0.
|
| 680 |
"learning_rate": 1.7701613410634365e-05,
|
| 681 |
-
"loss": 0.
|
| 682 |
"step": 430
|
| 683 |
},
|
| 684 |
{
|
| 685 |
"epoch": 8.285714285714286,
|
| 686 |
-
"grad_norm": 0.
|
| 687 |
"learning_rate": 1.584166950824061e-05,
|
| 688 |
-
"loss": 0.
|
| 689 |
"step": 435
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 8.380952380952381,
|
| 693 |
-
"grad_norm": 0.
|
| 694 |
"learning_rate": 1.4076524743778319e-05,
|
| 695 |
-
"loss": 0.
|
| 696 |
"step": 440
|
| 697 |
},
|
| 698 |
{
|
| 699 |
"epoch": 8.476190476190476,
|
| 700 |
-
"grad_norm": 0.
|
| 701 |
"learning_rate": 1.2408167443311214e-05,
|
| 702 |
-
"loss": 0.
|
| 703 |
"step": 445
|
| 704 |
},
|
| 705 |
{
|
| 706 |
"epoch": 8.571428571428571,
|
| 707 |
-
"grad_norm": 0.
|
| 708 |
"learning_rate": 1.083847690782972e-05,
|
| 709 |
-
"loss": 0.
|
| 710 |
"step": 450
|
| 711 |
},
|
| 712 |
{
|
| 713 |
"epoch": 8.666666666666666,
|
| 714 |
-
"grad_norm": 0.
|
| 715 |
"learning_rate": 9.369221296335006e-06,
|
| 716 |
-
"loss": 0.
|
| 717 |
"step": 455
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 8.761904761904763,
|
| 721 |
-
"grad_norm": 0.
|
| 722 |
"learning_rate": 8.002055634117578e-06,
|
| 723 |
-
"loss": 0.
|
| 724 |
"step": 460
|
| 725 |
},
|
| 726 |
{
|
| 727 |
"epoch": 8.857142857142858,
|
| 728 |
-
"grad_norm": 0.
|
| 729 |
"learning_rate": 6.738519948473976e-06,
|
| 730 |
-
"loss": 0.
|
| 731 |
"step": 465
|
| 732 |
},
|
| 733 |
{
|
| 734 |
"epoch": 8.952380952380953,
|
| 735 |
-
"grad_norm": 0.
|
| 736 |
"learning_rate": 5.580037533961546e-06,
|
| 737 |
-
"loss": 0.
|
| 738 |
"step": 470
|
| 739 |
},
|
| 740 |
{
|
| 741 |
"epoch": 8.99047619047619,
|
| 742 |
-
"eval_loss": 2.
|
| 743 |
-
"eval_runtime": 0.
|
| 744 |
-
"eval_samples_per_second":
|
| 745 |
-
"eval_steps_per_second":
|
| 746 |
"step": 472
|
| 747 |
},
|
| 748 |
{
|
| 749 |
"epoch": 9.047619047619047,
|
| 750 |
-
"grad_norm": 0.
|
| 751 |
"learning_rate": 4.527913349145441e-06,
|
| 752 |
-
"loss": 0.
|
| 753 |
"step": 475
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 9.142857142857142,
|
| 757 |
-
"grad_norm": 0.
|
| 758 |
"learning_rate": 3.5833325466437694e-06,
|
| 759 |
-
"loss": 0.
|
| 760 |
"step": 480
|
| 761 |
},
|
| 762 |
{
|
| 763 |
"epoch": 9.238095238095237,
|
| 764 |
-
"grad_norm": 0.
|
| 765 |
"learning_rate": 2.7473591381266708e-06,
|
| 766 |
-
"loss": 0.
|
| 767 |
"step": 485
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 9.333333333333334,
|
| 771 |
-
"grad_norm": 0.
|
| 772 |
"learning_rate": 2.0209347957732328e-06,
|
| 773 |
-
"loss": 0.
|
| 774 |
"step": 490
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 9.428571428571429,
|
| 778 |
-
"grad_norm": 0.
|
| 779 |
"learning_rate": 1.404877791536141e-06,
|
| 780 |
-
"loss": 0.
|
| 781 |
"step": 495
|
| 782 |
},
|
| 783 |
{
|
| 784 |
"epoch": 9.523809523809524,
|
| 785 |
-
"grad_norm": 0.
|
| 786 |
"learning_rate": 8.998820754091531e-07,
|
| 787 |
-
"loss": 0.
|
| 788 |
"step": 500
|
| 789 |
},
|
| 790 |
{
|
| 791 |
"epoch": 9.619047619047619,
|
| 792 |
-
"grad_norm": 0.
|
| 793 |
"learning_rate": 5.065164937354428e-07,
|
| 794 |
-
"loss": 0.
|
| 795 |
"step": 505
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 9.714285714285714,
|
| 799 |
-
"grad_norm": 0.
|
| 800 |
"learning_rate": 2.2522414843748618e-07,
|
| 801 |
-
"loss": 0.
|
| 802 |
"step": 510
|
| 803 |
},
|
| 804 |
{
|
| 805 |
"epoch": 9.80952380952381,
|
| 806 |
-
"grad_norm": 0.
|
| 807 |
"learning_rate": 5.632189789027687e-08,
|
| 808 |
-
"loss": 0.
|
| 809 |
"step": 515
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 9.904761904761905,
|
| 813 |
-
"grad_norm": 0.
|
| 814 |
"learning_rate": 0.0,
|
| 815 |
-
"loss": 0.
|
| 816 |
"step": 520
|
| 817 |
},
|
| 818 |
{
|
| 819 |
"epoch": 9.904761904761905,
|
| 820 |
-
"eval_loss": 2.
|
| 821 |
-
"eval_runtime": 0.
|
| 822 |
-
"eval_samples_per_second": 20.
|
| 823 |
-
"eval_steps_per_second": 2.
|
| 824 |
"step": 520
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 9.904761904761905,
|
| 828 |
"step": 520,
|
| 829 |
-
"total_flos": 8.
|
| 830 |
-
"train_loss": 1.
|
| 831 |
-
"train_runtime":
|
| 832 |
-
"train_samples_per_second":
|
| 833 |
-
"train_steps_per_second": 0.
|
| 834 |
}
|
| 835 |
],
|
| 836 |
"logging_steps": 5,
|
|
@@ -850,7 +850,7 @@
|
|
| 850 |
"attributes": {}
|
| 851 |
}
|
| 852 |
},
|
| 853 |
-
"total_flos": 8.
|
| 854 |
"train_batch_size": 8,
|
| 855 |
"trial_name": null,
|
| 856 |
"trial_params": null
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.01904761904761905,
|
| 13 |
+
"grad_norm": 4.625,
|
| 14 |
"learning_rate": 3.846153846153847e-06,
|
| 15 |
"loss": 3.0068,
|
| 16 |
"step": 1
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.09523809523809523,
|
| 20 |
+
"grad_norm": 7.28125,
|
| 21 |
"learning_rate": 1.923076923076923e-05,
|
| 22 |
+
"loss": 3.002,
|
| 23 |
"step": 5
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.19047619047619047,
|
| 27 |
+
"grad_norm": 2.9375,
|
| 28 |
"learning_rate": 3.846153846153846e-05,
|
| 29 |
+
"loss": 2.8237,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2857142857142857,
|
| 34 |
+
"grad_norm": 2.1875,
|
| 35 |
"learning_rate": 5.769230769230769e-05,
|
| 36 |
+
"loss": 2.4529,
|
| 37 |
"step": 15
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.38095238095238093,
|
| 41 |
+
"grad_norm": 3.0625,
|
| 42 |
"learning_rate": 7.692307692307693e-05,
|
| 43 |
+
"loss": 2.1685,
|
| 44 |
"step": 20
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.47619047619047616,
|
| 48 |
+
"grad_norm": 2.171875,
|
| 49 |
"learning_rate": 9.615384615384617e-05,
|
| 50 |
+
"loss": 1.9091,
|
| 51 |
"step": 25
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.5714285714285714,
|
| 55 |
+
"grad_norm": 0.73046875,
|
| 56 |
"learning_rate": 0.00011538461538461538,
|
| 57 |
+
"loss": 1.6688,
|
| 58 |
"step": 30
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.6666666666666666,
|
| 62 |
+
"grad_norm": 0.4921875,
|
| 63 |
"learning_rate": 0.00013461538461538464,
|
| 64 |
+
"loss": 1.4855,
|
| 65 |
"step": 35
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.7619047619047619,
|
| 69 |
+
"grad_norm": 0.59375,
|
| 70 |
"learning_rate": 0.00015384615384615385,
|
| 71 |
+
"loss": 1.3575,
|
| 72 |
"step": 40
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.8571428571428571,
|
| 76 |
+
"grad_norm": 0.734375,
|
| 77 |
"learning_rate": 0.0001730769230769231,
|
| 78 |
+
"loss": 1.2763,
|
| 79 |
"step": 45
|
| 80 |
},
|
| 81 |
{
|
| 82 |
"epoch": 0.9523809523809523,
|
| 83 |
+
"grad_norm": 1.1796875,
|
| 84 |
"learning_rate": 0.00019230769230769233,
|
| 85 |
+
"loss": 1.2288,
|
| 86 |
"step": 50
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.9904761904761905,
|
| 90 |
+
"eval_loss": 2.5090928077697754,
|
| 91 |
+
"eval_runtime": 0.6698,
|
| 92 |
+
"eval_samples_per_second": 14.929,
|
| 93 |
+
"eval_steps_per_second": 1.493,
|
| 94 |
"step": 52
|
| 95 |
},
|
| 96 |
{
|
| 97 |
"epoch": 1.0476190476190477,
|
| 98 |
+
"grad_norm": 0.3203125,
|
| 99 |
"learning_rate": 0.00019997972289848503,
|
| 100 |
+
"loss": 1.1945,
|
| 101 |
"step": 55
|
| 102 |
},
|
| 103 |
{
|
| 104 |
"epoch": 1.1428571428571428,
|
| 105 |
+
"grad_norm": 0.45703125,
|
| 106 |
"learning_rate": 0.00019985583705641418,
|
| 107 |
+
"loss": 1.1593,
|
| 108 |
"step": 60
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 1.2380952380952381,
|
| 112 |
+
"grad_norm": 0.76953125,
|
| 113 |
"learning_rate": 0.00019961946980917456,
|
| 114 |
+
"loss": 1.1262,
|
| 115 |
"step": 65
|
| 116 |
},
|
| 117 |
{
|
| 118 |
"epoch": 1.3333333333333333,
|
| 119 |
+
"grad_norm": 0.5703125,
|
| 120 |
"learning_rate": 0.0001992708874098054,
|
| 121 |
+
"loss": 1.1179,
|
| 122 |
"step": 70
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 1.4285714285714286,
|
| 126 |
+
"grad_norm": 0.357421875,
|
| 127 |
"learning_rate": 0.0001988104825147528,
|
| 128 |
+
"loss": 1.1067,
|
| 129 |
"step": 75
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 1.5238095238095237,
|
| 133 |
+
"grad_norm": 0.8984375,
|
| 134 |
"learning_rate": 0.00019823877374156647,
|
| 135 |
+
"loss": 1.0937,
|
| 136 |
"step": 80
|
| 137 |
},
|
| 138 |
{
|
| 139 |
"epoch": 1.619047619047619,
|
| 140 |
+
"grad_norm": 0.8046875,
|
| 141 |
"learning_rate": 0.00019755640508470942,
|
| 142 |
+
"loss": 1.0861,
|
| 143 |
"step": 85
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 1.7142857142857144,
|
| 147 |
+
"grad_norm": 0.400390625,
|
| 148 |
"learning_rate": 0.00019676414519013781,
|
| 149 |
+
"loss": 1.0766,
|
| 150 |
"step": 90
|
| 151 |
},
|
| 152 |
{
|
| 153 |
"epoch": 1.8095238095238095,
|
| 154 |
+
"grad_norm": 0.34765625,
|
| 155 |
"learning_rate": 0.00019586288648946947,
|
| 156 |
+
"loss": 1.0619,
|
| 157 |
"step": 95
|
| 158 |
},
|
| 159 |
{
|
| 160 |
"epoch": 1.9047619047619047,
|
| 161 |
+
"grad_norm": 0.314453125,
|
| 162 |
"learning_rate": 0.00019485364419471454,
|
| 163 |
+
"loss": 1.0651,
|
| 164 |
"step": 100
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 2.0,
|
| 168 |
+
"grad_norm": 0.2578125,
|
| 169 |
"learning_rate": 0.00019373755515470254,
|
| 170 |
+
"loss": 1.0492,
|
| 171 |
"step": 105
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 2.0,
|
| 175 |
+
"eval_loss": 2.4524765014648438,
|
| 176 |
+
"eval_runtime": 0.4793,
|
| 177 |
+
"eval_samples_per_second": 20.866,
|
| 178 |
+
"eval_steps_per_second": 2.087,
|
| 179 |
"step": 105
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 2.0952380952380953,
|
| 183 |
+
"grad_norm": 0.466796875,
|
| 184 |
"learning_rate": 0.00019251587657449236,
|
| 185 |
+
"loss": 1.0305,
|
| 186 |
"step": 110
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 2.1904761904761907,
|
| 190 |
+
"grad_norm": 0.333984375,
|
| 191 |
"learning_rate": 0.00019118998459920902,
|
| 192 |
+
"loss": 1.0189,
|
| 193 |
"step": 115
|
| 194 |
},
|
| 195 |
{
|
| 196 |
"epoch": 2.2857142857142856,
|
| 197 |
+
"grad_norm": 0.3515625,
|
| 198 |
"learning_rate": 0.0001897613727639014,
|
| 199 |
+
"loss": 1.0175,
|
| 200 |
"step": 120
|
| 201 |
},
|
| 202 |
{
|
| 203 |
"epoch": 2.380952380952381,
|
| 204 |
+
"grad_norm": 0.267578125,
|
| 205 |
"learning_rate": 0.0001882316503111678,
|
| 206 |
+
"loss": 1.015,
|
| 207 |
"step": 125
|
| 208 |
},
|
| 209 |
{
|
| 210 |
"epoch": 2.4761904761904763,
|
| 211 |
+
"grad_norm": 0.2314453125,
|
| 212 |
"learning_rate": 0.00018660254037844388,
|
| 213 |
+
"loss": 1.0152,
|
| 214 |
"step": 130
|
| 215 |
},
|
| 216 |
{
|
| 217 |
"epoch": 2.571428571428571,
|
| 218 |
+
"grad_norm": 0.3203125,
|
| 219 |
"learning_rate": 0.00018487587805699526,
|
| 220 |
+
"loss": 1.0047,
|
| 221 |
"step": 135
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 2.6666666666666665,
|
| 225 |
+
"grad_norm": 0.28125,
|
| 226 |
"learning_rate": 0.00018305360832480117,
|
| 227 |
+
"loss": 1.0046,
|
| 228 |
"step": 140
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 2.761904761904762,
|
| 232 |
+
"grad_norm": 0.267578125,
|
| 233 |
"learning_rate": 0.00018113778385565733,
|
| 234 |
+
"loss": 1.0092,
|
| 235 |
"step": 145
|
| 236 |
},
|
| 237 |
{
|
| 238 |
"epoch": 2.857142857142857,
|
| 239 |
+
"grad_norm": 0.306640625,
|
| 240 |
"learning_rate": 0.0001791305627069662,
|
| 241 |
+
"loss": 0.9981,
|
| 242 |
"step": 150
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 2.9523809523809526,
|
| 246 |
+
"grad_norm": 0.328125,
|
| 247 |
"learning_rate": 0.00017703420588881946,
|
| 248 |
+
"loss": 1.0025,
|
| 249 |
"step": 155
|
| 250 |
},
|
| 251 |
{
|
| 252 |
"epoch": 2.9904761904761905,
|
| 253 |
+
"eval_loss": 2.4599478244781494,
|
| 254 |
+
"eval_runtime": 0.4943,
|
| 255 |
+
"eval_samples_per_second": 20.231,
|
| 256 |
+
"eval_steps_per_second": 2.023,
|
| 257 |
"step": 157
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 3.0476190476190474,
|
| 261 |
+
"grad_norm": 0.310546875,
|
| 262 |
"learning_rate": 0.00017485107481711012,
|
| 263 |
+
"loss": 0.9874,
|
| 264 |
"step": 160
|
| 265 |
},
|
| 266 |
{
|
| 267 |
"epoch": 3.142857142857143,
|
| 268 |
+
"grad_norm": 0.3046875,
|
| 269 |
"learning_rate": 0.00017258362865354426,
|
| 270 |
+
"loss": 0.9747,
|
| 271 |
"step": 165
|
| 272 |
},
|
| 273 |
{
|
| 274 |
"epoch": 3.238095238095238,
|
| 275 |
+
"grad_norm": 0.2353515625,
|
| 276 |
"learning_rate": 0.00017023442153554777,
|
| 277 |
+
"loss": 0.9763,
|
| 278 |
"step": 170
|
| 279 |
},
|
| 280 |
{
|
| 281 |
"epoch": 3.3333333333333335,
|
| 282 |
+
"grad_norm": 0.318359375,
|
| 283 |
"learning_rate": 0.0001678060996991891,
|
| 284 |
+
"loss": 0.971,
|
| 285 |
"step": 175
|
| 286 |
},
|
| 287 |
{
|
| 288 |
"epoch": 3.4285714285714284,
|
| 289 |
+
"grad_norm": 0.47265625,
|
| 290 |
"learning_rate": 0.0001653013984983585,
|
| 291 |
+
"loss": 0.9666,
|
| 292 |
"step": 180
|
| 293 |
},
|
| 294 |
{
|
| 295 |
"epoch": 3.5238095238095237,
|
| 296 |
+
"grad_norm": 0.703125,
|
| 297 |
"learning_rate": 0.00016272313932356162,
|
| 298 |
+
"loss": 0.9646,
|
| 299 |
"step": 185
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 3.619047619047619,
|
| 303 |
+
"grad_norm": 0.796875,
|
| 304 |
"learning_rate": 0.0001600742264237979,
|
| 305 |
+
"loss": 0.9712,
|
| 306 |
"step": 190
|
| 307 |
},
|
| 308 |
{
|
| 309 |
"epoch": 3.7142857142857144,
|
| 310 |
+
"grad_norm": 0.2578125,
|
| 311 |
"learning_rate": 0.0001573576436351046,
|
| 312 |
+
"loss": 0.969,
|
| 313 |
"step": 195
|
| 314 |
},
|
| 315 |
{
|
| 316 |
"epoch": 3.8095238095238093,
|
| 317 |
+
"grad_norm": 0.33984375,
|
| 318 |
"learning_rate": 0.00015457645101945046,
|
| 319 |
+
"loss": 0.9594,
|
| 320 |
"step": 200
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 3.9047619047619047,
|
| 324 |
+
"grad_norm": 0.28515625,
|
| 325 |
"learning_rate": 0.00015173378141776568,
|
| 326 |
+
"loss": 0.9633,
|
| 327 |
"step": 205
|
| 328 |
},
|
| 329 |
{
|
| 330 |
"epoch": 4.0,
|
| 331 |
+
"grad_norm": 0.28515625,
|
| 332 |
"learning_rate": 0.00014883283692099112,
|
| 333 |
+
"loss": 0.9597,
|
| 334 |
"step": 210
|
| 335 |
},
|
| 336 |
{
|
| 337 |
"epoch": 4.0,
|
| 338 |
+
"eval_loss": 2.471402168273926,
|
| 339 |
+
"eval_runtime": 0.4759,
|
| 340 |
+
"eval_samples_per_second": 21.012,
|
| 341 |
+
"eval_steps_per_second": 2.101,
|
| 342 |
"step": 210
|
| 343 |
},
|
| 344 |
{
|
| 345 |
"epoch": 4.095238095238095,
|
| 346 |
+
"grad_norm": 0.31640625,
|
| 347 |
"learning_rate": 0.00014587688526312143,
|
| 348 |
+
"loss": 0.9325,
|
| 349 |
"step": 215
|
| 350 |
},
|
| 351 |
{
|
| 352 |
"epoch": 4.190476190476191,
|
| 353 |
+
"grad_norm": 0.287109375,
|
| 354 |
"learning_rate": 0.00014286925614030542,
|
| 355 |
+
"loss": 0.933,
|
| 356 |
"step": 220
|
| 357 |
},
|
| 358 |
{
|
| 359 |
"epoch": 4.285714285714286,
|
| 360 |
+
"grad_norm": 0.2421875,
|
| 361 |
"learning_rate": 0.0001398133374601501,
|
| 362 |
+
"loss": 0.9404,
|
| 363 |
"step": 225
|
| 364 |
},
|
| 365 |
{
|
| 366 |
"epoch": 4.380952380952381,
|
| 367 |
+
"grad_norm": 0.25390625,
|
| 368 |
"learning_rate": 0.00013671257152545277,
|
| 369 |
+
"loss": 0.9368,
|
| 370 |
"step": 230
|
| 371 |
},
|
| 372 |
{
|
| 373 |
"epoch": 4.476190476190476,
|
| 374 |
+
"grad_norm": 0.2314453125,
|
| 375 |
"learning_rate": 0.0001335704511566605,
|
| 376 |
+
"loss": 0.934,
|
| 377 |
"step": 235
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 4.571428571428571,
|
| 381 |
+
"grad_norm": 0.466796875,
|
| 382 |
"learning_rate": 0.0001303905157574247,
|
| 383 |
+
"loss": 0.9391,
|
| 384 |
"step": 240
|
| 385 |
},
|
| 386 |
{
|
| 387 |
"epoch": 4.666666666666667,
|
| 388 |
+
"grad_norm": 0.263671875,
|
| 389 |
"learning_rate": 0.00012717634732768243,
|
| 390 |
+
"loss": 0.9411,
|
| 391 |
"step": 245
|
| 392 |
},
|
| 393 |
{
|
| 394 |
"epoch": 4.761904761904762,
|
| 395 |
+
"grad_norm": 0.263671875,
|
| 396 |
"learning_rate": 0.0001239315664287558,
|
| 397 |
+
"loss": 0.9428,
|
| 398 |
"step": 250
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 4.857142857142857,
|
| 402 |
+
"grad_norm": 0.291015625,
|
| 403 |
"learning_rate": 0.00012065982810501404,
|
| 404 |
+
"loss": 0.9332,
|
| 405 |
"step": 255
|
| 406 |
},
|
| 407 |
{
|
| 408 |
"epoch": 4.9523809523809526,
|
| 409 |
+
"grad_norm": 0.24609375,
|
| 410 |
"learning_rate": 0.00011736481776669306,
|
| 411 |
+
"loss": 0.9394,
|
| 412 |
"step": 260
|
| 413 |
},
|
| 414 |
{
|
| 415 |
"epoch": 4.9904761904761905,
|
| 416 |
+
"eval_loss": 2.486337423324585,
|
| 417 |
+
"eval_runtime": 0.6707,
|
| 418 |
+
"eval_samples_per_second": 14.909,
|
| 419 |
+
"eval_steps_per_second": 1.491,
|
| 420 |
"step": 262
|
| 421 |
},
|
| 422 |
{
|
| 423 |
"epoch": 5.0476190476190474,
|
| 424 |
+
"grad_norm": 0.48828125,
|
| 425 |
"learning_rate": 0.00011405024703850929,
|
| 426 |
+
"loss": 0.9283,
|
| 427 |
"step": 265
|
| 428 |
},
|
| 429 |
{
|
| 430 |
"epoch": 5.142857142857143,
|
| 431 |
+
"grad_norm": 0.37890625,
|
| 432 |
"learning_rate": 0.00011071984957874479,
|
| 433 |
+
"loss": 0.9202,
|
| 434 |
"step": 270
|
| 435 |
},
|
| 436 |
{
|
| 437 |
"epoch": 5.238095238095238,
|
| 438 |
+
"grad_norm": 0.25,
|
| 439 |
"learning_rate": 0.00010737737687351284,
|
| 440 |
+
"loss": 0.916,
|
| 441 |
"step": 275
|
| 442 |
},
|
| 443 |
{
|
| 444 |
"epoch": 5.333333333333333,
|
| 445 |
+
"grad_norm": 0.25390625,
|
| 446 |
"learning_rate": 0.00010402659401094152,
|
| 447 |
+
"loss": 0.9109,
|
| 448 |
"step": 280
|
| 449 |
},
|
| 450 |
{
|
| 451 |
"epoch": 5.428571428571429,
|
| 452 |
+
"grad_norm": 0.2734375,
|
| 453 |
"learning_rate": 0.00010067127544003563,
|
| 454 |
+
"loss": 0.9156,
|
| 455 |
"step": 285
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 5.523809523809524,
|
| 459 |
+
"grad_norm": 0.259765625,
|
| 460 |
"learning_rate": 9.73152007189939e-05,
|
| 461 |
+
"loss": 0.9139,
|
| 462 |
"step": 290
|
| 463 |
},
|
| 464 |
{
|
| 465 |
"epoch": 5.619047619047619,
|
| 466 |
+
"grad_norm": 0.234375,
|
| 467 |
"learning_rate": 9.396215025777139e-05,
|
| 468 |
+
"loss": 0.9166,
|
| 469 |
"step": 295
|
| 470 |
},
|
| 471 |
{
|
| 472 |
"epoch": 5.714285714285714,
|
| 473 |
+
"grad_norm": 0.2314453125,
|
| 474 |
"learning_rate": 9.061590105968208e-05,
|
| 475 |
+
"loss": 0.9162,
|
| 476 |
"step": 300
|
| 477 |
},
|
| 478 |
{
|
| 479 |
"epoch": 5.809523809523809,
|
| 480 |
+
"grad_norm": 0.328125,
|
| 481 |
"learning_rate": 8.728022246683894e-05,
|
| 482 |
+
"loss": 0.9079,
|
| 483 |
"step": 305
|
| 484 |
},
|
| 485 |
{
|
| 486 |
"epoch": 5.904761904761905,
|
| 487 |
+
"grad_norm": 0.263671875,
|
| 488 |
"learning_rate": 8.395887191422397e-05,
|
| 489 |
+
"loss": 0.9135,
|
| 490 |
"step": 310
|
| 491 |
},
|
| 492 |
{
|
| 493 |
"epoch": 6.0,
|
| 494 |
+
"grad_norm": 0.2578125,
|
| 495 |
"learning_rate": 8.065559069717088e-05,
|
| 496 |
+
"loss": 0.912,
|
| 497 |
"step": 315
|
| 498 |
},
|
| 499 |
{
|
| 500 |
"epoch": 6.0,
|
| 501 |
+
"eval_loss": 2.5085232257843018,
|
| 502 |
+
"eval_runtime": 0.4762,
|
| 503 |
+
"eval_samples_per_second": 21.0,
|
| 504 |
+
"eval_steps_per_second": 2.1,
|
| 505 |
"step": 315
|
| 506 |
},
|
| 507 |
{
|
| 508 |
"epoch": 6.095238095238095,
|
| 509 |
+
"grad_norm": 0.2236328125,
|
| 510 |
"learning_rate": 7.73740997570278e-05,
|
| 511 |
+
"loss": 0.8989,
|
| 512 |
"step": 320
|
| 513 |
},
|
| 514 |
{
|
| 515 |
"epoch": 6.190476190476191,
|
| 516 |
+
"grad_norm": 0.2578125,
|
| 517 |
"learning_rate": 7.411809548974792e-05,
|
| 518 |
+
"loss": 0.897,
|
| 519 |
"step": 325
|
| 520 |
},
|
| 521 |
{
|
| 522 |
"epoch": 6.285714285714286,
|
| 523 |
+
"grad_norm": 0.31640625,
|
| 524 |
"learning_rate": 7.089124558212871e-05,
|
| 525 |
+
"loss": 0.9014,
|
| 526 |
"step": 330
|
| 527 |
},
|
| 528 |
{
|
| 529 |
"epoch": 6.380952380952381,
|
| 530 |
+
"grad_norm": 0.353515625,
|
| 531 |
"learning_rate": 6.769718488039023e-05,
|
| 532 |
+
"loss": 0.8964,
|
| 533 |
"step": 335
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 6.476190476190476,
|
| 537 |
+
"grad_norm": 0.267578125,
|
| 538 |
"learning_rate": 6.453951129574644e-05,
|
| 539 |
+
"loss": 0.894,
|
| 540 |
"step": 340
|
| 541 |
},
|
| 542 |
{
|
| 543 |
"epoch": 6.571428571428571,
|
| 544 |
+
"grad_norm": 0.25390625,
|
| 545 |
"learning_rate": 6.142178175158149e-05,
|
| 546 |
+
"loss": 0.9002,
|
| 547 |
"step": 345
|
| 548 |
},
|
| 549 |
{
|
| 550 |
"epoch": 6.666666666666667,
|
| 551 |
+
"grad_norm": 0.2294921875,
|
| 552 |
"learning_rate": 5.834750817679606e-05,
|
| 553 |
+
"loss": 0.8989,
|
| 554 |
"step": 350
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 6.761904761904762,
|
| 558 |
+
"grad_norm": 0.2265625,
|
| 559 |
"learning_rate": 5.5320153549837415e-05,
|
| 560 |
+
"loss": 0.8996,
|
| 561 |
"step": 355
|
| 562 |
},
|
| 563 |
{
|
| 564 |
"epoch": 6.857142857142857,
|
| 565 |
+
"grad_norm": 0.23046875,
|
| 566 |
"learning_rate": 5.234312799786921e-05,
|
| 567 |
+
"loss": 0.9,
|
| 568 |
"step": 360
|
| 569 |
},
|
| 570 |
{
|
| 571 |
"epoch": 6.9523809523809526,
|
| 572 |
+
"grad_norm": 0.2412109375,
|
| 573 |
"learning_rate": 4.9419784955474524e-05,
|
| 574 |
+
"loss": 0.8901,
|
| 575 |
"step": 365
|
| 576 |
},
|
| 577 |
{
|
| 578 |
"epoch": 6.9904761904761905,
|
| 579 |
+
"eval_loss": 2.517164707183838,
|
| 580 |
+
"eval_runtime": 0.6714,
|
| 581 |
+
"eval_samples_per_second": 14.895,
|
| 582 |
+
"eval_steps_per_second": 1.489,
|
| 583 |
"step": 367
|
| 584 |
},
|
| 585 |
{
|
| 586 |
"epoch": 7.0476190476190474,
|
| 587 |
+
"grad_norm": 0.21484375,
|
| 588 |
"learning_rate": 4.6553417387219886e-05,
|
| 589 |
+
"loss": 0.8844,
|
| 590 |
"step": 370
|
| 591 |
},
|
| 592 |
{
|
| 593 |
"epoch": 7.142857142857143,
|
| 594 |
+
"grad_norm": 0.2216796875,
|
| 595 |
"learning_rate": 4.374725407833532e-05,
|
| 596 |
+
"loss": 0.878,
|
| 597 |
"step": 375
|
| 598 |
},
|
| 599 |
{
|
| 600 |
"epoch": 7.238095238095238,
|
| 601 |
+
"grad_norm": 0.2294921875,
|
| 602 |
"learning_rate": 4.100445599768774e-05,
|
| 603 |
+
"loss": 0.8811,
|
| 604 |
"step": 380
|
| 605 |
},
|
| 606 |
{
|
| 607 |
"epoch": 7.333333333333333,
|
| 608 |
+
"grad_norm": 0.212890625,
|
| 609 |
"learning_rate": 3.832811273714569e-05,
|
| 610 |
+
"loss": 0.8853,
|
| 611 |
"step": 385
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 7.428571428571429,
|
| 615 |
+
"grad_norm": 0.34375,
|
| 616 |
"learning_rate": 3.5721239031346066e-05,
|
| 617 |
+
"loss": 0.8852,
|
| 618 |
"step": 390
|
| 619 |
},
|
| 620 |
{
|
| 621 |
"epoch": 7.523809523809524,
|
| 622 |
+
"grad_norm": 0.224609375,
|
| 623 |
"learning_rate": 3.318677136178228e-05,
|
| 624 |
+
"loss": 0.8829,
|
| 625 |
"step": 395
|
| 626 |
},
|
| 627 |
{
|
| 628 |
"epoch": 7.619047619047619,
|
| 629 |
"grad_norm": 0.232421875,
|
| 630 |
"learning_rate": 3.072756464904006e-05,
|
| 631 |
+
"loss": 0.8899,
|
| 632 |
"step": 400
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 7.714285714285714,
|
| 636 |
+
"grad_norm": 0.302734375,
|
| 637 |
"learning_rate": 2.8346389036906828e-05,
|
| 638 |
+
"loss": 0.888,
|
| 639 |
"step": 405
|
| 640 |
},
|
| 641 |
{
|
| 642 |
"epoch": 7.809523809523809,
|
| 643 |
+
"grad_norm": 0.21875,
|
| 644 |
"learning_rate": 2.6045926771976303e-05,
|
| 645 |
+
"loss": 0.8876,
|
| 646 |
"step": 410
|
| 647 |
},
|
| 648 |
{
|
| 649 |
"epoch": 7.904761904761905,
|
| 650 |
+
"grad_norm": 0.21875,
|
| 651 |
"learning_rate": 2.382876918226409e-05,
|
| 652 |
+
"loss": 0.8924,
|
| 653 |
"step": 415
|
| 654 |
},
|
| 655 |
{
|
| 656 |
"epoch": 8.0,
|
| 657 |
+
"grad_norm": 0.2138671875,
|
| 658 |
"learning_rate": 2.1697413758237784e-05,
|
| 659 |
+
"loss": 0.8887,
|
| 660 |
"step": 420
|
| 661 |
},
|
| 662 |
{
|
| 663 |
"epoch": 8.0,
|
| 664 |
+
"eval_loss": 2.529445171356201,
|
| 665 |
+
"eval_runtime": 0.4793,
|
| 666 |
+
"eval_samples_per_second": 20.866,
|
| 667 |
+
"eval_steps_per_second": 2.087,
|
| 668 |
"step": 420
|
| 669 |
},
|
| 670 |
{
|
| 671 |
"epoch": 8.095238095238095,
|
| 672 |
+
"grad_norm": 0.26171875,
|
| 673 |
"learning_rate": 1.965426133954854e-05,
|
| 674 |
+
"loss": 0.8761,
|
| 675 |
"step": 425
|
| 676 |
},
|
| 677 |
{
|
| 678 |
"epoch": 8.19047619047619,
|
| 679 |
+
"grad_norm": 0.2177734375,
|
| 680 |
"learning_rate": 1.7701613410634365e-05,
|
| 681 |
+
"loss": 0.8847,
|
| 682 |
"step": 430
|
| 683 |
},
|
| 684 |
{
|
| 685 |
"epoch": 8.285714285714286,
|
| 686 |
+
"grad_norm": 0.2197265625,
|
| 687 |
"learning_rate": 1.584166950824061e-05,
|
| 688 |
+
"loss": 0.8831,
|
| 689 |
"step": 435
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 8.380952380952381,
|
| 693 |
+
"grad_norm": 0.2080078125,
|
| 694 |
"learning_rate": 1.4076524743778319e-05,
|
| 695 |
+
"loss": 0.8769,
|
| 696 |
"step": 440
|
| 697 |
},
|
| 698 |
{
|
| 699 |
"epoch": 8.476190476190476,
|
| 700 |
+
"grad_norm": 0.205078125,
|
| 701 |
"learning_rate": 1.2408167443311214e-05,
|
| 702 |
+
"loss": 0.8773,
|
| 703 |
"step": 445
|
| 704 |
},
|
| 705 |
{
|
| 706 |
"epoch": 8.571428571428571,
|
| 707 |
+
"grad_norm": 0.21875,
|
| 708 |
"learning_rate": 1.083847690782972e-05,
|
| 709 |
+
"loss": 0.8772,
|
| 710 |
"step": 450
|
| 711 |
},
|
| 712 |
{
|
| 713 |
"epoch": 8.666666666666666,
|
| 714 |
+
"grad_norm": 0.236328125,
|
| 715 |
"learning_rate": 9.369221296335006e-06,
|
| 716 |
+
"loss": 0.8846,
|
| 717 |
"step": 455
|
| 718 |
},
|
| 719 |
{
|
| 720 |
"epoch": 8.761904761904763,
|
| 721 |
+
"grad_norm": 0.216796875,
|
| 722 |
"learning_rate": 8.002055634117578e-06,
|
| 723 |
+
"loss": 0.8759,
|
| 724 |
"step": 460
|
| 725 |
},
|
| 726 |
{
|
| 727 |
"epoch": 8.857142857142858,
|
| 728 |
+
"grad_norm": 0.203125,
|
| 729 |
"learning_rate": 6.738519948473976e-06,
|
| 730 |
+
"loss": 0.8798,
|
| 731 |
"step": 465
|
| 732 |
},
|
| 733 |
{
|
| 734 |
"epoch": 8.952380952380953,
|
| 735 |
+
"grad_norm": 0.20703125,
|
| 736 |
"learning_rate": 5.580037533961546e-06,
|
| 737 |
+
"loss": 0.8816,
|
| 738 |
"step": 470
|
| 739 |
},
|
| 740 |
{
|
| 741 |
"epoch": 8.99047619047619,
|
| 742 |
+
"eval_loss": 2.531066417694092,
|
| 743 |
+
"eval_runtime": 0.4939,
|
| 744 |
+
"eval_samples_per_second": 20.246,
|
| 745 |
+
"eval_steps_per_second": 2.025,
|
| 746 |
"step": 472
|
| 747 |
},
|
| 748 |
{
|
| 749 |
"epoch": 9.047619047619047,
|
| 750 |
+
"grad_norm": 0.203125,
|
| 751 |
"learning_rate": 4.527913349145441e-06,
|
| 752 |
+
"loss": 0.878,
|
| 753 |
"step": 475
|
| 754 |
},
|
| 755 |
{
|
| 756 |
"epoch": 9.142857142857142,
|
| 757 |
+
"grad_norm": 0.203125,
|
| 758 |
"learning_rate": 3.5833325466437694e-06,
|
| 759 |
+
"loss": 0.8736,
|
| 760 |
"step": 480
|
| 761 |
},
|
| 762 |
{
|
| 763 |
"epoch": 9.238095238095237,
|
| 764 |
+
"grad_norm": 0.208984375,
|
| 765 |
"learning_rate": 2.7473591381266708e-06,
|
| 766 |
+
"loss": 0.8763,
|
| 767 |
"step": 485
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 9.333333333333334,
|
| 771 |
+
"grad_norm": 0.20703125,
|
| 772 |
"learning_rate": 2.0209347957732328e-06,
|
| 773 |
+
"loss": 0.8832,
|
| 774 |
"step": 490
|
| 775 |
},
|
| 776 |
{
|
| 777 |
"epoch": 9.428571428571429,
|
| 778 |
+
"grad_norm": 0.203125,
|
| 779 |
"learning_rate": 1.404877791536141e-06,
|
| 780 |
+
"loss": 0.879,
|
| 781 |
"step": 495
|
| 782 |
},
|
| 783 |
{
|
| 784 |
"epoch": 9.523809523809524,
|
| 785 |
+
"grad_norm": 0.19921875,
|
| 786 |
"learning_rate": 8.998820754091531e-07,
|
| 787 |
+
"loss": 0.8787,
|
| 788 |
"step": 500
|
| 789 |
},
|
| 790 |
{
|
| 791 |
"epoch": 9.619047619047619,
|
| 792 |
+
"grad_norm": 0.201171875,
|
| 793 |
"learning_rate": 5.065164937354428e-07,
|
| 794 |
+
"loss": 0.8782,
|
| 795 |
"step": 505
|
| 796 |
},
|
| 797 |
{
|
| 798 |
"epoch": 9.714285714285714,
|
| 799 |
+
"grad_norm": 0.2080078125,
|
| 800 |
"learning_rate": 2.2522414843748618e-07,
|
| 801 |
+
"loss": 0.8803,
|
| 802 |
"step": 510
|
| 803 |
},
|
| 804 |
{
|
| 805 |
"epoch": 9.80952380952381,
|
| 806 |
+
"grad_norm": 0.201171875,
|
| 807 |
"learning_rate": 5.632189789027687e-08,
|
| 808 |
+
"loss": 0.8719,
|
| 809 |
"step": 515
|
| 810 |
},
|
| 811 |
{
|
| 812 |
"epoch": 9.904761904761905,
|
| 813 |
+
"grad_norm": 0.19921875,
|
| 814 |
"learning_rate": 0.0,
|
| 815 |
+
"loss": 0.8797,
|
| 816 |
"step": 520
|
| 817 |
},
|
| 818 |
{
|
| 819 |
"epoch": 9.904761904761905,
|
| 820 |
+
"eval_loss": 2.532215118408203,
|
| 821 |
+
"eval_runtime": 0.4799,
|
| 822 |
+
"eval_samples_per_second": 20.838,
|
| 823 |
+
"eval_steps_per_second": 2.084,
|
| 824 |
"step": 520
|
| 825 |
},
|
| 826 |
{
|
| 827 |
"epoch": 9.904761904761905,
|
| 828 |
"step": 520,
|
| 829 |
+
"total_flos": 8.265211656850637e+17,
|
| 830 |
+
"train_loss": 1.0387523100926326,
|
| 831 |
+
"train_runtime": 3563.5293,
|
| 832 |
+
"train_samples_per_second": 18.703,
|
| 833 |
+
"train_steps_per_second": 0.146
|
| 834 |
}
|
| 835 |
],
|
| 836 |
"logging_steps": 5,
|
|
|
|
| 850 |
"attributes": {}
|
| 851 |
}
|
| 852 |
},
|
| 853 |
+
"total_flos": 8.265211656850637e+17,
|
| 854 |
"train_batch_size": 8,
|
| 855 |
"trial_name": null,
|
| 856 |
"trial_params": null
|