End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 1.9438238143920898,
|
| 4 |
+
"eval_runtime": 45.4177,
|
| 5 |
+
"eval_samples": 2933,
|
| 6 |
+
"eval_samples_per_second": 64.578,
|
| 7 |
+
"eval_steps_per_second": 8.081,
|
| 8 |
+
"perplexity": 6.985410881541159,
|
| 9 |
+
"train_loss": 1.9284902631757455,
|
| 10 |
+
"train_runtime": 49525.3053,
|
| 11 |
+
"train_samples": 55736,
|
| 12 |
+
"train_samples_per_second": 16.881,
|
| 13 |
+
"train_steps_per_second": 0.264
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 1.9438238143920898,
|
| 4 |
+
"eval_runtime": 45.4177,
|
| 5 |
+
"eval_samples": 2933,
|
| 6 |
+
"eval_samples_per_second": 64.578,
|
| 7 |
+
"eval_steps_per_second": 8.081,
|
| 8 |
+
"perplexity": 6.985410881541159
|
| 9 |
+
}
|
runs/Dec04_08-11-51_jbuhmann/events.out.tfevents.1638605539.jbuhmann.21053.0
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 sha256:97cd5d688ce8f8eeff689dd30bfbd7b565ed89c48c6770f454910caaeb42d440
|
| 3 |
+
size 11669
|
runs/Dec04_08-11-51_jbuhmann/events.out.tfevents.1638655111.jbuhmann.21053.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b9d8ee390a6333fbbef640c12ca2b32a36a0563f6045f2eb88c5dec8c52d94fa
|
| 3 |
+
size 311
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 1.9284902631757455,
|
| 4 |
+
"train_runtime": 49525.3053,
|
| 5 |
+
"train_samples": 55736,
|
| 6 |
+
"train_samples_per_second": 16.881,
|
| 7 |
+
"train_steps_per_second": 0.264
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,301 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"global_step": 13065,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.57,
|
| 12 |
+
"learning_rate": 1.9234596249521625e-05,
|
| 13 |
+
"loss": 2.2122,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_loss": 2.0468506813049316,
|
| 19 |
+
"eval_runtime": 45.9615,
|
| 20 |
+
"eval_samples_per_second": 63.814,
|
| 21 |
+
"eval_steps_per_second": 7.985,
|
| 22 |
+
"step": 871
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 1.15,
|
| 26 |
+
"learning_rate": 1.8469192499043248e-05,
|
| 27 |
+
"loss": 2.1507,
|
| 28 |
+
"step": 1000
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 1.72,
|
| 32 |
+
"learning_rate": 1.770378874856487e-05,
|
| 33 |
+
"loss": 2.0961,
|
| 34 |
+
"step": 1500
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"eval_loss": 2.0117030143737793,
|
| 39 |
+
"eval_runtime": 45.9973,
|
| 40 |
+
"eval_samples_per_second": 63.765,
|
| 41 |
+
"eval_steps_per_second": 7.979,
|
| 42 |
+
"step": 1742
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 2.3,
|
| 46 |
+
"learning_rate": 1.693838499808649e-05,
|
| 47 |
+
"loss": 2.0572,
|
| 48 |
+
"step": 2000
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 2.87,
|
| 52 |
+
"learning_rate": 1.6172981247608114e-05,
|
| 53 |
+
"loss": 2.0628,
|
| 54 |
+
"step": 2500
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.0,
|
| 58 |
+
"eval_loss": 2.004037380218506,
|
| 59 |
+
"eval_runtime": 45.987,
|
| 60 |
+
"eval_samples_per_second": 63.779,
|
| 61 |
+
"eval_steps_per_second": 7.981,
|
| 62 |
+
"step": 2613
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 3.44,
|
| 66 |
+
"learning_rate": 1.5407577497129737e-05,
|
| 67 |
+
"loss": 2.0173,
|
| 68 |
+
"step": 3000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 4.0,
|
| 72 |
+
"eval_loss": 1.990085244178772,
|
| 73 |
+
"eval_runtime": 45.9573,
|
| 74 |
+
"eval_samples_per_second": 63.82,
|
| 75 |
+
"eval_steps_per_second": 7.986,
|
| 76 |
+
"step": 3484
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 4.02,
|
| 80 |
+
"learning_rate": 1.464217374665136e-05,
|
| 81 |
+
"loss": 2.0011,
|
| 82 |
+
"step": 3500
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 4.59,
|
| 86 |
+
"learning_rate": 1.3876769996172983e-05,
|
| 87 |
+
"loss": 1.9772,
|
| 88 |
+
"step": 4000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 5.0,
|
| 92 |
+
"eval_loss": 1.9711157083511353,
|
| 93 |
+
"eval_runtime": 45.9361,
|
| 94 |
+
"eval_samples_per_second": 63.85,
|
| 95 |
+
"eval_steps_per_second": 7.989,
|
| 96 |
+
"step": 4355
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 5.17,
|
| 100 |
+
"learning_rate": 1.3111366245694605e-05,
|
| 101 |
+
"loss": 1.9615,
|
| 102 |
+
"step": 4500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 5.74,
|
| 106 |
+
"learning_rate": 1.2345962495216228e-05,
|
| 107 |
+
"loss": 1.9455,
|
| 108 |
+
"step": 5000
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 6.0,
|
| 112 |
+
"eval_loss": 1.9784963130950928,
|
| 113 |
+
"eval_runtime": 45.9339,
|
| 114 |
+
"eval_samples_per_second": 63.853,
|
| 115 |
+
"eval_steps_per_second": 7.99,
|
| 116 |
+
"step": 5226
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 6.31,
|
| 120 |
+
"learning_rate": 1.1580558744737851e-05,
|
| 121 |
+
"loss": 1.9343,
|
| 122 |
+
"step": 5500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 6.89,
|
| 126 |
+
"learning_rate": 1.0815154994259474e-05,
|
| 127 |
+
"loss": 1.917,
|
| 128 |
+
"step": 6000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 7.0,
|
| 132 |
+
"eval_loss": 1.9379661083221436,
|
| 133 |
+
"eval_runtime": 45.9382,
|
| 134 |
+
"eval_samples_per_second": 63.847,
|
| 135 |
+
"eval_steps_per_second": 7.989,
|
| 136 |
+
"step": 6097
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 7.46,
|
| 140 |
+
"learning_rate": 1.0049751243781096e-05,
|
| 141 |
+
"loss": 1.8933,
|
| 142 |
+
"step": 6500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 8.0,
|
| 146 |
+
"eval_loss": 1.9650969505310059,
|
| 147 |
+
"eval_runtime": 45.9336,
|
| 148 |
+
"eval_samples_per_second": 63.853,
|
| 149 |
+
"eval_steps_per_second": 7.99,
|
| 150 |
+
"step": 6968
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 8.04,
|
| 154 |
+
"learning_rate": 9.284347493302717e-06,
|
| 155 |
+
"loss": 1.8992,
|
| 156 |
+
"step": 7000
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 8.61,
|
| 160 |
+
"learning_rate": 8.51894374282434e-06,
|
| 161 |
+
"loss": 1.8708,
|
| 162 |
+
"step": 7500
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 9.0,
|
| 166 |
+
"eval_loss": 1.9915274381637573,
|
| 167 |
+
"eval_runtime": 45.9645,
|
| 168 |
+
"eval_samples_per_second": 63.81,
|
| 169 |
+
"eval_steps_per_second": 7.984,
|
| 170 |
+
"step": 7839
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 9.18,
|
| 174 |
+
"learning_rate": 7.753539992345964e-06,
|
| 175 |
+
"loss": 1.8698,
|
| 176 |
+
"step": 8000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 9.76,
|
| 180 |
+
"learning_rate": 6.988136241867586e-06,
|
| 181 |
+
"loss": 1.862,
|
| 182 |
+
"step": 8500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 10.0,
|
| 186 |
+
"eval_loss": 1.9309507608413696,
|
| 187 |
+
"eval_runtime": 45.9471,
|
| 188 |
+
"eval_samples_per_second": 63.834,
|
| 189 |
+
"eval_steps_per_second": 7.987,
|
| 190 |
+
"step": 8710
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 10.33,
|
| 194 |
+
"learning_rate": 6.222732491389208e-06,
|
| 195 |
+
"loss": 1.862,
|
| 196 |
+
"step": 9000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 10.91,
|
| 200 |
+
"learning_rate": 5.457328740910831e-06,
|
| 201 |
+
"loss": 1.8545,
|
| 202 |
+
"step": 9500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 11.0,
|
| 206 |
+
"eval_loss": 1.9422342777252197,
|
| 207 |
+
"eval_runtime": 45.9437,
|
| 208 |
+
"eval_samples_per_second": 63.839,
|
| 209 |
+
"eval_steps_per_second": 7.988,
|
| 210 |
+
"step": 9581
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 11.48,
|
| 214 |
+
"learning_rate": 4.691924990432454e-06,
|
| 215 |
+
"loss": 1.8231,
|
| 216 |
+
"step": 10000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 12.0,
|
| 220 |
+
"eval_loss": 1.931045651435852,
|
| 221 |
+
"eval_runtime": 45.9553,
|
| 222 |
+
"eval_samples_per_second": 63.823,
|
| 223 |
+
"eval_steps_per_second": 7.986,
|
| 224 |
+
"step": 10452
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 12.06,
|
| 228 |
+
"learning_rate": 3.926521239954076e-06,
|
| 229 |
+
"loss": 1.8175,
|
| 230 |
+
"step": 10500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 12.63,
|
| 234 |
+
"learning_rate": 3.1611174894756987e-06,
|
| 235 |
+
"loss": 1.8141,
|
| 236 |
+
"step": 11000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 13.0,
|
| 240 |
+
"eval_loss": 1.936198115348816,
|
| 241 |
+
"eval_runtime": 45.947,
|
| 242 |
+
"eval_samples_per_second": 63.834,
|
| 243 |
+
"eval_steps_per_second": 7.987,
|
| 244 |
+
"step": 11323
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 13.2,
|
| 248 |
+
"learning_rate": 2.395713738997321e-06,
|
| 249 |
+
"loss": 1.84,
|
| 250 |
+
"step": 11500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 13.78,
|
| 254 |
+
"learning_rate": 1.630309988518944e-06,
|
| 255 |
+
"loss": 1.7939,
|
| 256 |
+
"step": 12000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 14.0,
|
| 260 |
+
"eval_loss": 1.9334322214126587,
|
| 261 |
+
"eval_runtime": 45.9955,
|
| 262 |
+
"eval_samples_per_second": 63.767,
|
| 263 |
+
"eval_steps_per_second": 7.979,
|
| 264 |
+
"step": 12194
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 14.35,
|
| 268 |
+
"learning_rate": 8.649062380405665e-07,
|
| 269 |
+
"loss": 1.8165,
|
| 270 |
+
"step": 12500
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 14.93,
|
| 274 |
+
"learning_rate": 9.950248756218906e-08,
|
| 275 |
+
"loss": 1.8035,
|
| 276 |
+
"step": 13000
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 15.0,
|
| 280 |
+
"eval_loss": 1.9196784496307373,
|
| 281 |
+
"eval_runtime": 45.9399,
|
| 282 |
+
"eval_samples_per_second": 63.844,
|
| 283 |
+
"eval_steps_per_second": 7.989,
|
| 284 |
+
"step": 13065
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 15.0,
|
| 288 |
+
"step": 13065,
|
| 289 |
+
"total_flos": 2.2004859851993088e+17,
|
| 290 |
+
"train_loss": 1.9284902631757455,
|
| 291 |
+
"train_runtime": 49525.3053,
|
| 292 |
+
"train_samples_per_second": 16.881,
|
| 293 |
+
"train_steps_per_second": 0.264
|
| 294 |
+
}
|
| 295 |
+
],
|
| 296 |
+
"max_steps": 13065,
|
| 297 |
+
"num_train_epochs": 15,
|
| 298 |
+
"total_flos": 2.2004859851993088e+17,
|
| 299 |
+
"trial_name": null,
|
| 300 |
+
"trial_params": null
|
| 301 |
+
}
|