End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.11,
|
| 3 |
+
"eval_bleu": 15.1824,
|
| 4 |
+
"eval_gen_len": 56.1138,
|
| 5 |
+
"eval_loss": 1.5099098682403564,
|
| 6 |
+
"eval_runtime": 39547.6625,
|
| 7 |
+
"eval_samples": 1894096,
|
| 8 |
+
"eval_samples_per_second": 47.894,
|
| 9 |
+
"eval_steps_per_second": 0.374,
|
| 10 |
+
"train_loss": 1.7310921798706054,
|
| 11 |
+
"train_runtime": 38945.4898,
|
| 12 |
+
"train_samples": 13168780,
|
| 13 |
+
"train_samples_per_second": 1051.726,
|
| 14 |
+
"train_steps_per_second": 0.257
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.11,
|
| 3 |
+
"eval_bleu": 15.1824,
|
| 4 |
+
"eval_gen_len": 56.1138,
|
| 5 |
+
"eval_loss": 1.5099098682403564,
|
| 6 |
+
"eval_runtime": 39547.6625,
|
| 7 |
+
"eval_samples": 1894096,
|
| 8 |
+
"eval_samples_per_second": 47.894,
|
| 9 |
+
"eval_steps_per_second": 0.374
|
| 10 |
+
}
|
runs/Mar28_10-56-04_bookbot-torch/events.out.tfevents.1680080453.bookbot-torch.24245.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:489735bc9f55fd1fa5d1a73f761df0507d3f673921c54a087f4e5a3156aa1afc
|
| 3 |
+
size 458
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.11,
|
| 3 |
+
"train_loss": 1.7310921798706054,
|
| 4 |
+
"train_runtime": 38945.4898,
|
| 5 |
+
"train_samples": 13168780,
|
| 6 |
+
"train_samples_per_second": 1051.726,
|
| 7 |
+
"train_steps_per_second": 0.257
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,345 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.1103594409128905,
|
| 5 |
+
"global_step": 10000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.06,
|
| 12 |
+
"learning_rate": 0.00016,
|
| 13 |
+
"loss": 8.4604,
|
| 14 |
+
"step": 200
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.12,
|
| 18 |
+
"learning_rate": 0.00032,
|
| 19 |
+
"loss": 4.2921,
|
| 20 |
+
"step": 400
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.19,
|
| 24 |
+
"learning_rate": 0.00048,
|
| 25 |
+
"loss": 3.3774,
|
| 26 |
+
"step": 600
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.25,
|
| 30 |
+
"learning_rate": 0.00064,
|
| 31 |
+
"loss": 2.9308,
|
| 32 |
+
"step": 800
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.31,
|
| 36 |
+
"learning_rate": 0.0008,
|
| 37 |
+
"loss": 2.641,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.37,
|
| 42 |
+
"learning_rate": 0.0007822222222222222,
|
| 43 |
+
"loss": 2.4213,
|
| 44 |
+
"step": 1200
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.44,
|
| 48 |
+
"learning_rate": 0.0007644444444444445,
|
| 49 |
+
"loss": 2.2593,
|
| 50 |
+
"step": 1400
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.5,
|
| 54 |
+
"learning_rate": 0.0007466666666666667,
|
| 55 |
+
"loss": 2.1346,
|
| 56 |
+
"step": 1600
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.56,
|
| 60 |
+
"learning_rate": 0.0007288888888888889,
|
| 61 |
+
"loss": 2.0382,
|
| 62 |
+
"step": 1800
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.62,
|
| 66 |
+
"learning_rate": 0.0007111111111111111,
|
| 67 |
+
"loss": 1.955,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.68,
|
| 72 |
+
"learning_rate": 0.0006933333333333333,
|
| 73 |
+
"loss": 1.8814,
|
| 74 |
+
"step": 2200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.75,
|
| 78 |
+
"learning_rate": 0.0006755555555555555,
|
| 79 |
+
"loss": 1.8179,
|
| 80 |
+
"step": 2400
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.81,
|
| 84 |
+
"learning_rate": 0.0006577777777777777,
|
| 85 |
+
"loss": 1.7623,
|
| 86 |
+
"step": 2600
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.87,
|
| 90 |
+
"learning_rate": 0.00064,
|
| 91 |
+
"loss": 1.7135,
|
| 92 |
+
"step": 2800
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.93,
|
| 96 |
+
"learning_rate": 0.0006222222222222223,
|
| 97 |
+
"loss": 1.6689,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.0,
|
| 102 |
+
"learning_rate": 0.0006044444444444445,
|
| 103 |
+
"loss": 1.6295,
|
| 104 |
+
"step": 3200
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.06,
|
| 108 |
+
"learning_rate": 0.0005866666666666667,
|
| 109 |
+
"loss": 1.5857,
|
| 110 |
+
"step": 3400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.12,
|
| 114 |
+
"learning_rate": 0.000568888888888889,
|
| 115 |
+
"loss": 1.5541,
|
| 116 |
+
"step": 3600
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.18,
|
| 120 |
+
"learning_rate": 0.0005511111111111112,
|
| 121 |
+
"loss": 1.5221,
|
| 122 |
+
"step": 3800
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.24,
|
| 126 |
+
"learning_rate": 0.0005333333333333334,
|
| 127 |
+
"loss": 1.4947,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.31,
|
| 132 |
+
"learning_rate": 0.0005155555555555557,
|
| 133 |
+
"loss": 1.4683,
|
| 134 |
+
"step": 4200
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.37,
|
| 138 |
+
"learning_rate": 0.0004977777777777778,
|
| 139 |
+
"loss": 1.4415,
|
| 140 |
+
"step": 4400
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.43,
|
| 144 |
+
"learning_rate": 0.00048,
|
| 145 |
+
"loss": 1.4183,
|
| 146 |
+
"step": 4600
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.49,
|
| 150 |
+
"learning_rate": 0.0004622222222222222,
|
| 151 |
+
"loss": 1.3975,
|
| 152 |
+
"step": 4800
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.56,
|
| 156 |
+
"learning_rate": 0.00044444444444444447,
|
| 157 |
+
"loss": 1.3756,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.56,
|
| 162 |
+
"eval_bleu": 2.3992,
|
| 163 |
+
"eval_gen_len": 18.9403,
|
| 164 |
+
"eval_loss": 1.535530686378479,
|
| 165 |
+
"eval_runtime": 7626.9892,
|
| 166 |
+
"eval_samples_per_second": 248.341,
|
| 167 |
+
"eval_steps_per_second": 1.94,
|
| 168 |
+
"step": 5000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.62,
|
| 172 |
+
"learning_rate": 0.00042666666666666667,
|
| 173 |
+
"loss": 1.358,
|
| 174 |
+
"step": 5200
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.68,
|
| 178 |
+
"learning_rate": 0.00040888888888888887,
|
| 179 |
+
"loss": 1.339,
|
| 180 |
+
"step": 5400
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.74,
|
| 184 |
+
"learning_rate": 0.0003911111111111111,
|
| 185 |
+
"loss": 1.3213,
|
| 186 |
+
"step": 5600
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.8,
|
| 190 |
+
"learning_rate": 0.0003733333333333334,
|
| 191 |
+
"loss": 1.3068,
|
| 192 |
+
"step": 5800
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.87,
|
| 196 |
+
"learning_rate": 0.00035555555555555557,
|
| 197 |
+
"loss": 1.2927,
|
| 198 |
+
"step": 6000
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.93,
|
| 202 |
+
"learning_rate": 0.00033777777777777777,
|
| 203 |
+
"loss": 1.279,
|
| 204 |
+
"step": 6200
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 1.99,
|
| 208 |
+
"learning_rate": 0.00032,
|
| 209 |
+
"loss": 1.2641,
|
| 210 |
+
"step": 6400
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.05,
|
| 214 |
+
"learning_rate": 0.0003022222222222222,
|
| 215 |
+
"loss": 1.2455,
|
| 216 |
+
"step": 6600
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.12,
|
| 220 |
+
"learning_rate": 0.0002844444444444445,
|
| 221 |
+
"loss": 1.2329,
|
| 222 |
+
"step": 6800
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.18,
|
| 226 |
+
"learning_rate": 0.0002666666666666667,
|
| 227 |
+
"loss": 1.2239,
|
| 228 |
+
"step": 7000
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.24,
|
| 232 |
+
"learning_rate": 0.0002488888888888889,
|
| 233 |
+
"loss": 1.2123,
|
| 234 |
+
"step": 7200
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.3,
|
| 238 |
+
"learning_rate": 0.0002311111111111111,
|
| 239 |
+
"loss": 1.2036,
|
| 240 |
+
"step": 7400
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.36,
|
| 244 |
+
"learning_rate": 0.00021333333333333333,
|
| 245 |
+
"loss": 1.1937,
|
| 246 |
+
"step": 7600
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.43,
|
| 250 |
+
"learning_rate": 0.00019555555555555556,
|
| 251 |
+
"loss": 1.1851,
|
| 252 |
+
"step": 7800
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 2.49,
|
| 256 |
+
"learning_rate": 0.00017777777777777779,
|
| 257 |
+
"loss": 1.1772,
|
| 258 |
+
"step": 8000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.55,
|
| 262 |
+
"learning_rate": 0.00016,
|
| 263 |
+
"loss": 1.1708,
|
| 264 |
+
"step": 8200
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.61,
|
| 268 |
+
"learning_rate": 0.00014222222222222224,
|
| 269 |
+
"loss": 1.1634,
|
| 270 |
+
"step": 8400
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.67,
|
| 274 |
+
"learning_rate": 0.00012444444444444444,
|
| 275 |
+
"loss": 1.158,
|
| 276 |
+
"step": 8600
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 2.74,
|
| 280 |
+
"learning_rate": 0.00010666666666666667,
|
| 281 |
+
"loss": 1.1541,
|
| 282 |
+
"step": 8800
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 2.8,
|
| 286 |
+
"learning_rate": 8.888888888888889e-05,
|
| 287 |
+
"loss": 1.1494,
|
| 288 |
+
"step": 9000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.86,
|
| 292 |
+
"learning_rate": 7.111111111111112e-05,
|
| 293 |
+
"loss": 1.1435,
|
| 294 |
+
"step": 9200
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 2.92,
|
| 298 |
+
"learning_rate": 5.333333333333333e-05,
|
| 299 |
+
"loss": 1.1406,
|
| 300 |
+
"step": 9400
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 2.99,
|
| 304 |
+
"learning_rate": 3.555555555555556e-05,
|
| 305 |
+
"loss": 1.1375,
|
| 306 |
+
"step": 9600
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 3.05,
|
| 310 |
+
"learning_rate": 1.777777777777778e-05,
|
| 311 |
+
"loss": 1.1315,
|
| 312 |
+
"step": 9800
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 3.11,
|
| 316 |
+
"learning_rate": 0.0,
|
| 317 |
+
"loss": 1.1295,
|
| 318 |
+
"step": 10000
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 3.11,
|
| 322 |
+
"eval_bleu": 2.543,
|
| 323 |
+
"eval_gen_len": 18.9389,
|
| 324 |
+
"eval_loss": 1.5099098682403564,
|
| 325 |
+
"eval_runtime": 7653.3773,
|
| 326 |
+
"eval_samples_per_second": 247.485,
|
| 327 |
+
"eval_steps_per_second": 1.934,
|
| 328 |
+
"step": 10000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 3.11,
|
| 332 |
+
"step": 10000,
|
| 333 |
+
"total_flos": 1.2278131895086088e+19,
|
| 334 |
+
"train_loss": 1.7310921798706054,
|
| 335 |
+
"train_runtime": 38945.4898,
|
| 336 |
+
"train_samples_per_second": 1051.726,
|
| 337 |
+
"train_steps_per_second": 0.257
|
| 338 |
+
}
|
| 339 |
+
],
|
| 340 |
+
"max_steps": 10000,
|
| 341 |
+
"num_train_epochs": 4,
|
| 342 |
+
"total_flos": 1.2278131895086088e+19,
|
| 343 |
+
"trial_name": null,
|
| 344 |
+
"trial_params": null
|
| 345 |
+
}
|