Commit
•
61752de
1
Parent(s):
b6994be
sync
Browse files- README.md +38 -38
- all_results.json +9 -9
- eval_results.json +5 -5
- train_results.json +4 -4
- trainer_state.json +513 -513
README.md
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
---
|
| 2 |
language:
|
| 3 |
- sv-SE
|
| 4 |
-
license:
|
| 5 |
tags:
|
| 6 |
- automatic-speech-recognition
|
| 7 |
- mozilla-foundation/common_voice_7_0
|
|
@@ -18,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
#
|
| 20 |
|
| 21 |
-
This model is a fine-tuned version of [
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
-
- Loss: 0.
|
| 24 |
-
- Wer: 0.
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
@@ -56,40 +56,40 @@ The following hyperparameters were used during training:
|
|
| 56 |
|
| 57 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 58 |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 59 |
-
| 3.
|
| 60 |
-
| 2.
|
| 61 |
-
|
|
| 62 |
-
| 1.
|
| 63 |
-
| 1.
|
| 64 |
-
| 1.
|
| 65 |
-
| 1.
|
| 66 |
-
| 1.
|
| 67 |
-
| 1.
|
| 68 |
-
| 1.
|
| 69 |
-
| 1.
|
| 70 |
-
| 1.
|
| 71 |
-
| 1.
|
| 72 |
-
| 1.
|
| 73 |
-
| 1.
|
| 74 |
-
| 1.
|
| 75 |
-
| 1.
|
| 76 |
-
| 1.
|
| 77 |
-
| 1.
|
| 78 |
-
| 1.
|
| 79 |
-
| 1.
|
| 80 |
-
| 1.
|
| 81 |
-
| 1.
|
| 82 |
-
| 1.
|
| 83 |
-
| 1.
|
| 84 |
-
| 1.
|
| 85 |
-
| 1.
|
| 86 |
-
| 1.
|
| 87 |
-
| 1.
|
| 88 |
-
| 1.
|
| 89 |
-
| 1.
|
| 90 |
-
| 1.
|
| 91 |
-
| 1.
|
| 92 |
-
|
|
| 93 |
|
| 94 |
|
| 95 |
### Framework versions
|
|
|
|
| 1 |
---
|
| 2 |
language:
|
| 3 |
- sv-SE
|
| 4 |
+
license: cc0-1.0
|
| 5 |
tags:
|
| 6 |
- automatic-speech-recognition
|
| 7 |
- mozilla-foundation/common_voice_7_0
|
|
|
|
| 18 |
|
| 19 |
#
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [KBLab/wav2vec2-large-voxrex](https://huggingface.co/KBLab/wav2vec2-large-voxrex) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - SV-SE dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.2201
|
| 24 |
+
- Wer: 0.1778
|
| 25 |
|
| 26 |
## Model description
|
| 27 |
|
|
|
|
| 56 |
|
| 57 |
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
| 58 |
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
| 59 |
+
| 3.1522 | 1.45 | 500 | 3.1290 | 1.0 |
|
| 60 |
+
| 2.9576 | 2.91 | 1000 | 2.9633 | 1.0 |
|
| 61 |
+
| 1.9853 | 4.36 | 1500 | 0.8902 | 0.6104 |
|
| 62 |
+
| 1.5867 | 5.81 | 2000 | 0.4793 | 0.3664 |
|
| 63 |
+
| 1.4608 | 7.27 | 2500 | 0.3816 | 0.3095 |
|
| 64 |
+
| 1.3496 | 8.72 | 3000 | 0.3415 | 0.2783 |
|
| 65 |
+
| 1.3058 | 10.17 | 3500 | 0.3072 | 0.2519 |
|
| 66 |
+
| 1.2533 | 11.63 | 4000 | 0.2877 | 0.2381 |
|
| 67 |
+
| 1.2535 | 13.08 | 4500 | 0.2791 | 0.2320 |
|
| 68 |
+
| 1.2273 | 14.53 | 5000 | 0.2726 | 0.2282 |
|
| 69 |
+
| 1.2083 | 15.99 | 5500 | 0.2638 | 0.2212 |
|
| 70 |
+
| 1.1606 | 17.44 | 6000 | 0.2531 | 0.2174 |
|
| 71 |
+
| 1.1545 | 18.89 | 6500 | 0.2468 | 0.2109 |
|
| 72 |
+
| 1.1344 | 20.35 | 7000 | 0.2494 | 0.2050 |
|
| 73 |
+
| 1.1173 | 21.8 | 7500 | 0.2447 | 0.1980 |
|
| 74 |
+
| 1.1081 | 23.26 | 8000 | 0.2428 | 0.1998 |
|
| 75 |
+
| 1.1023 | 24.71 | 8500 | 0.2329 | 0.1951 |
|
| 76 |
+
| 1.0923 | 26.16 | 9000 | 0.2388 | 0.1962 |
|
| 77 |
+
| 1.0798 | 27.61 | 9500 | 0.2363 | 0.1944 |
|
| 78 |
+
| 1.0769 | 29.07 | 10000 | 0.2342 | 0.1913 |
|
| 79 |
+
| 1.0672 | 30.52 | 10500 | 0.2250 | 0.1875 |
|
| 80 |
+
| 1.0735 | 31.97 | 11000 | 0.2305 | 0.1874 |
|
| 81 |
+
| 1.0628 | 33.43 | 11500 | 0.2291 | 0.1851 |
|
| 82 |
+
| 1.0451 | 34.88 | 12000 | 0.2263 | 0.1856 |
|
| 83 |
+
| 1.0299 | 36.34 | 12500 | 0.2257 | 0.1834 |
|
| 84 |
+
| 1.0368 | 37.79 | 13000 | 0.2230 | 0.1808 |
|
| 85 |
+
| 1.0322 | 39.24 | 13500 | 0.2231 | 0.1833 |
|
| 86 |
+
| 1.0451 | 40.7 | 14000 | 0.2197 | 0.1817 |
|
| 87 |
+
| 1.0304 | 42.15 | 14500 | 0.2241 | 0.1813 |
|
| 88 |
+
| 1.0102 | 43.6 | 15000 | 0.2233 | 0.1795 |
|
| 89 |
+
| 1.0135 | 45.06 | 15500 | 0.2200 | 0.1794 |
|
| 90 |
+
| 1.014 | 46.51 | 16000 | 0.2207 | 0.1779 |
|
| 91 |
+
| 1.0071 | 47.96 | 16500 | 0.2205 | 0.1784 |
|
| 92 |
+
| 0.9729 | 49.42 | 17000 | 0.2204 | 0.1777 |
|
| 93 |
|
| 94 |
|
| 95 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 50.0,
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 4620,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 4.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
-
"train_loss": 1.
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 11030,
|
| 12 |
-
"train_samples_per_second": 18.
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.2201116979122162,
|
| 4 |
+
"eval_runtime": 130.7896,
|
| 5 |
"eval_samples": 4620,
|
| 6 |
+
"eval_samples_per_second": 35.324,
|
| 7 |
+
"eval_steps_per_second": 4.419,
|
| 8 |
+
"eval_wer": 0.17781313100409962,
|
| 9 |
+
"train_loss": 1.4085150359397711,
|
| 10 |
+
"train_runtime": 29405.2698,
|
| 11 |
"train_samples": 11030,
|
| 12 |
+
"train_samples_per_second": 18.755,
|
| 13 |
+
"train_steps_per_second": 0.585
|
| 14 |
}
|
eval_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 50.0,
|
| 3 |
-
"eval_loss": 0.
|
| 4 |
-
"eval_runtime":
|
| 5 |
"eval_samples": 4620,
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 4.
|
| 8 |
-
"eval_wer": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 50.0,
|
| 3 |
+
"eval_loss": 0.2201116979122162,
|
| 4 |
+
"eval_runtime": 130.7896,
|
| 5 |
"eval_samples": 4620,
|
| 6 |
+
"eval_samples_per_second": 35.324,
|
| 7 |
+
"eval_steps_per_second": 4.419,
|
| 8 |
+
"eval_wer": 0.17781313100409962
|
| 9 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 50.0,
|
| 3 |
-
"train_loss": 1.
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 11030,
|
| 6 |
-
"train_samples_per_second": 18.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 50.0,
|
| 3 |
+
"train_loss": 1.4085150359397711,
|
| 4 |
+
"train_runtime": 29405.2698,
|
| 5 |
"train_samples": 11030,
|
| 6 |
+
"train_samples_per_second": 18.755,
|
| 7 |
+
"train_steps_per_second": 0.585
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -9,1350 +9,1350 @@
|
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.29,
|
| 12 |
-
"learning_rate": 3.
|
| 13 |
-
"loss": 12.
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.58,
|
| 18 |
-
"learning_rate": 7.
|
| 19 |
-
"loss":
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.87,
|
| 24 |
-
"learning_rate": 1.
|
| 25 |
-
"loss": 4.
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 1.16,
|
| 30 |
-
"learning_rate": 1.
|
| 31 |
-
"loss": 3.
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.45,
|
| 36 |
-
"learning_rate": 1.
|
| 37 |
-
"loss": 3.
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.45,
|
| 42 |
-
"eval_loss": 3.
|
| 43 |
-
"eval_runtime":
|
| 44 |
-
"eval_samples_per_second":
|
| 45 |
-
"eval_steps_per_second": 4.
|
| 46 |
"eval_wer": 1.0,
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 1.74,
|
| 51 |
-
"learning_rate": 2.
|
| 52 |
-
"loss": 3.
|
| 53 |
"step": 600
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 2.03,
|
| 57 |
-
"learning_rate": 2.
|
| 58 |
-
"loss": 3.
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 2.32,
|
| 63 |
-
"learning_rate": 2.
|
| 64 |
-
"loss": 3.
|
| 65 |
"step": 800
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 2.61,
|
| 69 |
-
"learning_rate": 3.
|
| 70 |
-
"loss":
|
| 71 |
"step": 900
|
| 72 |
},
|
| 73 |
{
|
| 74 |
"epoch": 2.91,
|
| 75 |
-
"learning_rate": 3.
|
| 76 |
-
"loss": 2.
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 2.91,
|
| 81 |
-
"eval_loss": 2.
|
| 82 |
-
"eval_runtime":
|
| 83 |
-
"eval_samples_per_second":
|
| 84 |
-
"eval_steps_per_second":
|
| 85 |
-
"eval_wer": 0
|
| 86 |
"step": 1000
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 3.2,
|
| 90 |
-
"learning_rate": 4.
|
| 91 |
-
"loss": 2.
|
| 92 |
"step": 1100
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 3.49,
|
| 96 |
-
"learning_rate": 4.
|
| 97 |
-
"loss": 2.
|
| 98 |
"step": 1200
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 3.78,
|
| 102 |
-
"learning_rate": 4.
|
| 103 |
-
"loss": 2.
|
| 104 |
"step": 1300
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 4.07,
|
| 108 |
-
"learning_rate": 5.
|
| 109 |
-
"loss": 2.
|
| 110 |
"step": 1400
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 4.36,
|
| 114 |
-
"learning_rate": 5.
|
| 115 |
-
"loss":
|
| 116 |
"step": 1500
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 4.36,
|
| 120 |
-
"eval_loss":
|
| 121 |
-
"eval_runtime":
|
| 122 |
-
"eval_samples_per_second":
|
| 123 |
-
"eval_steps_per_second": 4.
|
| 124 |
-
"eval_wer": 0.
|
| 125 |
"step": 1500
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 4.65,
|
| 129 |
-
"learning_rate": 5.
|
| 130 |
-
"loss": 1.
|
| 131 |
"step": 1600
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 4.94,
|
| 135 |
-
"learning_rate": 6.
|
| 136 |
-
"loss": 1.
|
| 137 |
"step": 1700
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 5.23,
|
| 141 |
-
"learning_rate": 6.
|
| 142 |
-
"loss": 1.
|
| 143 |
"step": 1800
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 5.52,
|
| 147 |
-
"learning_rate": 7.
|
| 148 |
-
"loss": 1.
|
| 149 |
"step": 1900
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 5.81,
|
| 153 |
-
"learning_rate": 7.
|
| 154 |
-
"loss": 1.
|
| 155 |
"step": 2000
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 5.81,
|
| 159 |
-
"eval_loss": 0.
|
| 160 |
-
"eval_runtime":
|
| 161 |
-
"eval_samples_per_second":
|
| 162 |
-
"eval_steps_per_second":
|
| 163 |
-
"eval_wer": 0.
|
| 164 |
"step": 2000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 6.1,
|
| 168 |
-
"learning_rate": 7.
|
| 169 |
-
"loss": 1.
|
| 170 |
"step": 2100
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 6.39,
|
| 174 |
-
"learning_rate": 7.
|
| 175 |
-
"loss": 1.
|
| 176 |
"step": 2200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 6.68,
|
| 180 |
-
"learning_rate": 7.
|
| 181 |
-
"loss": 1.
|
| 182 |
"step": 2300
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 6.97,
|
| 186 |
-
"learning_rate": 7.
|
| 187 |
-
"loss": 1.
|
| 188 |
"step": 2400
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 7.27,
|
| 192 |
-
"learning_rate": 7.
|
| 193 |
-
"loss": 1.
|
| 194 |
"step": 2500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 7.27,
|
| 198 |
-
"eval_loss": 0.
|
| 199 |
-
"eval_runtime":
|
| 200 |
-
"eval_samples_per_second":
|
| 201 |
-
"eval_steps_per_second": 4.
|
| 202 |
-
"eval_wer": 0.
|
| 203 |
"step": 2500
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 7.56,
|
| 207 |
-
"learning_rate": 7.
|
| 208 |
-
"loss": 1.
|
| 209 |
"step": 2600
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 7.85,
|
| 213 |
-
"learning_rate": 7.
|
| 214 |
-
"loss": 1.
|
| 215 |
"step": 2700
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 8.14,
|
| 219 |
-
"learning_rate": 7.
|
| 220 |
-
"loss": 1.
|
| 221 |
"step": 2800
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 8.43,
|
| 225 |
-
"learning_rate": 7.
|
| 226 |
-
"loss": 1.
|
| 227 |
"step": 2900
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 8.72,
|
| 231 |
-
"learning_rate": 7.
|
| 232 |
-
"loss": 1.
|
| 233 |
"step": 3000
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 8.72,
|
| 237 |
-
"eval_loss": 0.
|
| 238 |
-
"eval_runtime":
|
| 239 |
-
"eval_samples_per_second":
|
| 240 |
-
"eval_steps_per_second": 4.
|
| 241 |
-
"eval_wer": 0.
|
| 242 |
"step": 3000
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 9.01,
|
| 246 |
-
"learning_rate": 6.
|
| 247 |
-
"loss": 1.
|
| 248 |
"step": 3100
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 9.3,
|
| 252 |
-
"learning_rate": 6.
|
| 253 |
-
"loss": 1.
|
| 254 |
"step": 3200
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 9.59,
|
| 258 |
-
"learning_rate": 6.
|
| 259 |
-
"loss": 1.
|
| 260 |
"step": 3300
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 9.88,
|
| 264 |
-
"learning_rate": 6.
|
| 265 |
-
"loss": 1.
|
| 266 |
"step": 3400
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 10.17,
|
| 270 |
-
"learning_rate": 6.
|
| 271 |
-
"loss": 1.
|
| 272 |
"step": 3500
|
| 273 |
},
|
| 274 |
{
|
| 275 |
"epoch": 10.17,
|
| 276 |
-
"eval_loss": 0.
|
| 277 |
-
"eval_runtime":
|
| 278 |
-
"eval_samples_per_second": 34.
|
| 279 |
-
"eval_steps_per_second": 4.
|
| 280 |
-
"eval_wer": 0.
|
| 281 |
"step": 3500
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 10.46,
|
| 285 |
-
"learning_rate": 6.
|
| 286 |
-
"loss": 1.
|
| 287 |
"step": 3600
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 10.75,
|
| 291 |
-
"learning_rate": 6.
|
| 292 |
-
"loss": 1.
|
| 293 |
"step": 3700
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"epoch": 11.05,
|
| 297 |
-
"learning_rate": 6.
|
| 298 |
-
"loss": 1.
|
| 299 |
"step": 3800
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 11.34,
|
| 303 |
-
"learning_rate": 6.
|
| 304 |
-
"loss": 1.
|
| 305 |
"step": 3900
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 11.63,
|
| 309 |
-
"learning_rate": 6.
|
| 310 |
-
"loss": 1.
|
| 311 |
"step": 4000
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 11.63,
|
| 315 |
-
"eval_loss": 0.
|
| 316 |
-
"eval_runtime":
|
| 317 |
-
"eval_samples_per_second":
|
| 318 |
-
"eval_steps_per_second": 4.
|
| 319 |
-
"eval_wer": 0.
|
| 320 |
"step": 4000
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 11.92,
|
| 324 |
-
"learning_rate": 6.
|
| 325 |
-
"loss": 1.
|
| 326 |
"step": 4100
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 12.21,
|
| 330 |
-
"learning_rate": 6.
|
| 331 |
-
"loss": 1.
|
| 332 |
"step": 4200
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 12.5,
|
| 336 |
-
"learning_rate": 6.
|
| 337 |
-
"loss": 1.
|
| 338 |
"step": 4300
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 12.79,
|
| 342 |
-
"learning_rate": 6.
|
| 343 |
-
"loss": 1.
|
| 344 |
"step": 4400
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 13.08,
|
| 348 |
-
"learning_rate": 6.
|
| 349 |
-
"loss": 1.
|
| 350 |
"step": 4500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 13.08,
|
| 354 |
-
"eval_loss": 0.
|
| 355 |
-
"eval_runtime":
|
| 356 |
-
"eval_samples_per_second":
|
| 357 |
-
"eval_steps_per_second": 4.
|
| 358 |
-
"eval_wer": 0.
|
| 359 |
"step": 4500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 13.37,
|
| 363 |
-
"learning_rate": 6.
|
| 364 |
-
"loss": 1.
|
| 365 |
"step": 4600
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 13.66,
|
| 369 |
-
"learning_rate": 6.
|
| 370 |
-
"loss": 1.
|
| 371 |
"step": 4700
|
| 372 |
},
|
| 373 |
{
|
| 374 |
"epoch": 13.95,
|
| 375 |
-
"learning_rate": 6.
|
| 376 |
-
"loss": 1.
|
| 377 |
"step": 4800
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 14.24,
|
| 381 |
-
"learning_rate": 6.
|
| 382 |
-
"loss": 1.
|
| 383 |
"step": 4900
|
| 384 |
},
|
| 385 |
{
|
| 386 |
"epoch": 14.53,
|
| 387 |
-
"learning_rate": 6.
|
| 388 |
-
"loss": 1.
|
| 389 |
"step": 5000
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 14.53,
|
| 393 |
-
"eval_loss": 0.
|
| 394 |
-
"eval_runtime":
|
| 395 |
-
"eval_samples_per_second":
|
| 396 |
-
"eval_steps_per_second": 4.
|
| 397 |
-
"eval_wer": 0.
|
| 398 |
"step": 5000
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 14.82,
|
| 402 |
-
"learning_rate": 5.
|
| 403 |
-
"loss": 1.
|
| 404 |
"step": 5100
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 15.12,
|
| 408 |
-
"learning_rate": 5.
|
| 409 |
-
"loss": 1.
|
| 410 |
"step": 5200
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 15.41,
|
| 414 |
-
"learning_rate": 5.
|
| 415 |
-
"loss": 1.
|
| 416 |
"step": 5300
|
| 417 |
},
|
| 418 |
{
|
| 419 |
"epoch": 15.7,
|
| 420 |
-
"learning_rate": 5.
|
| 421 |
-
"loss": 1.
|
| 422 |
"step": 5400
|
| 423 |
},
|
| 424 |
{
|
| 425 |
"epoch": 15.99,
|
| 426 |
-
"learning_rate": 5.
|
| 427 |
-
"loss": 1.
|
| 428 |
"step": 5500
|
| 429 |
},
|
| 430 |
{
|
| 431 |
"epoch": 15.99,
|
| 432 |
-
"eval_loss": 0.
|
| 433 |
-
"eval_runtime":
|
| 434 |
-
"eval_samples_per_second":
|
| 435 |
-
"eval_steps_per_second": 4.
|
| 436 |
-
"eval_wer": 0.
|
| 437 |
"step": 5500
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 16.28,
|
| 441 |
-
"learning_rate": 5.
|
| 442 |
-
"loss": 1.
|
| 443 |
"step": 5600
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 16.57,
|
| 447 |
-
"learning_rate": 5.
|
| 448 |
-
"loss": 1.
|
| 449 |
"step": 5700
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 16.86,
|
| 453 |
-
"learning_rate": 5.
|
| 454 |
-
"loss": 1.
|
| 455 |
"step": 5800
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 17.15,
|
| 459 |
-
"learning_rate": 5.
|
| 460 |
-
"loss": 1.
|
| 461 |
"step": 5900
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 17.44,
|
| 465 |
-
"learning_rate": 5.
|
| 466 |
-
"loss": 1.
|
| 467 |
"step": 6000
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 17.44,
|
| 471 |
-
"eval_loss": 0.
|
| 472 |
-
"eval_runtime":
|
| 473 |
-
"eval_samples_per_second":
|
| 474 |
-
"eval_steps_per_second": 4.
|
| 475 |
-
"eval_wer": 0.
|
| 476 |
"step": 6000
|
| 477 |
},
|
| 478 |
{
|
| 479 |
"epoch": 17.73,
|
| 480 |
-
"learning_rate": 5.
|
| 481 |
-
"loss": 1.
|
| 482 |
"step": 6100
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 18.02,
|
| 486 |
-
"learning_rate": 5.
|
| 487 |
-
"loss": 1.
|
| 488 |
"step": 6200
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 18.31,
|
| 492 |
-
"learning_rate": 5.
|
| 493 |
-
"loss": 1.
|
| 494 |
"step": 6300
|
| 495 |
},
|
| 496 |
{
|
| 497 |
"epoch": 18.6,
|
| 498 |
-
"learning_rate": 5.
|
| 499 |
-
"loss": 1.
|
| 500 |
"step": 6400
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 18.89,
|
| 504 |
-
"learning_rate": 5.
|
| 505 |
-
"loss": 1.
|
| 506 |
"step": 6500
|
| 507 |
},
|
| 508 |
{
|
| 509 |
"epoch": 18.89,
|
| 510 |
-
"eval_loss": 0.
|
| 511 |
-
"eval_runtime":
|
| 512 |
-
"eval_samples_per_second":
|
| 513 |
-
"eval_steps_per_second": 4.
|
| 514 |
-
"eval_wer": 0.
|
| 515 |
"step": 6500
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 19.19,
|
| 519 |
-
"learning_rate": 5.
|
| 520 |
-
"loss": 1.
|
| 521 |
"step": 6600
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 19.48,
|
| 525 |
"learning_rate": 5.182894736842105e-05,
|
| 526 |
-
"loss": 1.
|
| 527 |
"step": 6700
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 19.77,
|
| 531 |
"learning_rate": 5.133552631578947e-05,
|
| 532 |
-
"loss": 1.
|
| 533 |
"step": 6800
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 20.06,
|
| 537 |
"learning_rate": 5.084210526315789e-05,
|
| 538 |
-
"loss": 1.
|
| 539 |
"step": 6900
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 20.35,
|
| 543 |
-
"learning_rate": 5.
|
| 544 |
-
"loss": 1.
|
| 545 |
"step": 7000
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 20.35,
|
| 549 |
-
"eval_loss": 0.
|
| 550 |
-
"eval_runtime":
|
| 551 |
-
"eval_samples_per_second":
|
| 552 |
-
"eval_steps_per_second": 4.
|
| 553 |
-
"eval_wer": 0.
|
| 554 |
"step": 7000
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 20.64,
|
| 558 |
-
"learning_rate": 4.
|
| 559 |
-
"loss": 1.
|
| 560 |
"step": 7100
|
| 561 |
},
|
| 562 |
{
|
| 563 |
"epoch": 20.93,
|
| 564 |
-
"learning_rate": 4.
|
| 565 |
-
"loss": 1.
|
| 566 |
"step": 7200
|
| 567 |
},
|
| 568 |
{
|
| 569 |
"epoch": 21.22,
|
| 570 |
-
"learning_rate": 4.
|
| 571 |
-
"loss": 1.
|
| 572 |
"step": 7300
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 21.51,
|
| 576 |
-
"learning_rate": 4.
|
| 577 |
-
"loss": 1.
|
| 578 |
"step": 7400
|
| 579 |
},
|
| 580 |
{
|
| 581 |
"epoch": 21.8,
|
| 582 |
-
"learning_rate": 4.
|
| 583 |
-
"loss": 1.
|
| 584 |
"step": 7500
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 21.8,
|
| 588 |
-
"eval_loss": 0.
|
| 589 |
-
"eval_runtime":
|
| 590 |
-
"eval_samples_per_second":
|
| 591 |
-
"eval_steps_per_second":
|
| 592 |
-
"eval_wer": 0.
|
| 593 |
"step": 7500
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 22.09,
|
| 597 |
"learning_rate": 4.739309210526315e-05,
|
| 598 |
-
"loss": 1.
|
| 599 |
"step": 7600
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 22.38,
|
| 603 |
-
"learning_rate": 4.
|
| 604 |
-
"loss": 1.
|
| 605 |
"step": 7700
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 22.67,
|
| 609 |
-
"learning_rate": 4.
|
| 610 |
-
"loss": 1.
|
| 611 |
"step": 7800
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 22.96,
|
| 615 |
-
"learning_rate": 4.
|
| 616 |
-
"loss": 1.
|
| 617 |
"step": 7900
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 23.26,
|
| 621 |
-
"learning_rate": 4.
|
| 622 |
-
"loss": 1.
|
| 623 |
"step": 8000
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 23.26,
|
| 627 |
-
"eval_loss": 0.
|
| 628 |
-
"eval_runtime":
|
| 629 |
-
"eval_samples_per_second":
|
| 630 |
-
"eval_steps_per_second": 4.
|
| 631 |
-
"eval_wer": 0.
|
| 632 |
"step": 8000
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 23.55,
|
| 636 |
-
"learning_rate": 4.
|
| 637 |
-
"loss": 1.
|
| 638 |
"step": 8100
|
| 639 |
},
|
| 640 |
{
|
| 641 |
"epoch": 23.84,
|
| 642 |
-
"learning_rate": 4.
|
| 643 |
-
"loss": 1.
|
| 644 |
"step": 8200
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 24.13,
|
| 648 |
-
"learning_rate": 4.
|
| 649 |
-
"loss": 1.
|
| 650 |
"step": 8300
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 24.42,
|
| 654 |
-
"learning_rate": 4.
|
| 655 |
-
"loss": 1.
|
| 656 |
"step": 8400
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 24.71,
|
| 660 |
-
"learning_rate": 4.
|
| 661 |
-
"loss": 1.
|
| 662 |
"step": 8500
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 24.71,
|
| 666 |
-
"eval_loss": 0.
|
| 667 |
-
"eval_runtime":
|
| 668 |
-
"eval_samples_per_second":
|
| 669 |
-
"eval_steps_per_second": 4.
|
| 670 |
-
"eval_wer": 0.
|
| 671 |
"step": 8500
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 25.0,
|
| 675 |
-
"learning_rate": 4.
|
| 676 |
-
"loss": 1.
|
| 677 |
"step": 8600
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 25.29,
|
| 681 |
-
"learning_rate": 4.
|
| 682 |
-
"loss": 1.
|
| 683 |
"step": 8700
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 25.58,
|
| 687 |
-
"learning_rate": 4.
|
| 688 |
-
"loss": 1.
|
| 689 |
"step": 8800
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 25.87,
|
| 693 |
-
"learning_rate": 4.
|
| 694 |
-
"loss": 1.
|
| 695 |
"step": 8900
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 26.16,
|
| 699 |
-
"learning_rate": 4.
|
| 700 |
-
"loss": 1.
|
| 701 |
"step": 9000
|
| 702 |
},
|
| 703 |
{
|
| 704 |
"epoch": 26.16,
|
| 705 |
-
"eval_loss": 0.
|
| 706 |
-
"eval_runtime":
|
| 707 |
-
"eval_samples_per_second":
|
| 708 |
-
"eval_steps_per_second":
|
| 709 |
-
"eval_wer": 0.
|
| 710 |
"step": 9000
|
| 711 |
},
|
| 712 |
{
|
| 713 |
"epoch": 26.45,
|
| 714 |
-
"learning_rate": 3.
|
| 715 |
-
"loss": 1.
|
| 716 |
"step": 9100
|
| 717 |
},
|
| 718 |
{
|
| 719 |
"epoch": 26.74,
|
| 720 |
-
"learning_rate": 3.
|
| 721 |
-
"loss": 1.
|
| 722 |
"step": 9200
|
| 723 |
},
|
| 724 |
{
|
| 725 |
"epoch": 27.03,
|
| 726 |
-
"learning_rate": 3.
|
| 727 |
-
"loss": 1.
|
| 728 |
"step": 9300
|
| 729 |
},
|
| 730 |
{
|
| 731 |
"epoch": 27.32,
|
| 732 |
-
"learning_rate": 3.
|
| 733 |
-
"loss": 1.
|
| 734 |
"step": 9400
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 27.61,
|
| 738 |
-
"learning_rate": 3.
|
| 739 |
-
"loss": 1.
|
| 740 |
"step": 9500
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 27.61,
|
| 744 |
-
"eval_loss": 0.
|
| 745 |
-
"eval_runtime":
|
| 746 |
-
"eval_samples_per_second":
|
| 747 |
-
"eval_steps_per_second":
|
| 748 |
-
"eval_wer": 0.
|
| 749 |
"step": 9500
|
| 750 |
},
|
| 751 |
{
|
| 752 |
"epoch": 27.91,
|
| 753 |
-
"learning_rate": 3.
|
| 754 |
-
"loss": 1.
|
| 755 |
"step": 9600
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 28.2,
|
| 759 |
-
"learning_rate": 3.
|
| 760 |
-
"loss": 1.
|
| 761 |
"step": 9700
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 28.49,
|
| 765 |
-
"learning_rate": 3.
|
| 766 |
-
"loss": 1.
|
| 767 |
"step": 9800
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 28.78,
|
| 771 |
-
"learning_rate": 3.
|
| 772 |
-
"loss": 1.
|
| 773 |
"step": 9900
|
| 774 |
},
|
| 775 |
{
|
| 776 |
"epoch": 29.07,
|
| 777 |
-
"learning_rate": 3.
|
| 778 |
-
"loss": 1.
|
| 779 |
"step": 10000
|
| 780 |
},
|
| 781 |
{
|
| 782 |
"epoch": 29.07,
|
| 783 |
-
"eval_loss": 0.
|
| 784 |
-
"eval_runtime":
|
| 785 |
-
"eval_samples_per_second":
|
| 786 |
-
"eval_steps_per_second": 4.
|
| 787 |
-
"eval_wer": 0.
|
| 788 |
"step": 10000
|
| 789 |
},
|
| 790 |
{
|
| 791 |
"epoch": 29.36,
|
| 792 |
-
"learning_rate": 3.
|
| 793 |
-
"loss": 1.
|
| 794 |
"step": 10100
|
| 795 |
},
|
| 796 |
{
|
| 797 |
"epoch": 29.65,
|
| 798 |
-
"learning_rate": 3.
|
| 799 |
-
"loss": 1.
|
| 800 |
"step": 10200
|
| 801 |
},
|
| 802 |
{
|
| 803 |
"epoch": 29.94,
|
| 804 |
-
"learning_rate": 3.
|
| 805 |
-
"loss": 1.
|
| 806 |
"step": 10300
|
| 807 |
},
|
| 808 |
{
|
| 809 |
"epoch": 30.23,
|
| 810 |
-
"learning_rate": 3.
|
| 811 |
-
"loss": 1.
|
| 812 |
"step": 10400
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 30.52,
|
| 816 |
-
"learning_rate": 3.
|
| 817 |
-
"loss": 1.
|
| 818 |
"step": 10500
|
| 819 |
},
|
| 820 |
{
|
| 821 |
"epoch": 30.52,
|
| 822 |
-
"eval_loss": 0.
|
| 823 |
-
"eval_runtime":
|
| 824 |
-
"eval_samples_per_second":
|
| 825 |
-
"eval_steps_per_second":
|
| 826 |
-
"eval_wer": 0.
|
| 827 |
"step": 10500
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 30.81,
|
| 831 |
-
"learning_rate": 3.
|
| 832 |
-
"loss": 1.
|
| 833 |
"step": 10600
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 31.1,
|
| 837 |
-
"learning_rate": 3.
|
| 838 |
-
"loss": 1.
|
| 839 |
"step": 10700
|
| 840 |
},
|
| 841 |
{
|
| 842 |
"epoch": 31.39,
|
| 843 |
-
"learning_rate": 3.
|
| 844 |
-
"loss": 1.
|
| 845 |
"step": 10800
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 31.68,
|
| 849 |
-
"learning_rate": 3.
|
| 850 |
-
"loss": 1.
|
| 851 |
"step": 10900
|
| 852 |
},
|
| 853 |
{
|
| 854 |
"epoch": 31.97,
|
| 855 |
-
"learning_rate": 3.
|
| 856 |
-
"loss": 1.
|
| 857 |
"step": 11000
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 31.97,
|
| 861 |
-
"eval_loss": 0.
|
| 862 |
-
"eval_runtime":
|
| 863 |
-
"eval_samples_per_second":
|
| 864 |
-
"eval_steps_per_second":
|
| 865 |
-
"eval_wer": 0.
|
| 866 |
"step": 11000
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 32.27,
|
| 870 |
-
"learning_rate": 3.
|
| 871 |
-
"loss": 1.
|
| 872 |
"step": 11100
|
| 873 |
},
|
| 874 |
{
|
| 875 |
"epoch": 32.56,
|
| 876 |
-
"learning_rate": 2.
|
| 877 |
-
"loss": 1.
|
| 878 |
"step": 11200
|
| 879 |
},
|
| 880 |
{
|
| 881 |
"epoch": 32.85,
|
| 882 |
-
"learning_rate": 2.
|
| 883 |
-
"loss": 1.
|
| 884 |
"step": 11300
|
| 885 |
},
|
| 886 |
{
|
| 887 |
"epoch": 33.14,
|
| 888 |
-
"learning_rate": 2.
|
| 889 |
-
"loss": 1.
|
| 890 |
"step": 11400
|
| 891 |
},
|
| 892 |
{
|
| 893 |
"epoch": 33.43,
|
| 894 |
-
"learning_rate": 2.
|
| 895 |
-
"loss": 1.
|
| 896 |
"step": 11500
|
| 897 |
},
|
| 898 |
{
|
| 899 |
"epoch": 33.43,
|
| 900 |
-
"eval_loss": 0.
|
| 901 |
-
"eval_runtime":
|
| 902 |
-
"eval_samples_per_second":
|
| 903 |
-
"eval_steps_per_second":
|
| 904 |
-
"eval_wer": 0.
|
| 905 |
"step": 11500
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 33.72,
|
| 909 |
-
"learning_rate": 2.
|
| 910 |
-
"loss": 1.
|
| 911 |
"step": 11600
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 34.01,
|
| 915 |
-
"learning_rate": 2.
|
| 916 |
-
"loss": 1.
|
| 917 |
"step": 11700
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 34.3,
|
| 921 |
-
"learning_rate": 2.
|
| 922 |
-
"loss": 1.
|
| 923 |
"step": 11800
|
| 924 |
},
|
| 925 |
{
|
| 926 |
"epoch": 34.59,
|
| 927 |
-
"learning_rate": 2.
|
| 928 |
-
"loss": 1.
|
| 929 |
"step": 11900
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 34.88,
|
| 933 |
-
"learning_rate": 2.
|
| 934 |
-
"loss": 1.
|
| 935 |
"step": 12000
|
| 936 |
},
|
| 937 |
{
|
| 938 |
"epoch": 34.88,
|
| 939 |
-
"eval_loss": 0.
|
| 940 |
-
"eval_runtime":
|
| 941 |
-
"eval_samples_per_second":
|
| 942 |
-
"eval_steps_per_second":
|
| 943 |
-
"eval_wer": 0.
|
| 944 |
"step": 12000
|
| 945 |
},
|
| 946 |
{
|
| 947 |
"epoch": 35.17,
|
| 948 |
-
"learning_rate": 2.
|
| 949 |
-
"loss": 1.
|
| 950 |
"step": 12100
|
| 951 |
},
|
| 952 |
{
|
| 953 |
"epoch": 35.46,
|
| 954 |
-
"learning_rate": 2.
|
| 955 |
-
"loss": 1.
|
| 956 |
"step": 12200
|
| 957 |
},
|
| 958 |
{
|
| 959 |
"epoch": 35.75,
|
| 960 |
-
"learning_rate": 2.
|
| 961 |
-
"loss": 1.
|
| 962 |
"step": 12300
|
| 963 |
},
|
| 964 |
{
|
| 965 |
"epoch": 36.05,
|
| 966 |
-
"learning_rate": 2.
|
| 967 |
-
"loss": 1.
|
| 968 |
"step": 12400
|
| 969 |
},
|
| 970 |
{
|
| 971 |
"epoch": 36.34,
|
| 972 |
-
"learning_rate": 2.
|
| 973 |
-
"loss": 1.
|
| 974 |
"step": 12500
|
| 975 |
},
|
| 976 |
{
|
| 977 |
"epoch": 36.34,
|
| 978 |
-
"eval_loss": 0.
|
| 979 |
-
"eval_runtime":
|
| 980 |
-
"eval_samples_per_second":
|
| 981 |
-
"eval_steps_per_second":
|
| 982 |
-
"eval_wer": 0.
|
| 983 |
"step": 12500
|
| 984 |
},
|
| 985 |
{
|
| 986 |
"epoch": 36.63,
|
| 987 |
-
"learning_rate": 2.
|
| 988 |
-
"loss": 1.
|
| 989 |
"step": 12600
|
| 990 |
},
|
| 991 |
{
|
| 992 |
"epoch": 36.92,
|
| 993 |
-
"learning_rate": 2.
|
| 994 |
-
"loss": 1.
|
| 995 |
"step": 12700
|
| 996 |
},
|
| 997 |
{
|
| 998 |
"epoch": 37.21,
|
| 999 |
-
"learning_rate": 2.
|
| 1000 |
-
"loss": 1.
|
| 1001 |
"step": 12800
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
"epoch": 37.5,
|
| 1005 |
-
"learning_rate": 2.
|
| 1006 |
-
"loss": 1.
|
| 1007 |
"step": 12900
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
"epoch": 37.79,
|
| 1011 |
-
"learning_rate": 2.
|
| 1012 |
-
"loss": 1.
|
| 1013 |
"step": 13000
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
"epoch": 37.79,
|
| 1017 |
-
"eval_loss": 0.
|
| 1018 |
-
"eval_runtime":
|
| 1019 |
-
"eval_samples_per_second":
|
| 1020 |
-
"eval_steps_per_second":
|
| 1021 |
-
"eval_wer": 0.
|
| 1022 |
"step": 13000
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
"epoch": 38.08,
|
| 1026 |
-
"learning_rate": 2.
|
| 1027 |
-
"loss": 1.
|
| 1028 |
"step": 13100
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
"epoch": 38.37,
|
| 1032 |
-
"learning_rate": 1.
|
| 1033 |
-
"loss": 1.
|
| 1034 |
"step": 13200
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
"epoch": 38.66,
|
| 1038 |
-
"learning_rate": 1.
|
| 1039 |
-
"loss": 1.
|
| 1040 |
"step": 13300
|
| 1041 |
},
|
| 1042 |
{
|
| 1043 |
"epoch": 38.95,
|
| 1044 |
-
"learning_rate": 1.
|
| 1045 |
-
"loss": 1.
|
| 1046 |
"step": 13400
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
"epoch": 39.24,
|
| 1050 |
-
"learning_rate": 1.
|
| 1051 |
-
"loss": 1.
|
| 1052 |
"step": 13500
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
"epoch": 39.24,
|
| 1056 |
-
"eval_loss": 0.
|
| 1057 |
-
"eval_runtime":
|
| 1058 |
-
"eval_samples_per_second":
|
| 1059 |
-
"eval_steps_per_second":
|
| 1060 |
-
"eval_wer": 0.
|
| 1061 |
"step": 13500
|
| 1062 |
},
|
| 1063 |
{
|
| 1064 |
"epoch": 39.53,
|
| 1065 |
-
"learning_rate": 1.
|
| 1066 |
-
"loss": 1.
|
| 1067 |
"step": 13600
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
"epoch": 39.82,
|
| 1071 |
-
"learning_rate": 1.
|
| 1072 |
-
"loss": 1.
|
| 1073 |
"step": 13700
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
"epoch": 40.12,
|
| 1077 |
-
"learning_rate": 1.
|
| 1078 |
-
"loss": 1.
|
| 1079 |
"step": 13800
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
"epoch": 40.41,
|
| 1083 |
-
"learning_rate": 1.
|
| 1084 |
-
"loss": 1.
|
| 1085 |
"step": 13900
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
"epoch": 40.7,
|
| 1089 |
-
"learning_rate": 1.
|
| 1090 |
-
"loss": 1.
|
| 1091 |
"step": 14000
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
"epoch": 40.7,
|
| 1095 |
-
"eval_loss": 0.
|
| 1096 |
-
"eval_runtime":
|
| 1097 |
-
"eval_samples_per_second":
|
| 1098 |
-
"eval_steps_per_second":
|
| 1099 |
-
"eval_wer": 0.
|
| 1100 |
"step": 14000
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
"epoch": 40.99,
|
| 1104 |
-
"learning_rate": 1.
|
| 1105 |
-
"loss": 1.
|
| 1106 |
"step": 14100
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
"epoch": 41.28,
|
| 1110 |
-
"learning_rate": 1.
|
| 1111 |
-
"loss": 1.
|
| 1112 |
"step": 14200
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
"epoch": 41.57,
|
| 1116 |
-
"learning_rate": 1.
|
| 1117 |
-
"loss": 1.
|
| 1118 |
"step": 14300
|
| 1119 |
},
|
| 1120 |
{
|
| 1121 |
"epoch": 41.86,
|
| 1122 |
-
"learning_rate": 1.
|
| 1123 |
-
"loss":
|
| 1124 |
"step": 14400
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
"epoch": 42.15,
|
| 1128 |
-
"learning_rate": 1.
|
| 1129 |
-
"loss": 1.
|
| 1130 |
"step": 14500
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
"epoch": 42.15,
|
| 1134 |
-
"eval_loss": 0.
|
| 1135 |
-
"eval_runtime":
|
| 1136 |
-
"eval_samples_per_second":
|
| 1137 |
-
"eval_steps_per_second":
|
| 1138 |
-
"eval_wer": 0.
|
| 1139 |
"step": 14500
|
| 1140 |
},
|
| 1141 |
{
|
| 1142 |
"epoch": 42.44,
|
| 1143 |
-
"learning_rate": 1.
|
| 1144 |
-
"loss": 1.
|
| 1145 |
"step": 14600
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
"epoch": 42.73,
|
| 1149 |
-
"learning_rate": 1.
|
| 1150 |
-
"loss":
|
| 1151 |
"step": 14700
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
"epoch": 43.02,
|
| 1155 |
-
"learning_rate": 1.
|
| 1156 |
-
"loss": 1.
|
| 1157 |
"step": 14800
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 43.31,
|
| 1161 |
-
"learning_rate": 1.
|
| 1162 |
-
"loss": 1.
|
| 1163 |
"step": 14900
|
| 1164 |
},
|
| 1165 |
{
|
| 1166 |
"epoch": 43.6,
|
| 1167 |
-
"learning_rate": 1.
|
| 1168 |
-
"loss": 1.
|
| 1169 |
"step": 15000
|
| 1170 |
},
|
| 1171 |
{
|
| 1172 |
"epoch": 43.6,
|
| 1173 |
-
"eval_loss": 0.
|
| 1174 |
-
"eval_runtime":
|
| 1175 |
-
"eval_samples_per_second":
|
| 1176 |
-
"eval_steps_per_second": 4.
|
| 1177 |
-
"eval_wer": 0.
|
| 1178 |
"step": 15000
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 43.89,
|
| 1182 |
-
"learning_rate": 1.
|
| 1183 |
-
"loss": 1.
|
| 1184 |
"step": 15100
|
| 1185 |
},
|
| 1186 |
{
|
| 1187 |
"epoch": 44.19,
|
| 1188 |
-
"learning_rate": 9.
|
| 1189 |
-
"loss": 1.
|
| 1190 |
"step": 15200
|
| 1191 |
},
|
| 1192 |
{
|
| 1193 |
"epoch": 44.48,
|
| 1194 |
-
"learning_rate": 9.
|
| 1195 |
-
"loss": 1.
|
| 1196 |
"step": 15300
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
"epoch": 44.77,
|
| 1200 |
-
"learning_rate": 8.
|
| 1201 |
-
"loss": 1.
|
| 1202 |
"step": 15400
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
"epoch": 45.06,
|
| 1206 |
-
"learning_rate": 8.
|
| 1207 |
-
"loss": 1.
|
| 1208 |
"step": 15500
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
"epoch": 45.06,
|
| 1212 |
-
"eval_loss": 0.
|
| 1213 |
-
"eval_runtime":
|
| 1214 |
-
"eval_samples_per_second":
|
| 1215 |
-
"eval_steps_per_second": 4.
|
| 1216 |
-
"eval_wer": 0.
|
| 1217 |
"step": 15500
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
"epoch": 45.35,
|
| 1221 |
-
"learning_rate": 7.
|
| 1222 |
-
"loss": 1.
|
| 1223 |
"step": 15600
|
| 1224 |
},
|
| 1225 |
{
|
| 1226 |
"epoch": 45.64,
|
| 1227 |
-
"learning_rate": 7.
|
| 1228 |
-
"loss": 1.
|
| 1229 |
"step": 15700
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
"epoch": 45.93,
|
| 1233 |
-
"learning_rate": 6.
|
| 1234 |
-
"loss": 1.
|
| 1235 |
"step": 15800
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
"epoch": 46.22,
|
| 1239 |
-
"learning_rate": 6.
|
| 1240 |
-
"loss": 1.
|
| 1241 |
"step": 15900
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 46.51,
|
| 1245 |
-
"learning_rate": 5.
|
| 1246 |
-
"loss": 1.
|
| 1247 |
"step": 16000
|
| 1248 |
},
|
| 1249 |
{
|
| 1250 |
"epoch": 46.51,
|
| 1251 |
-
"eval_loss": 0.
|
| 1252 |
-
"eval_runtime":
|
| 1253 |
-
"eval_samples_per_second":
|
| 1254 |
-
"eval_steps_per_second": 4.
|
| 1255 |
-
"eval_wer": 0.
|
| 1256 |
"step": 16000
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
"epoch": 46.8,
|
| 1260 |
-
"learning_rate": 5.
|
| 1261 |
-
"loss": 1.
|
| 1262 |
"step": 16100
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
"epoch": 47.09,
|
| 1266 |
-
"learning_rate": 4.
|
| 1267 |
-
"loss": 1.
|
| 1268 |
"step": 16200
|
| 1269 |
},
|
| 1270 |
{
|
| 1271 |
"epoch": 47.38,
|
| 1272 |
-
"learning_rate": 4.
|
| 1273 |
-
"loss":
|
| 1274 |
"step": 16300
|
| 1275 |
},
|
| 1276 |
{
|
| 1277 |
"epoch": 47.67,
|
| 1278 |
-
"learning_rate": 3.
|
| 1279 |
-
"loss":
|
| 1280 |
"step": 16400
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
"epoch": 47.96,
|
| 1284 |
-
"learning_rate": 3.
|
| 1285 |
-
"loss": 1.
|
| 1286 |
"step": 16500
|
| 1287 |
},
|
| 1288 |
{
|
| 1289 |
"epoch": 47.96,
|
| 1290 |
-
"eval_loss": 0.
|
| 1291 |
-
"eval_runtime":
|
| 1292 |
-
"eval_samples_per_second":
|
| 1293 |
-
"eval_steps_per_second": 4.
|
| 1294 |
-
"eval_wer": 0.
|
| 1295 |
"step": 16500
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
"epoch": 48.26,
|
| 1299 |
-
"learning_rate":
|
| 1300 |
-
"loss": 1.
|
| 1301 |
"step": 16600
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
"epoch": 48.55,
|
| 1305 |
-
"learning_rate": 2.
|
| 1306 |
-
"loss": 1.
|
| 1307 |
"step": 16700
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
"epoch": 48.84,
|
| 1311 |
-
"learning_rate": 2.
|
| 1312 |
-
"loss": 1.
|
| 1313 |
"step": 16800
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
"epoch": 49.13,
|
| 1317 |
-
"learning_rate": 1.
|
| 1318 |
-
"loss": 1.
|
| 1319 |
"step": 16900
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
"epoch": 49.42,
|
| 1323 |
-
"learning_rate": 1.
|
| 1324 |
-
"loss":
|
| 1325 |
"step": 17000
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
"epoch": 49.42,
|
| 1329 |
-
"eval_loss": 0.
|
| 1330 |
-
"eval_runtime":
|
| 1331 |
-
"eval_samples_per_second":
|
| 1332 |
-
"eval_steps_per_second": 4.
|
| 1333 |
-
"eval_wer": 0.
|
| 1334 |
"step": 17000
|
| 1335 |
},
|
| 1336 |
{
|
| 1337 |
"epoch": 49.71,
|
| 1338 |
-
"learning_rate": 5.
|
| 1339 |
-
"loss": 1.
|
| 1340 |
"step": 17100
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
"epoch": 50.0,
|
| 1344 |
-
"learning_rate":
|
| 1345 |
-
"loss": 1.
|
| 1346 |
"step": 17200
|
| 1347 |
},
|
| 1348 |
{
|
| 1349 |
"epoch": 50.0,
|
| 1350 |
"step": 17200,
|
| 1351 |
"total_flos": 5.074861809449127e+19,
|
| 1352 |
-
"train_loss": 1.
|
| 1353 |
-
"train_runtime":
|
| 1354 |
-
"train_samples_per_second": 18.
|
| 1355 |
-
"train_steps_per_second": 0.
|
| 1356 |
}
|
| 1357 |
],
|
| 1358 |
"max_steps": 17200,
|
|
|
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 0.29,
|
| 12 |
+
"learning_rate": 3.6375e-06,
|
| 13 |
+
"loss": 12.958,
|
| 14 |
"step": 100
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.58,
|
| 18 |
+
"learning_rate": 7.35e-06,
|
| 19 |
+
"loss": 10.9696,
|
| 20 |
"step": 200
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.87,
|
| 24 |
+
"learning_rate": 1.1099999999999999e-05,
|
| 25 |
+
"loss": 4.9039,
|
| 26 |
"step": 300
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 1.16,
|
| 30 |
+
"learning_rate": 1.485e-05,
|
| 31 |
+
"loss": 3.4263,
|
| 32 |
"step": 400
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 1.45,
|
| 36 |
+
"learning_rate": 1.8599999999999998e-05,
|
| 37 |
+
"loss": 3.1522,
|
| 38 |
"step": 500
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 1.45,
|
| 42 |
+
"eval_loss": 3.128952741622925,
|
| 43 |
+
"eval_runtime": 133.9857,
|
| 44 |
+
"eval_samples_per_second": 34.481,
|
| 45 |
+
"eval_steps_per_second": 4.314,
|
| 46 |
"eval_wer": 1.0,
|
| 47 |
"step": 500
|
| 48 |
},
|
| 49 |
{
|
| 50 |
"epoch": 1.74,
|
| 51 |
+
"learning_rate": 2.2349999999999998e-05,
|
| 52 |
+
"loss": 3.0702,
|
| 53 |
"step": 600
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 2.03,
|
| 57 |
+
"learning_rate": 2.6099999999999997e-05,
|
| 58 |
+
"loss": 3.0589,
|
| 59 |
"step": 700
|
| 60 |
},
|
| 61 |
{
|
| 62 |
"epoch": 2.32,
|
| 63 |
+
"learning_rate": 2.985e-05,
|
| 64 |
+
"loss": 3.0223,
|
| 65 |
"step": 800
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 2.61,
|
| 69 |
+
"learning_rate": 3.36e-05,
|
| 70 |
+
"loss": 3.0021,
|
| 71 |
"step": 900
|
| 72 |
},
|
| 73 |
{
|
| 74 |
"epoch": 2.91,
|
| 75 |
+
"learning_rate": 3.735e-05,
|
| 76 |
+
"loss": 2.9576,
|
| 77 |
"step": 1000
|
| 78 |
},
|
| 79 |
{
|
| 80 |
"epoch": 2.91,
|
| 81 |
+
"eval_loss": 2.963273763656616,
|
| 82 |
+
"eval_runtime": 130.7049,
|
| 83 |
+
"eval_samples_per_second": 35.347,
|
| 84 |
+
"eval_steps_per_second": 4.422,
|
| 85 |
+
"eval_wer": 1.0,
|
| 86 |
"step": 1000
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 3.2,
|
| 90 |
+
"learning_rate": 4.11e-05,
|
| 91 |
+
"loss": 2.9499,
|
| 92 |
"step": 1100
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"epoch": 3.49,
|
| 96 |
+
"learning_rate": 4.484999999999999e-05,
|
| 97 |
+
"loss": 2.8923,
|
| 98 |
"step": 1200
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 3.78,
|
| 102 |
+
"learning_rate": 4.8599999999999995e-05,
|
| 103 |
+
"loss": 2.7406,
|
| 104 |
"step": 1300
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 4.07,
|
| 108 |
+
"learning_rate": 5.234999999999999e-05,
|
| 109 |
+
"loss": 2.3473,
|
| 110 |
"step": 1400
|
| 111 |
},
|
| 112 |
{
|
| 113 |
"epoch": 4.36,
|
| 114 |
+
"learning_rate": 5.6099999999999995e-05,
|
| 115 |
+
"loss": 1.9853,
|
| 116 |
"step": 1500
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 4.36,
|
| 120 |
+
"eval_loss": 0.8902471661567688,
|
| 121 |
+
"eval_runtime": 128.9711,
|
| 122 |
+
"eval_samples_per_second": 35.822,
|
| 123 |
+
"eval_steps_per_second": 4.482,
|
| 124 |
+
"eval_wer": 0.6104448387688919,
|
| 125 |
"step": 1500
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 4.65,
|
| 129 |
+
"learning_rate": 5.985e-05,
|
| 130 |
+
"loss": 1.8166,
|
| 131 |
"step": 1600
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 4.94,
|
| 135 |
+
"learning_rate": 6.359999999999999e-05,
|
| 136 |
+
"loss": 1.7352,
|
| 137 |
"step": 1700
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 5.23,
|
| 141 |
+
"learning_rate": 6.735e-05,
|
| 142 |
+
"loss": 1.6603,
|
| 143 |
"step": 1800
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 5.52,
|
| 147 |
+
"learning_rate": 7.11e-05,
|
| 148 |
+
"loss": 1.6183,
|
| 149 |
"step": 1900
|
| 150 |
},
|
| 151 |
{
|
| 152 |
"epoch": 5.81,
|
| 153 |
+
"learning_rate": 7.484999999999999e-05,
|
| 154 |
+
"loss": 1.5867,
|
| 155 |
"step": 2000
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 5.81,
|
| 159 |
+
"eval_loss": 0.47926634550094604,
|
| 160 |
+
"eval_runtime": 128.7387,
|
| 161 |
+
"eval_samples_per_second": 35.887,
|
| 162 |
+
"eval_steps_per_second": 4.49,
|
| 163 |
+
"eval_wer": 0.36639539864162024,
|
| 164 |
"step": 2000
|
| 165 |
},
|
| 166 |
{
|
| 167 |
"epoch": 6.1,
|
| 168 |
+
"learning_rate": 7.452631578947368e-05,
|
| 169 |
+
"loss": 1.539,
|
| 170 |
"step": 2100
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 6.39,
|
| 174 |
+
"learning_rate": 7.40328947368421e-05,
|
| 175 |
+
"loss": 1.4847,
|
| 176 |
"step": 2200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 6.68,
|
| 180 |
+
"learning_rate": 7.353947368421052e-05,
|
| 181 |
+
"loss": 1.4785,
|
| 182 |
"step": 2300
|
| 183 |
},
|
| 184 |
{
|
| 185 |
"epoch": 6.97,
|
| 186 |
+
"learning_rate": 7.304605263157895e-05,
|
| 187 |
+
"loss": 1.4361,
|
| 188 |
"step": 2400
|
| 189 |
},
|
| 190 |
{
|
| 191 |
"epoch": 7.27,
|
| 192 |
+
"learning_rate": 7.255263157894736e-05,
|
| 193 |
+
"loss": 1.4608,
|
| 194 |
"step": 2500
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 7.27,
|
| 198 |
+
"eval_loss": 0.38162761926651,
|
| 199 |
+
"eval_runtime": 130.0928,
|
| 200 |
+
"eval_samples_per_second": 35.513,
|
| 201 |
+
"eval_steps_per_second": 4.443,
|
| 202 |
+
"eval_wer": 0.3094903016582023,
|
| 203 |
"step": 2500
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 7.56,
|
| 207 |
+
"learning_rate": 7.205921052631578e-05,
|
| 208 |
+
"loss": 1.4239,
|
| 209 |
"step": 2600
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 7.85,
|
| 213 |
+
"learning_rate": 7.15657894736842e-05,
|
| 214 |
+
"loss": 1.3778,
|
| 215 |
"step": 2700
|
| 216 |
},
|
| 217 |
{
|
| 218 |
"epoch": 8.14,
|
| 219 |
+
"learning_rate": 7.107236842105262e-05,
|
| 220 |
+
"loss": 1.414,
|
| 221 |
"step": 2800
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 8.43,
|
| 225 |
+
"learning_rate": 7.057894736842104e-05,
|
| 226 |
+
"loss": 1.3957,
|
| 227 |
"step": 2900
|
| 228 |
},
|
| 229 |
{
|
| 230 |
"epoch": 8.72,
|
| 231 |
+
"learning_rate": 7.008552631578947e-05,
|
| 232 |
+
"loss": 1.3496,
|
| 233 |
"step": 3000
|
| 234 |
},
|
| 235 |
{
|
| 236 |
"epoch": 8.72,
|
| 237 |
+
"eval_loss": 0.34154650568962097,
|
| 238 |
+
"eval_runtime": 129.9374,
|
| 239 |
+
"eval_samples_per_second": 35.556,
|
| 240 |
+
"eval_steps_per_second": 4.448,
|
| 241 |
+
"eval_wer": 0.27825368659364863,
|
| 242 |
"step": 3000
|
| 243 |
},
|
| 244 |
{
|
| 245 |
"epoch": 9.01,
|
| 246 |
+
"learning_rate": 6.959210526315788e-05,
|
| 247 |
+
"loss": 1.3661,
|
| 248 |
"step": 3100
|
| 249 |
},
|
| 250 |
{
|
| 251 |
"epoch": 9.3,
|
| 252 |
+
"learning_rate": 6.909868421052631e-05,
|
| 253 |
+
"loss": 1.3144,
|
| 254 |
"step": 3200
|
| 255 |
},
|
| 256 |
{
|
| 257 |
"epoch": 9.59,
|
| 258 |
+
"learning_rate": 6.860526315789474e-05,
|
| 259 |
+
"loss": 1.3253,
|
| 260 |
"step": 3300
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 9.88,
|
| 264 |
+
"learning_rate": 6.811184210526315e-05,
|
| 265 |
+
"loss": 1.3247,
|
| 266 |
"step": 3400
|
| 267 |
},
|
| 268 |
{
|
| 269 |
"epoch": 10.17,
|
| 270 |
+
"learning_rate": 6.761842105263158e-05,
|
| 271 |
+
"loss": 1.3058,
|
| 272 |
"step": 3500
|
| 273 |
},
|
| 274 |
{
|
| 275 |
"epoch": 10.17,
|
| 276 |
+
"eval_loss": 0.307181179523468,
|
| 277 |
+
"eval_runtime": 133.008,
|
| 278 |
+
"eval_samples_per_second": 34.735,
|
| 279 |
+
"eval_steps_per_second": 4.346,
|
| 280 |
+
"eval_wer": 0.2519121336351955,
|
| 281 |
"step": 3500
|
| 282 |
},
|
| 283 |
{
|
| 284 |
"epoch": 10.46,
|
| 285 |
+
"learning_rate": 6.712499999999999e-05,
|
| 286 |
+
"loss": 1.2943,
|
| 287 |
"step": 3600
|
| 288 |
},
|
| 289 |
{
|
| 290 |
"epoch": 10.75,
|
| 291 |
+
"learning_rate": 6.663157894736842e-05,
|
| 292 |
+
"loss": 1.2986,
|
| 293 |
"step": 3700
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"epoch": 11.05,
|
| 297 |
+
"learning_rate": 6.613815789473683e-05,
|
| 298 |
+
"loss": 1.3045,
|
| 299 |
"step": 3800
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 11.34,
|
| 303 |
+
"learning_rate": 6.564473684210526e-05,
|
| 304 |
+
"loss": 1.281,
|
| 305 |
"step": 3900
|
| 306 |
},
|
| 307 |
{
|
| 308 |
"epoch": 11.63,
|
| 309 |
+
"learning_rate": 6.515131578947369e-05,
|
| 310 |
+
"loss": 1.2533,
|
| 311 |
"step": 4000
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 11.63,
|
| 315 |
+
"eval_loss": 0.2877204418182373,
|
| 316 |
+
"eval_runtime": 130.1763,
|
| 317 |
+
"eval_samples_per_second": 35.49,
|
| 318 |
+
"eval_steps_per_second": 4.44,
|
| 319 |
+
"eval_wer": 0.23814477146178792,
|
| 320 |
"step": 4000
|
| 321 |
},
|
| 322 |
{
|
| 323 |
"epoch": 11.92,
|
| 324 |
+
"learning_rate": 6.46578947368421e-05,
|
| 325 |
+
"loss": 1.268,
|
| 326 |
"step": 4100
|
| 327 |
},
|
| 328 |
{
|
| 329 |
"epoch": 12.21,
|
| 330 |
+
"learning_rate": 6.416447368421053e-05,
|
| 331 |
+
"loss": 1.2522,
|
| 332 |
"step": 4200
|
| 333 |
},
|
| 334 |
{
|
| 335 |
"epoch": 12.5,
|
| 336 |
+
"learning_rate": 6.367105263157894e-05,
|
| 337 |
+
"loss": 1.2473,
|
| 338 |
"step": 4300
|
| 339 |
},
|
| 340 |
{
|
| 341 |
"epoch": 12.79,
|
| 342 |
+
"learning_rate": 6.317763157894737e-05,
|
| 343 |
+
"loss": 1.2416,
|
| 344 |
"step": 4400
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 13.08,
|
| 348 |
+
"learning_rate": 6.268421052631578e-05,
|
| 349 |
+
"loss": 1.2535,
|
| 350 |
"step": 4500
|
| 351 |
},
|
| 352 |
{
|
| 353 |
"epoch": 13.08,
|
| 354 |
+
"eval_loss": 0.27913230657577515,
|
| 355 |
+
"eval_runtime": 130.449,
|
| 356 |
+
"eval_samples_per_second": 35.416,
|
| 357 |
+
"eval_steps_per_second": 4.431,
|
| 358 |
+
"eval_wer": 0.23202594382916233,
|
| 359 |
"step": 4500
|
| 360 |
},
|
| 361 |
{
|
| 362 |
"epoch": 13.37,
|
| 363 |
+
"learning_rate": 6.21907894736842e-05,
|
| 364 |
+
"loss": 1.2196,
|
| 365 |
"step": 4600
|
| 366 |
},
|
| 367 |
{
|
| 368 |
"epoch": 13.66,
|
| 369 |
+
"learning_rate": 6.169736842105262e-05,
|
| 370 |
+
"loss": 1.2203,
|
| 371 |
"step": 4700
|
| 372 |
},
|
| 373 |
{
|
| 374 |
"epoch": 13.95,
|
| 375 |
+
"learning_rate": 6.120394736842105e-05,
|
| 376 |
+
"loss": 1.2238,
|
| 377 |
"step": 4800
|
| 378 |
},
|
| 379 |
{
|
| 380 |
"epoch": 14.24,
|
| 381 |
+
"learning_rate": 6.0710526315789474e-05,
|
| 382 |
+
"loss": 1.2226,
|
| 383 |
"step": 4900
|
| 384 |
},
|
| 385 |
{
|
| 386 |
"epoch": 14.53,
|
| 387 |
+
"learning_rate": 6.021710526315789e-05,
|
| 388 |
+
"loss": 1.2273,
|
| 389 |
"step": 5000
|
| 390 |
},
|
| 391 |
{
|
| 392 |
"epoch": 14.53,
|
| 393 |
+
"eval_loss": 0.27258577942848206,
|
| 394 |
+
"eval_runtime": 129.7722,
|
| 395 |
+
"eval_samples_per_second": 35.601,
|
| 396 |
+
"eval_steps_per_second": 4.454,
|
| 397 |
+
"eval_wer": 0.22823227069693447,
|
| 398 |
"step": 5000
|
| 399 |
},
|
| 400 |
{
|
| 401 |
"epoch": 14.82,
|
| 402 |
+
"learning_rate": 5.972368421052631e-05,
|
| 403 |
+
"loss": 1.2016,
|
| 404 |
"step": 5100
|
| 405 |
},
|
| 406 |
{
|
| 407 |
"epoch": 15.12,
|
| 408 |
+
"learning_rate": 5.9230263157894734e-05,
|
| 409 |
+
"loss": 1.2156,
|
| 410 |
"step": 5200
|
| 411 |
},
|
| 412 |
{
|
| 413 |
"epoch": 15.41,
|
| 414 |
+
"learning_rate": 5.873684210526315e-05,
|
| 415 |
+
"loss": 1.2005,
|
| 416 |
"step": 5300
|
| 417 |
},
|
| 418 |
{
|
| 419 |
"epoch": 15.7,
|
| 420 |
+
"learning_rate": 5.8243421052631574e-05,
|
| 421 |
+
"loss": 1.187,
|
| 422 |
"step": 5400
|
| 423 |
},
|
| 424 |
{
|
| 425 |
"epoch": 15.99,
|
| 426 |
+
"learning_rate": 5.7749999999999994e-05,
|
| 427 |
+
"loss": 1.2083,
|
| 428 |
"step": 5500
|
| 429 |
},
|
| 430 |
{
|
| 431 |
"epoch": 15.99,
|
| 432 |
+
"eval_loss": 0.26375535130500793,
|
| 433 |
+
"eval_runtime": 128.7955,
|
| 434 |
+
"eval_samples_per_second": 35.871,
|
| 435 |
+
"eval_steps_per_second": 4.488,
|
| 436 |
+
"eval_wer": 0.22119561891941503,
|
| 437 |
"step": 5500
|
| 438 |
},
|
| 439 |
{
|
| 440 |
"epoch": 16.28,
|
| 441 |
+
"learning_rate": 5.725657894736842e-05,
|
| 442 |
+
"loss": 1.1987,
|
| 443 |
"step": 5600
|
| 444 |
},
|
| 445 |
{
|
| 446 |
"epoch": 16.57,
|
| 447 |
+
"learning_rate": 5.6763157894736834e-05,
|
| 448 |
+
"loss": 1.1832,
|
| 449 |
"step": 5700
|
| 450 |
},
|
| 451 |
{
|
| 452 |
"epoch": 16.86,
|
| 453 |
+
"learning_rate": 5.626973684210526e-05,
|
| 454 |
+
"loss": 1.142,
|
| 455 |
"step": 5800
|
| 456 |
},
|
| 457 |
{
|
| 458 |
"epoch": 17.15,
|
| 459 |
+
"learning_rate": 5.577631578947368e-05,
|
| 460 |
+
"loss": 1.1863,
|
| 461 |
"step": 5900
|
| 462 |
},
|
| 463 |
{
|
| 464 |
"epoch": 17.44,
|
| 465 |
+
"learning_rate": 5.52828947368421e-05,
|
| 466 |
+
"loss": 1.1606,
|
| 467 |
"step": 6000
|
| 468 |
},
|
| 469 |
{
|
| 470 |
"epoch": 17.44,
|
| 471 |
+
"eval_loss": 0.253081351518631,
|
| 472 |
+
"eval_runtime": 130.3816,
|
| 473 |
+
"eval_samples_per_second": 35.434,
|
| 474 |
+
"eval_steps_per_second": 4.433,
|
| 475 |
+
"eval_wer": 0.21740194578718716,
|
| 476 |
"step": 6000
|
| 477 |
},
|
| 478 |
{
|
| 479 |
"epoch": 17.73,
|
| 480 |
+
"learning_rate": 5.478947368421052e-05,
|
| 481 |
+
"loss": 1.154,
|
| 482 |
"step": 6100
|
| 483 |
},
|
| 484 |
{
|
| 485 |
"epoch": 18.02,
|
| 486 |
+
"learning_rate": 5.429605263157895e-05,
|
| 487 |
+
"loss": 1.1842,
|
| 488 |
"step": 6200
|
| 489 |
},
|
| 490 |
{
|
| 491 |
"epoch": 18.31,
|
| 492 |
+
"learning_rate": 5.380263157894736e-05,
|
| 493 |
+
"loss": 1.1603,
|
| 494 |
"step": 6300
|
| 495 |
},
|
| 496 |
{
|
| 497 |
"epoch": 18.6,
|
| 498 |
+
"learning_rate": 5.330921052631578e-05,
|
| 499 |
+
"loss": 1.1692,
|
| 500 |
"step": 6400
|
| 501 |
},
|
| 502 |
{
|
| 503 |
"epoch": 18.89,
|
| 504 |
+
"learning_rate": 5.281578947368421e-05,
|
| 505 |
+
"loss": 1.1545,
|
| 506 |
"step": 6500
|
| 507 |
},
|
| 508 |
{
|
| 509 |
"epoch": 18.89,
|
| 510 |
+
"eval_loss": 0.2468416541814804,
|
| 511 |
+
"eval_runtime": 130.515,
|
| 512 |
+
"eval_samples_per_second": 35.398,
|
| 513 |
+
"eval_steps_per_second": 4.429,
|
| 514 |
+
"eval_wer": 0.2108548002202778,
|
| 515 |
"step": 6500
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 19.19,
|
| 519 |
+
"learning_rate": 5.232236842105262e-05,
|
| 520 |
+
"loss": 1.1389,
|
| 521 |
"step": 6600
|
| 522 |
},
|
| 523 |
{
|
| 524 |
"epoch": 19.48,
|
| 525 |
"learning_rate": 5.182894736842105e-05,
|
| 526 |
+
"loss": 1.1635,
|
| 527 |
"step": 6700
|
| 528 |
},
|
| 529 |
{
|
| 530 |
"epoch": 19.77,
|
| 531 |
"learning_rate": 5.133552631578947e-05,
|
| 532 |
+
"loss": 1.1547,
|
| 533 |
"step": 6800
|
| 534 |
},
|
| 535 |
{
|
| 536 |
"epoch": 20.06,
|
| 537 |
"learning_rate": 5.084210526315789e-05,
|
| 538 |
+
"loss": 1.1323,
|
| 539 |
"step": 6900
|
| 540 |
},
|
| 541 |
{
|
| 542 |
"epoch": 20.35,
|
| 543 |
+
"learning_rate": 5.034868421052631e-05,
|
| 544 |
+
"loss": 1.1344,
|
| 545 |
"step": 7000
|
| 546 |
},
|
| 547 |
{
|
| 548 |
"epoch": 20.35,
|
| 549 |
+
"eval_loss": 0.2494264841079712,
|
| 550 |
+
"eval_runtime": 131.1616,
|
| 551 |
+
"eval_samples_per_second": 35.224,
|
| 552 |
+
"eval_steps_per_second": 4.407,
|
| 553 |
+
"eval_wer": 0.20504191396928348,
|
| 554 |
"step": 7000
|
| 555 |
},
|
| 556 |
{
|
| 557 |
"epoch": 20.64,
|
| 558 |
+
"learning_rate": 4.9855263157894736e-05,
|
| 559 |
+
"loss": 1.1423,
|
| 560 |
"step": 7100
|
| 561 |
},
|
| 562 |
{
|
| 563 |
"epoch": 20.93,
|
| 564 |
+
"learning_rate": 4.936184210526315e-05,
|
| 565 |
+
"loss": 1.1264,
|
| 566 |
"step": 7200
|
| 567 |
},
|
| 568 |
{
|
| 569 |
"epoch": 21.22,
|
| 570 |
+
"learning_rate": 4.886842105263158e-05,
|
| 571 |
+
"loss": 1.1362,
|
| 572 |
"step": 7300
|
| 573 |
},
|
| 574 |
{
|
| 575 |
"epoch": 21.51,
|
| 576 |
+
"learning_rate": 4.8375e-05,
|
| 577 |
+
"loss": 1.1156,
|
| 578 |
"step": 7400
|
| 579 |
},
|
| 580 |
{
|
| 581 |
"epoch": 21.8,
|
| 582 |
+
"learning_rate": 4.7881578947368424e-05,
|
| 583 |
+
"loss": 1.1173,
|
| 584 |
"step": 7500
|
| 585 |
},
|
| 586 |
{
|
| 587 |
"epoch": 21.8,
|
| 588 |
+
"eval_loss": 0.24467654526233673,
|
| 589 |
+
"eval_runtime": 161.8558,
|
| 590 |
+
"eval_samples_per_second": 28.544,
|
| 591 |
+
"eval_steps_per_second": 3.571,
|
| 592 |
+
"eval_wer": 0.19800526219176406,
|
| 593 |
"step": 7500
|
| 594 |
},
|
| 595 |
{
|
| 596 |
"epoch": 22.09,
|
| 597 |
"learning_rate": 4.739309210526315e-05,
|
| 598 |
+
"loss": 1.1319,
|
| 599 |
"step": 7600
|
| 600 |
},
|
| 601 |
{
|
| 602 |
"epoch": 22.38,
|
| 603 |
+
"learning_rate": 4.690460526315789e-05,
|
| 604 |
+
"loss": 1.1065,
|
| 605 |
"step": 7700
|
| 606 |
},
|
| 607 |
{
|
| 608 |
"epoch": 22.67,
|
| 609 |
+
"learning_rate": 4.641118421052631e-05,
|
| 610 |
+
"loss": 1.1189,
|
| 611 |
"step": 7800
|
| 612 |
},
|
| 613 |
{
|
| 614 |
"epoch": 22.96,
|
| 615 |
+
"learning_rate": 4.5917763157894736e-05,
|
| 616 |
+
"loss": 1.1204,
|
| 617 |
"step": 7900
|
| 618 |
},
|
| 619 |
{
|
| 620 |
"epoch": 23.26,
|
| 621 |
+
"learning_rate": 4.542434210526315e-05,
|
| 622 |
+
"loss": 1.1081,
|
| 623 |
"step": 8000
|
| 624 |
},
|
| 625 |
{
|
| 626 |
"epoch": 23.26,
|
| 627 |
+
"eval_loss": 0.24276763200759888,
|
| 628 |
+
"eval_runtime": 130.7165,
|
| 629 |
+
"eval_samples_per_second": 35.344,
|
| 630 |
+
"eval_steps_per_second": 4.422,
|
| 631 |
+
"eval_wer": 0.19977972220522547,
|
| 632 |
"step": 8000
|
| 633 |
},
|
| 634 |
{
|
| 635 |
"epoch": 23.55,
|
| 636 |
+
"learning_rate": 4.4930921052631576e-05,
|
| 637 |
+
"loss": 1.1097,
|
| 638 |
"step": 8100
|
| 639 |
},
|
| 640 |
{
|
| 641 |
"epoch": 23.84,
|
| 642 |
+
"learning_rate": 4.4437499999999996e-05,
|
| 643 |
+
"loss": 1.0912,
|
| 644 |
"step": 8200
|
| 645 |
},
|
| 646 |
{
|
| 647 |
"epoch": 24.13,
|
| 648 |
+
"learning_rate": 4.3944078947368416e-05,
|
| 649 |
+
"loss": 1.1305,
|
| 650 |
"step": 8300
|
| 651 |
},
|
| 652 |
{
|
| 653 |
"epoch": 24.42,
|
| 654 |
+
"learning_rate": 4.3450657894736836e-05,
|
| 655 |
+
"loss": 1.1219,
|
| 656 |
"step": 8400
|
| 657 |
},
|
| 658 |
{
|
| 659 |
"epoch": 24.71,
|
| 660 |
+
"learning_rate": 4.295723684210526e-05,
|
| 661 |
+
"loss": 1.1023,
|
| 662 |
"step": 8500
|
| 663 |
},
|
| 664 |
{
|
| 665 |
"epoch": 24.71,
|
| 666 |
+
"eval_loss": 0.23292256891727448,
|
| 667 |
+
"eval_runtime": 133.1697,
|
| 668 |
+
"eval_samples_per_second": 34.693,
|
| 669 |
+
"eval_steps_per_second": 4.34,
|
| 670 |
+
"eval_wer": 0.19512941320443003,
|
| 671 |
"step": 8500
|
| 672 |
},
|
| 673 |
{
|
| 674 |
"epoch": 25.0,
|
| 675 |
+
"learning_rate": 4.2463815789473676e-05,
|
| 676 |
+
"loss": 1.1049,
|
| 677 |
"step": 8600
|
| 678 |
},
|
| 679 |
{
|
| 680 |
"epoch": 25.29,
|
| 681 |
+
"learning_rate": 4.19703947368421e-05,
|
| 682 |
+
"loss": 1.1194,
|
| 683 |
"step": 8700
|
| 684 |
},
|
| 685 |
{
|
| 686 |
"epoch": 25.58,
|
| 687 |
+
"learning_rate": 4.1476973684210523e-05,
|
| 688 |
+
"loss": 1.1028,
|
| 689 |
"step": 8800
|
| 690 |
},
|
| 691 |
{
|
| 692 |
"epoch": 25.87,
|
| 693 |
+
"learning_rate": 4.098355263157895e-05,
|
| 694 |
+
"loss": 1.0614,
|
| 695 |
"step": 8900
|
| 696 |
},
|
| 697 |
{
|
| 698 |
"epoch": 26.16,
|
| 699 |
+
"learning_rate": 4.0490131578947364e-05,
|
| 700 |
+
"loss": 1.0923,
|
| 701 |
"step": 9000
|
| 702 |
},
|
| 703 |
{
|
| 704 |
"epoch": 26.16,
|
| 705 |
+
"eval_loss": 0.23875781893730164,
|
| 706 |
+
"eval_runtime": 132.625,
|
| 707 |
+
"eval_samples_per_second": 34.835,
|
| 708 |
+
"eval_steps_per_second": 4.358,
|
| 709 |
+
"eval_wer": 0.19616961390197638,
|
| 710 |
"step": 9000
|
| 711 |
},
|
| 712 |
{
|
| 713 |
"epoch": 26.45,
|
| 714 |
+
"learning_rate": 3.9996710526315784e-05,
|
| 715 |
+
"loss": 1.0789,
|
| 716 |
"step": 9100
|
| 717 |
},
|
| 718 |
{
|
| 719 |
"epoch": 26.74,
|
| 720 |
+
"learning_rate": 3.950328947368421e-05,
|
| 721 |
+
"loss": 1.0844,
|
| 722 |
"step": 9200
|
| 723 |
},
|
| 724 |
{
|
| 725 |
"epoch": 27.03,
|
| 726 |
+
"learning_rate": 3.9009868421052624e-05,
|
| 727 |
+
"loss": 1.0961,
|
| 728 |
"step": 9300
|
| 729 |
},
|
| 730 |
{
|
| 731 |
"epoch": 27.32,
|
| 732 |
+
"learning_rate": 3.851644736842105e-05,
|
| 733 |
+
"loss": 1.0691,
|
| 734 |
"step": 9400
|
| 735 |
},
|
| 736 |
{
|
| 737 |
"epoch": 27.61,
|
| 738 |
+
"learning_rate": 3.802302631578947e-05,
|
| 739 |
+
"loss": 1.0798,
|
| 740 |
"step": 9500
|
| 741 |
},
|
| 742 |
{
|
| 743 |
"epoch": 27.61,
|
| 744 |
+
"eval_loss": 0.23633554577827454,
|
| 745 |
+
"eval_runtime": 130.4004,
|
| 746 |
+
"eval_samples_per_second": 35.429,
|
| 747 |
+
"eval_steps_per_second": 4.433,
|
| 748 |
+
"eval_wer": 0.19442574802667809,
|
| 749 |
"step": 9500
|
| 750 |
},
|
| 751 |
{
|
| 752 |
"epoch": 27.91,
|
| 753 |
+
"learning_rate": 3.752960526315789e-05,
|
| 754 |
+
"loss": 1.0609,
|
| 755 |
"step": 9600
|
| 756 |
},
|
| 757 |
{
|
| 758 |
"epoch": 28.2,
|
| 759 |
+
"learning_rate": 3.703618421052631e-05,
|
| 760 |
+
"loss": 1.0779,
|
| 761 |
"step": 9700
|
| 762 |
},
|
| 763 |
{
|
| 764 |
"epoch": 28.49,
|
| 765 |
+
"learning_rate": 3.654276315789473e-05,
|
| 766 |
+
"loss": 1.0693,
|
| 767 |
"step": 9800
|
| 768 |
},
|
| 769 |
{
|
| 770 |
"epoch": 28.78,
|
| 771 |
+
"learning_rate": 3.604934210526316e-05,
|
| 772 |
+
"loss": 1.0786,
|
| 773 |
"step": 9900
|
| 774 |
},
|
| 775 |
{
|
| 776 |
"epoch": 29.07,
|
| 777 |
+
"learning_rate": 3.5560855263157894e-05,
|
| 778 |
+
"loss": 1.0769,
|
| 779 |
"step": 10000
|
| 780 |
},
|
| 781 |
{
|
| 782 |
"epoch": 29.07,
|
| 783 |
+
"eval_loss": 0.23424996435642242,
|
| 784 |
+
"eval_runtime": 129.7477,
|
| 785 |
+
"eval_samples_per_second": 35.608,
|
| 786 |
+
"eval_steps_per_second": 4.455,
|
| 787 |
+
"eval_wer": 0.19130514593403905,
|
| 788 |
"step": 10000
|
| 789 |
},
|
| 790 |
{
|
| 791 |
"epoch": 29.36,
|
| 792 |
+
"learning_rate": 3.5067434210526314e-05,
|
| 793 |
+
"loss": 1.0655,
|
| 794 |
"step": 10100
|
| 795 |
},
|
| 796 |
{
|
| 797 |
"epoch": 29.65,
|
| 798 |
+
"learning_rate": 3.4574013157894734e-05,
|
| 799 |
+
"loss": 1.0618,
|
| 800 |
"step": 10200
|
| 801 |
},
|
| 802 |
{
|
| 803 |
"epoch": 29.94,
|
| 804 |
+
"learning_rate": 3.4080592105263154e-05,
|
| 805 |
+
"loss": 1.071,
|
| 806 |
"step": 10300
|
| 807 |
},
|
| 808 |
{
|
| 809 |
"epoch": 30.23,
|
| 810 |
+
"learning_rate": 3.3587171052631575e-05,
|
| 811 |
+
"loss": 1.0618,
|
| 812 |
"step": 10400
|
| 813 |
},
|
| 814 |
{
|
| 815 |
"epoch": 30.52,
|
| 816 |
+
"learning_rate": 3.3093749999999995e-05,
|
| 817 |
+
"loss": 1.0672,
|
| 818 |
"step": 10500
|
| 819 |
},
|
| 820 |
{
|
| 821 |
"epoch": 30.52,
|
| 822 |
+
"eval_loss": 0.22497320175170898,
|
| 823 |
+
"eval_runtime": 129.3889,
|
| 824 |
+
"eval_samples_per_second": 35.706,
|
| 825 |
+
"eval_steps_per_second": 4.467,
|
| 826 |
+
"eval_wer": 0.18745028452548493,
|
| 827 |
"step": 10500
|
| 828 |
},
|
| 829 |
{
|
| 830 |
"epoch": 30.81,
|
| 831 |
+
"learning_rate": 3.260032894736842e-05,
|
| 832 |
+
"loss": 1.0642,
|
| 833 |
"step": 10600
|
| 834 |
},
|
| 835 |
{
|
| 836 |
"epoch": 31.1,
|
| 837 |
+
"learning_rate": 3.210690789473684e-05,
|
| 838 |
+
"loss": 1.0567,
|
| 839 |
"step": 10700
|
| 840 |
},
|
| 841 |
{
|
| 842 |
"epoch": 31.39,
|
| 843 |
+
"learning_rate": 3.161348684210526e-05,
|
| 844 |
+
"loss": 1.0566,
|
| 845 |
"step": 10800
|
| 846 |
},
|
| 847 |
{
|
| 848 |
"epoch": 31.68,
|
| 849 |
+
"learning_rate": 3.112006578947368e-05,
|
| 850 |
+
"loss": 1.0553,
|
| 851 |
"step": 10900
|
| 852 |
},
|
| 853 |
{
|
| 854 |
"epoch": 31.97,
|
| 855 |
+
"learning_rate": 3.06266447368421e-05,
|
| 856 |
+
"loss": 1.0735,
|
| 857 |
"step": 11000
|
| 858 |
},
|
| 859 |
{
|
| 860 |
"epoch": 31.97,
|
| 861 |
+
"eval_loss": 0.23052582144737244,
|
| 862 |
+
"eval_runtime": 130.8716,
|
| 863 |
+
"eval_samples_per_second": 35.302,
|
| 864 |
+
"eval_steps_per_second": 4.417,
|
| 865 |
+
"eval_wer": 0.18741969038732179,
|
| 866 |
"step": 11000
|
| 867 |
},
|
| 868 |
{
|
| 869 |
"epoch": 32.27,
|
| 870 |
+
"learning_rate": 3.0133223684210525e-05,
|
| 871 |
+
"loss": 1.0583,
|
| 872 |
"step": 11100
|
| 873 |
},
|
| 874 |
{
|
| 875 |
"epoch": 32.56,
|
| 876 |
+
"learning_rate": 2.9639802631578946e-05,
|
| 877 |
+
"loss": 1.0502,
|
| 878 |
"step": 11200
|
| 879 |
},
|
| 880 |
{
|
| 881 |
"epoch": 32.85,
|
| 882 |
+
"learning_rate": 2.9146381578947366e-05,
|
| 883 |
+
"loss": 1.0529,
|
| 884 |
"step": 11300
|
| 885 |
},
|
| 886 |
{
|
| 887 |
"epoch": 33.14,
|
| 888 |
+
"learning_rate": 2.865296052631579e-05,
|
| 889 |
+
"loss": 1.0436,
|
| 890 |
"step": 11400
|
| 891 |
},
|
| 892 |
{
|
| 893 |
"epoch": 33.43,
|
| 894 |
+
"learning_rate": 2.8159539473684206e-05,
|
| 895 |
+
"loss": 1.0628,
|
| 896 |
"step": 11500
|
| 897 |
},
|
| 898 |
{
|
| 899 |
"epoch": 33.43,
|
| 900 |
+
"eval_loss": 0.22911641001701355,
|
| 901 |
+
"eval_runtime": 129.7033,
|
| 902 |
+
"eval_samples_per_second": 35.62,
|
| 903 |
+
"eval_steps_per_second": 4.456,
|
| 904 |
+
"eval_wer": 0.18509453588692407,
|
| 905 |
"step": 11500
|
| 906 |
},
|
| 907 |
{
|
| 908 |
"epoch": 33.72,
|
| 909 |
+
"learning_rate": 2.7666118421052626e-05,
|
| 910 |
+
"loss": 1.0228,
|
| 911 |
"step": 11600
|
| 912 |
},
|
| 913 |
{
|
| 914 |
"epoch": 34.01,
|
| 915 |
+
"learning_rate": 2.717269736842105e-05,
|
| 916 |
+
"loss": 1.0556,
|
| 917 |
"step": 11700
|
| 918 |
},
|
| 919 |
{
|
| 920 |
"epoch": 34.3,
|
| 921 |
+
"learning_rate": 2.667927631578947e-05,
|
| 922 |
+
"loss": 1.0616,
|
| 923 |
"step": 11800
|
| 924 |
},
|
| 925 |
{
|
| 926 |
"epoch": 34.59,
|
| 927 |
+
"learning_rate": 2.6185855263157893e-05,
|
| 928 |
+
"loss": 1.0397,
|
| 929 |
"step": 11900
|
| 930 |
},
|
| 931 |
{
|
| 932 |
"epoch": 34.88,
|
| 933 |
+
"learning_rate": 2.5692434210526313e-05,
|
| 934 |
+
"loss": 1.0451,
|
| 935 |
"step": 12000
|
| 936 |
},
|
| 937 |
{
|
| 938 |
"epoch": 34.88,
|
| 939 |
+
"eval_loss": 0.22632543742656708,
|
| 940 |
+
"eval_runtime": 130.5561,
|
| 941 |
+
"eval_samples_per_second": 35.387,
|
| 942 |
+
"eval_steps_per_second": 4.427,
|
| 943 |
+
"eval_wer": 0.18558404209753412,
|
| 944 |
"step": 12000
|
| 945 |
},
|
| 946 |
{
|
| 947 |
"epoch": 35.17,
|
| 948 |
+
"learning_rate": 2.5199013157894733e-05,
|
| 949 |
+
"loss": 1.0387,
|
| 950 |
"step": 12100
|
| 951 |
},
|
| 952 |
{
|
| 953 |
"epoch": 35.46,
|
| 954 |
+
"learning_rate": 2.4705592105263157e-05,
|
| 955 |
+
"loss": 1.043,
|
| 956 |
"step": 12200
|
| 957 |
},
|
| 958 |
{
|
| 959 |
"epoch": 35.75,
|
| 960 |
+
"learning_rate": 2.4212171052631577e-05,
|
| 961 |
+
"loss": 1.019,
|
| 962 |
"step": 12300
|
| 963 |
},
|
| 964 |
{
|
| 965 |
"epoch": 36.05,
|
| 966 |
+
"learning_rate": 2.3718749999999997e-05,
|
| 967 |
+
"loss": 1.052,
|
| 968 |
"step": 12400
|
| 969 |
},
|
| 970 |
{
|
| 971 |
"epoch": 36.34,
|
| 972 |
+
"learning_rate": 2.322532894736842e-05,
|
| 973 |
+
"loss": 1.0299,
|
| 974 |
"step": 12500
|
| 975 |
},
|
| 976 |
{
|
| 977 |
"epoch": 36.34,
|
| 978 |
+
"eval_loss": 0.22569817304611206,
|
| 979 |
+
"eval_runtime": 131.0957,
|
| 980 |
+
"eval_samples_per_second": 35.241,
|
| 981 |
+
"eval_steps_per_second": 4.409,
|
| 982 |
+
"eval_wer": 0.1833812641497889,
|
| 983 |
"step": 12500
|
| 984 |
},
|
| 985 |
{
|
| 986 |
"epoch": 36.63,
|
| 987 |
+
"learning_rate": 2.273190789473684e-05,
|
| 988 |
+
"loss": 1.0492,
|
| 989 |
"step": 12600
|
| 990 |
},
|
| 991 |
{
|
| 992 |
"epoch": 36.92,
|
| 993 |
+
"learning_rate": 2.2238486842105264e-05,
|
| 994 |
+
"loss": 1.0405,
|
| 995 |
"step": 12700
|
| 996 |
},
|
| 997 |
{
|
| 998 |
"epoch": 37.21,
|
| 999 |
+
"learning_rate": 2.174506578947368e-05,
|
| 1000 |
+
"loss": 1.0405,
|
| 1001 |
"step": 12800
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
"epoch": 37.5,
|
| 1005 |
+
"learning_rate": 2.12516447368421e-05,
|
| 1006 |
+
"loss": 1.0186,
|
| 1007 |
"step": 12900
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
"epoch": 37.79,
|
| 1011 |
+
"learning_rate": 2.076315789473684e-05,
|
| 1012 |
+
"loss": 1.0368,
|
| 1013 |
"step": 13000
|
| 1014 |
},
|
| 1015 |
{
|
| 1016 |
"epoch": 37.79,
|
| 1017 |
+
"eval_loss": 0.2230287492275238,
|
| 1018 |
+
"eval_runtime": 141.6123,
|
| 1019 |
+
"eval_samples_per_second": 32.624,
|
| 1020 |
+
"eval_steps_per_second": 4.082,
|
| 1021 |
+
"eval_wer": 0.1808419506822493,
|
| 1022 |
"step": 13000
|
| 1023 |
},
|
| 1024 |
{
|
| 1025 |
"epoch": 38.08,
|
| 1026 |
+
"learning_rate": 2.026973684210526e-05,
|
| 1027 |
+
"loss": 1.0346,
|
| 1028 |
"step": 13100
|
| 1029 |
},
|
| 1030 |
{
|
| 1031 |
"epoch": 38.37,
|
| 1032 |
+
"learning_rate": 1.9776315789473684e-05,
|
| 1033 |
+
"loss": 1.0348,
|
| 1034 |
"step": 13200
|
| 1035 |
},
|
| 1036 |
{
|
| 1037 |
"epoch": 38.66,
|
| 1038 |
+
"learning_rate": 1.9282894736842104e-05,
|
| 1039 |
+
"loss": 1.016,
|
| 1040 |
"step": 13300
|
| 1041 |
},
|
| 1042 |
{
|
| 1043 |
"epoch": 38.95,
|
| 1044 |
+
"learning_rate": 1.8789473684210528e-05,
|
| 1045 |
+
"loss": 1.047,
|
| 1046 |
"step": 13400
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
"epoch": 39.24,
|
| 1050 |
+
"learning_rate": 1.8296052631578948e-05,
|
| 1051 |
+
"loss": 1.0322,
|
| 1052 |
"step": 13500
|
| 1053 |
},
|
| 1054 |
{
|
| 1055 |
"epoch": 39.24,
|
| 1056 |
+
"eval_loss": 0.22309929132461548,
|
| 1057 |
+
"eval_runtime": 135.3243,
|
| 1058 |
+
"eval_samples_per_second": 34.14,
|
| 1059 |
+
"eval_steps_per_second": 4.271,
|
| 1060 |
+
"eval_wer": 0.18332007587346263,
|
| 1061 |
"step": 13500
|
| 1062 |
},
|
| 1063 |
{
|
| 1064 |
"epoch": 39.53,
|
| 1065 |
+
"learning_rate": 1.7802631578947368e-05,
|
| 1066 |
+
"loss": 1.0091,
|
| 1067 |
"step": 13600
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
"epoch": 39.82,
|
| 1071 |
+
"learning_rate": 1.7309210526315788e-05,
|
| 1072 |
+
"loss": 1.0311,
|
| 1073 |
"step": 13700
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
"epoch": 40.12,
|
| 1077 |
+
"learning_rate": 1.6815789473684208e-05,
|
| 1078 |
+
"loss": 1.0436,
|
| 1079 |
"step": 13800
|
| 1080 |
},
|
| 1081 |
{
|
| 1082 |
"epoch": 40.41,
|
| 1083 |
+
"learning_rate": 1.632236842105263e-05,
|
| 1084 |
+
"loss": 1.0089,
|
| 1085 |
"step": 13900
|
| 1086 |
},
|
| 1087 |
{
|
| 1088 |
"epoch": 40.7,
|
| 1089 |
+
"learning_rate": 1.582894736842105e-05,
|
| 1090 |
+
"loss": 1.0451,
|
| 1091 |
"step": 14000
|
| 1092 |
},
|
| 1093 |
{
|
| 1094 |
"epoch": 40.7,
|
| 1095 |
+
"eval_loss": 0.2196696549654007,
|
| 1096 |
+
"eval_runtime": 131.6599,
|
| 1097 |
+
"eval_samples_per_second": 35.09,
|
| 1098 |
+
"eval_steps_per_second": 4.39,
|
| 1099 |
+
"eval_wer": 0.18172918068898,
|
| 1100 |
"step": 14000
|
| 1101 |
},
|
| 1102 |
{
|
| 1103 |
"epoch": 40.99,
|
| 1104 |
+
"learning_rate": 1.5335526315789472e-05,
|
| 1105 |
+
"loss": 1.0268,
|
| 1106 |
"step": 14100
|
| 1107 |
},
|
| 1108 |
{
|
| 1109 |
"epoch": 41.28,
|
| 1110 |
+
"learning_rate": 1.4842105263157894e-05,
|
| 1111 |
+
"loss": 1.0357,
|
| 1112 |
"step": 14200
|
| 1113 |
},
|
| 1114 |
{
|
| 1115 |
"epoch": 41.57,
|
| 1116 |
+
"learning_rate": 1.4348684210526315e-05,
|
| 1117 |
+
"loss": 1.0211,
|
| 1118 |
"step": 14300
|
| 1119 |
},
|
| 1120 |
{
|
| 1121 |
"epoch": 41.86,
|
| 1122 |
+
"learning_rate": 1.3855263157894737e-05,
|
| 1123 |
+
"loss": 0.9896,
|
| 1124 |
"step": 14400
|
| 1125 |
},
|
| 1126 |
{
|
| 1127 |
"epoch": 42.15,
|
| 1128 |
+
"learning_rate": 1.3361842105263156e-05,
|
| 1129 |
+
"loss": 1.0304,
|
| 1130 |
"step": 14500
|
| 1131 |
},
|
| 1132 |
{
|
| 1133 |
"epoch": 42.15,
|
| 1134 |
+
"eval_loss": 0.22407057881355286,
|
| 1135 |
+
"eval_runtime": 130.69,
|
| 1136 |
+
"eval_samples_per_second": 35.351,
|
| 1137 |
+
"eval_steps_per_second": 4.423,
|
| 1138 |
+
"eval_wer": 0.18127026861653306,
|
| 1139 |
"step": 14500
|
| 1140 |
},
|
| 1141 |
{
|
| 1142 |
"epoch": 42.44,
|
| 1143 |
+
"learning_rate": 1.2868421052631577e-05,
|
| 1144 |
+
"loss": 1.0092,
|
| 1145 |
"step": 14600
|
| 1146 |
},
|
| 1147 |
{
|
| 1148 |
"epoch": 42.73,
|
| 1149 |
+
"learning_rate": 1.2375e-05,
|
| 1150 |
+
"loss": 0.9983,
|
| 1151 |
"step": 14700
|
| 1152 |
},
|
| 1153 |
{
|
| 1154 |
"epoch": 43.02,
|
| 1155 |
+
"learning_rate": 1.188157894736842e-05,
|
| 1156 |
+
"loss": 1.0297,
|
| 1157 |
"step": 14800
|
| 1158 |
},
|
| 1159 |
{
|
| 1160 |
"epoch": 43.31,
|
| 1161 |
+
"learning_rate": 1.1388157894736841e-05,
|
| 1162 |
+
"loss": 1.0079,
|
| 1163 |
"step": 14900
|
| 1164 |
},
|
| 1165 |
{
|
| 1166 |
"epoch": 43.6,
|
| 1167 |
+
"learning_rate": 1.0894736842105263e-05,
|
| 1168 |
+
"loss": 1.0102,
|
| 1169 |
"step": 15000
|
| 1170 |
},
|
| 1171 |
{
|
| 1172 |
"epoch": 43.6,
|
| 1173 |
+
"eval_loss": 0.22326770424842834,
|
| 1174 |
+
"eval_runtime": 129.9475,
|
| 1175 |
+
"eval_samples_per_second": 35.553,
|
| 1176 |
+
"eval_steps_per_second": 4.448,
|
| 1177 |
+
"eval_wer": 0.17949580860307165,
|
| 1178 |
"step": 15000
|
| 1179 |
},
|
| 1180 |
{
|
| 1181 |
"epoch": 43.89,
|
| 1182 |
+
"learning_rate": 1.0401315789473685e-05,
|
| 1183 |
+
"loss": 1.0303,
|
| 1184 |
"step": 15100
|
| 1185 |
},
|
| 1186 |
{
|
| 1187 |
"epoch": 44.19,
|
| 1188 |
+
"learning_rate": 9.912828947368419e-06,
|
| 1189 |
+
"loss": 1.0099,
|
| 1190 |
"step": 15200
|
| 1191 |
},
|
| 1192 |
{
|
| 1193 |
"epoch": 44.48,
|
| 1194 |
+
"learning_rate": 9.41940789473684e-06,
|
| 1195 |
+
"loss": 1.0098,
|
| 1196 |
"step": 15300
|
| 1197 |
},
|
| 1198 |
{
|
| 1199 |
"epoch": 44.77,
|
| 1200 |
+
"learning_rate": 8.925986842105263e-06,
|
| 1201 |
+
"loss": 1.0068,
|
| 1202 |
"step": 15400
|
| 1203 |
},
|
| 1204 |
{
|
| 1205 |
"epoch": 45.06,
|
| 1206 |
+
"learning_rate": 8.432565789473683e-06,
|
| 1207 |
+
"loss": 1.0135,
|
| 1208 |
"step": 15500
|
| 1209 |
},
|
| 1210 |
{
|
| 1211 |
"epoch": 45.06,
|
| 1212 |
+
"eval_loss": 0.2199595868587494,
|
| 1213 |
+
"eval_runtime": 129.6997,
|
| 1214 |
+
"eval_samples_per_second": 35.621,
|
| 1215 |
+
"eval_steps_per_second": 4.456,
|
| 1216 |
+
"eval_wer": 0.1794346203267454,
|
| 1217 |
"step": 15500
|
| 1218 |
},
|
| 1219 |
{
|
| 1220 |
"epoch": 45.35,
|
| 1221 |
+
"learning_rate": 7.939144736842104e-06,
|
| 1222 |
+
"loss": 1.0279,
|
| 1223 |
"step": 15600
|
| 1224 |
},
|
| 1225 |
{
|
| 1226 |
"epoch": 45.64,
|
| 1227 |
+
"learning_rate": 7.4457236842105255e-06,
|
| 1228 |
+
"loss": 1.0116,
|
| 1229 |
"step": 15700
|
| 1230 |
},
|
| 1231 |
{
|
| 1232 |
"epoch": 45.93,
|
| 1233 |
+
"learning_rate": 6.952302631578946e-06,
|
| 1234 |
+
"loss": 1.0106,
|
| 1235 |
"step": 15800
|
| 1236 |
},
|
| 1237 |
{
|
| 1238 |
"epoch": 46.22,
|
| 1239 |
+
"learning_rate": 6.458881578947368e-06,
|
| 1240 |
+
"loss": 1.0132,
|
| 1241 |
"step": 15900
|
| 1242 |
},
|
| 1243 |
{
|
| 1244 |
"epoch": 46.51,
|
| 1245 |
+
"learning_rate": 5.965460526315788e-06,
|
| 1246 |
+
"loss": 1.014,
|
| 1247 |
"step": 16000
|
| 1248 |
},
|
| 1249 |
{
|
| 1250 |
"epoch": 46.51,
|
| 1251 |
+
"eval_loss": 0.22073601186275482,
|
| 1252 |
+
"eval_runtime": 130.9629,
|
| 1253 |
+
"eval_samples_per_second": 35.277,
|
| 1254 |
+
"eval_steps_per_second": 4.413,
|
| 1255 |
+
"eval_wer": 0.177904913418589,
|
| 1256 |
"step": 16000
|
| 1257 |
},
|
| 1258 |
{
|
| 1259 |
"epoch": 46.8,
|
| 1260 |
+
"learning_rate": 5.47203947368421e-06,
|
| 1261 |
+
"loss": 1.0231,
|
| 1262 |
"step": 16100
|
| 1263 |
},
|
| 1264 |
{
|
| 1265 |
"epoch": 47.09,
|
| 1266 |
+
"learning_rate": 4.978618421052631e-06,
|
| 1267 |
+
"loss": 1.0161,
|
| 1268 |
"step": 16200
|
| 1269 |
},
|
| 1270 |
{
|
| 1271 |
"epoch": 47.38,
|
| 1272 |
+
"learning_rate": 4.485197368421052e-06,
|
| 1273 |
+
"loss": 0.9842,
|
| 1274 |
"step": 16300
|
| 1275 |
},
|
| 1276 |
{
|
| 1277 |
"epoch": 47.67,
|
| 1278 |
+
"learning_rate": 3.991776315789474e-06,
|
| 1279 |
+
"loss": 0.9901,
|
| 1280 |
"step": 16400
|
| 1281 |
},
|
| 1282 |
{
|
| 1283 |
"epoch": 47.96,
|
| 1284 |
+
"learning_rate": 3.498355263157895e-06,
|
| 1285 |
+
"loss": 1.0071,
|
| 1286 |
"step": 16500
|
| 1287 |
},
|
| 1288 |
{
|
| 1289 |
"epoch": 47.96,
|
| 1290 |
+
"eval_loss": 0.2204526960849762,
|
| 1291 |
+
"eval_runtime": 130.0967,
|
| 1292 |
+
"eval_samples_per_second": 35.512,
|
| 1293 |
+
"eval_steps_per_second": 4.443,
|
| 1294 |
+
"eval_wer": 0.1783638254910359,
|
| 1295 |
"step": 16500
|
| 1296 |
},
|
| 1297 |
{
|
| 1298 |
"epoch": 48.26,
|
| 1299 |
+
"learning_rate": 3.0049342105263154e-06,
|
| 1300 |
+
"loss": 1.0114,
|
| 1301 |
"step": 16600
|
| 1302 |
},
|
| 1303 |
{
|
| 1304 |
"epoch": 48.55,
|
| 1305 |
+
"learning_rate": 2.5115131578947363e-06,
|
| 1306 |
+
"loss": 1.0035,
|
| 1307 |
"step": 16700
|
| 1308 |
},
|
| 1309 |
{
|
| 1310 |
"epoch": 48.84,
|
| 1311 |
+
"learning_rate": 2.0180921052631577e-06,
|
| 1312 |
+
"loss": 1.0091,
|
| 1313 |
"step": 16800
|
| 1314 |
},
|
| 1315 |
{
|
| 1316 |
"epoch": 49.13,
|
| 1317 |
+
"learning_rate": 1.524671052631579e-06,
|
| 1318 |
+
"loss": 1.0022,
|
| 1319 |
"step": 16900
|
| 1320 |
},
|
| 1321 |
{
|
| 1322 |
"epoch": 49.42,
|
| 1323 |
+
"learning_rate": 1.0312499999999999e-06,
|
| 1324 |
+
"loss": 0.9729,
|
| 1325 |
"step": 17000
|
| 1326 |
},
|
| 1327 |
{
|
| 1328 |
"epoch": 49.42,
|
| 1329 |
+
"eval_loss": 0.22042310237884521,
|
| 1330 |
+
"eval_runtime": 130.006,
|
| 1331 |
+
"eval_samples_per_second": 35.537,
|
| 1332 |
+
"eval_steps_per_second": 4.446,
|
| 1333 |
+
"eval_wer": 0.17766016031328397,
|
| 1334 |
"step": 17000
|
| 1335 |
},
|
| 1336 |
{
|
| 1337 |
"epoch": 49.71,
|
| 1338 |
+
"learning_rate": 5.378289473684209e-07,
|
| 1339 |
+
"loss": 1.0161,
|
| 1340 |
"step": 17100
|
| 1341 |
},
|
| 1342 |
{
|
| 1343 |
"epoch": 50.0,
|
| 1344 |
+
"learning_rate": 4.9342105263157887e-08,
|
| 1345 |
+
"loss": 1.0062,
|
| 1346 |
"step": 17200
|
| 1347 |
},
|
| 1348 |
{
|
| 1349 |
"epoch": 50.0,
|
| 1350 |
"step": 17200,
|
| 1351 |
"total_flos": 5.074861809449127e+19,
|
| 1352 |
+
"train_loss": 1.4085150359397711,
|
| 1353 |
+
"train_runtime": 29405.2698,
|
| 1354 |
+
"train_samples_per_second": 18.755,
|
| 1355 |
+
"train_steps_per_second": 0.585
|
| 1356 |
}
|
| 1357 |
],
|
| 1358 |
"max_steps": 17200,
|