End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"eval_loss": 2.015979766845703,
|
| 4 |
+
"eval_runtime": 80.6758,
|
| 5 |
+
"eval_samples": 5193,
|
| 6 |
+
"eval_samples_per_second": 64.369,
|
| 7 |
+
"eval_steps_per_second": 8.057,
|
| 8 |
+
"perplexity": 7.508079946521416,
|
| 9 |
+
"train_loss": 2.304505779894719,
|
| 10 |
+
"train_runtime": 46564.1697,
|
| 11 |
+
"train_samples": 98673,
|
| 12 |
+
"train_samples_per_second": 16.953,
|
| 13 |
+
"train_steps_per_second": 0.265
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"eval_loss": 2.015979766845703,
|
| 4 |
+
"eval_runtime": 80.6758,
|
| 5 |
+
"eval_samples": 5193,
|
| 6 |
+
"eval_samples_per_second": 64.369,
|
| 7 |
+
"eval_steps_per_second": 8.057,
|
| 8 |
+
"perplexity": 7.508079946521416
|
| 9 |
+
}
|
runs/Dec17_15-41-51_jbuhmann/events.out.tfevents.1639755731.jbuhmann.28634.0
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 sha256:1cd8976a4a50e870086570bb17c02866a122c006684fcaedb03db3775fe4f1f5
|
| 3 |
+
size 9473
|
runs/Dec17_15-41-51_jbuhmann/events.out.tfevents.1639802377.jbuhmann.28634.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:842aae514ae11a8ed6000c7052427a242c6affeb66ab060d45a4a0d05255f0da
|
| 3 |
+
size 311
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"train_loss": 2.304505779894719,
|
| 4 |
+
"train_runtime": 46564.1697,
|
| 5 |
+
"train_samples": 98673,
|
| 6 |
+
"train_samples_per_second": 16.953,
|
| 7 |
+
"train_steps_per_second": 0.265
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,233 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"global_step": 12336,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.32,
|
| 12 |
+
"learning_rate": 1.9189364461738006e-05,
|
| 13 |
+
"loss": 2.5752,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.65,
|
| 18 |
+
"learning_rate": 1.8378728923476004e-05,
|
| 19 |
+
"loss": 2.4758,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.97,
|
| 24 |
+
"learning_rate": 1.756809338521401e-05,
|
| 25 |
+
"loss": 2.4695,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"eval_loss": 2.0689399242401123,
|
| 31 |
+
"eval_runtime": 81.3366,
|
| 32 |
+
"eval_samples_per_second": 63.846,
|
| 33 |
+
"eval_steps_per_second": 7.991,
|
| 34 |
+
"step": 1542
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 1.3,
|
| 38 |
+
"learning_rate": 1.675745784695201e-05,
|
| 39 |
+
"loss": 2.4094,
|
| 40 |
+
"step": 2000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 1.62,
|
| 44 |
+
"learning_rate": 1.5946822308690015e-05,
|
| 45 |
+
"loss": 2.3802,
|
| 46 |
+
"step": 2500
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.95,
|
| 50 |
+
"learning_rate": 1.5136186770428017e-05,
|
| 51 |
+
"loss": 2.3622,
|
| 52 |
+
"step": 3000
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 2.0,
|
| 56 |
+
"eval_loss": 2.0509471893310547,
|
| 57 |
+
"eval_runtime": 81.3625,
|
| 58 |
+
"eval_samples_per_second": 63.825,
|
| 59 |
+
"eval_steps_per_second": 7.989,
|
| 60 |
+
"step": 3084
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.27,
|
| 64 |
+
"learning_rate": 1.432555123216602e-05,
|
| 65 |
+
"loss": 2.3547,
|
| 66 |
+
"step": 3500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 2.59,
|
| 70 |
+
"learning_rate": 1.3514915693904023e-05,
|
| 71 |
+
"loss": 2.3319,
|
| 72 |
+
"step": 4000
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.92,
|
| 76 |
+
"learning_rate": 1.2704280155642024e-05,
|
| 77 |
+
"loss": 2.3303,
|
| 78 |
+
"step": 4500
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.0,
|
| 82 |
+
"eval_loss": 2.0795907974243164,
|
| 83 |
+
"eval_runtime": 81.2667,
|
| 84 |
+
"eval_samples_per_second": 63.901,
|
| 85 |
+
"eval_steps_per_second": 7.998,
|
| 86 |
+
"step": 4626
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.24,
|
| 90 |
+
"learning_rate": 1.1893644617380028e-05,
|
| 91 |
+
"loss": 2.293,
|
| 92 |
+
"step": 5000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.57,
|
| 96 |
+
"learning_rate": 1.108300907911803e-05,
|
| 97 |
+
"loss": 2.2819,
|
| 98 |
+
"step": 5500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 3.89,
|
| 102 |
+
"learning_rate": 1.027237354085603e-05,
|
| 103 |
+
"loss": 2.2865,
|
| 104 |
+
"step": 6000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.0,
|
| 108 |
+
"eval_loss": 1.9920992851257324,
|
| 109 |
+
"eval_runtime": 81.5086,
|
| 110 |
+
"eval_samples_per_second": 63.711,
|
| 111 |
+
"eval_steps_per_second": 7.975,
|
| 112 |
+
"step": 6168
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 4.22,
|
| 116 |
+
"learning_rate": 9.461738002594033e-06,
|
| 117 |
+
"loss": 2.2856,
|
| 118 |
+
"step": 6500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 4.54,
|
| 122 |
+
"learning_rate": 8.651102464332037e-06,
|
| 123 |
+
"loss": 2.2669,
|
| 124 |
+
"step": 7000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 4.86,
|
| 128 |
+
"learning_rate": 7.84046692607004e-06,
|
| 129 |
+
"loss": 2.2598,
|
| 130 |
+
"step": 7500
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 5.0,
|
| 134 |
+
"eval_loss": 1.9843554496765137,
|
| 135 |
+
"eval_runtime": 81.4242,
|
| 136 |
+
"eval_samples_per_second": 63.777,
|
| 137 |
+
"eval_steps_per_second": 7.983,
|
| 138 |
+
"step": 7710
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 5.19,
|
| 142 |
+
"learning_rate": 7.029831387808041e-06,
|
| 143 |
+
"loss": 2.2482,
|
| 144 |
+
"step": 8000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 5.51,
|
| 148 |
+
"learning_rate": 6.219195849546045e-06,
|
| 149 |
+
"loss": 2.2285,
|
| 150 |
+
"step": 8500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 5.84,
|
| 154 |
+
"learning_rate": 5.4085603112840465e-06,
|
| 155 |
+
"loss": 2.2544,
|
| 156 |
+
"step": 9000
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 6.0,
|
| 160 |
+
"eval_loss": 2.0076751708984375,
|
| 161 |
+
"eval_runtime": 81.3076,
|
| 162 |
+
"eval_samples_per_second": 63.869,
|
| 163 |
+
"eval_steps_per_second": 7.994,
|
| 164 |
+
"step": 9252
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 6.16,
|
| 168 |
+
"learning_rate": 4.5979247730220496e-06,
|
| 169 |
+
"loss": 2.2357,
|
| 170 |
+
"step": 9500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 6.49,
|
| 174 |
+
"learning_rate": 3.7872892347600522e-06,
|
| 175 |
+
"loss": 2.218,
|
| 176 |
+
"step": 10000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.81,
|
| 180 |
+
"learning_rate": 2.9766536964980545e-06,
|
| 181 |
+
"loss": 2.2128,
|
| 182 |
+
"step": 10500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 7.0,
|
| 186 |
+
"eval_loss": 1.9913570880889893,
|
| 187 |
+
"eval_runtime": 81.2325,
|
| 188 |
+
"eval_samples_per_second": 63.928,
|
| 189 |
+
"eval_steps_per_second": 8.002,
|
| 190 |
+
"step": 10794
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 7.13,
|
| 194 |
+
"learning_rate": 2.166018158236057e-06,
|
| 195 |
+
"loss": 2.2075,
|
| 196 |
+
"step": 11000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 7.46,
|
| 200 |
+
"learning_rate": 1.3553826199740596e-06,
|
| 201 |
+
"loss": 2.1936,
|
| 202 |
+
"step": 11500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 7.78,
|
| 206 |
+
"learning_rate": 5.447470817120623e-07,
|
| 207 |
+
"loss": 2.1963,
|
| 208 |
+
"step": 12000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 8.0,
|
| 212 |
+
"eval_loss": 1.9882222414016724,
|
| 213 |
+
"eval_runtime": 81.2899,
|
| 214 |
+
"eval_samples_per_second": 63.883,
|
| 215 |
+
"eval_steps_per_second": 7.996,
|
| 216 |
+
"step": 12336
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 8.0,
|
| 220 |
+
"step": 12336,
|
| 221 |
+
"total_flos": 2.0776857912786125e+17,
|
| 222 |
+
"train_loss": 2.304505779894719,
|
| 223 |
+
"train_runtime": 46564.1697,
|
| 224 |
+
"train_samples_per_second": 16.953,
|
| 225 |
+
"train_steps_per_second": 0.265
|
| 226 |
+
}
|
| 227 |
+
],
|
| 228 |
+
"max_steps": 12336,
|
| 229 |
+
"num_train_epochs": 8,
|
| 230 |
+
"total_flos": 2.0776857912786125e+17,
|
| 231 |
+
"trial_name": null,
|
| 232 |
+
"trial_params": null
|
| 233 |
+
}
|