Commit
•
abf7966
1
Parent(s):
1506ee3
End of training
Browse files- all_results.json +6 -6
- test_results.json +6 -6
- trainer_state.json +765 -75
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 5.
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.06,
|
| 3 |
+
"eval_accuracy": 0.8292682926829268,
|
| 4 |
+
"eval_loss": 0.4281897246837616,
|
| 5 |
+
"eval_runtime": 7.7423,
|
| 6 |
+
"eval_samples_per_second": 5.296,
|
| 7 |
+
"eval_steps_per_second": 1.421
|
| 8 |
}
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 5.
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 15.06,
|
| 3 |
+
"eval_accuracy": 0.8292682926829268,
|
| 4 |
+
"eval_loss": 0.4281897246837616,
|
| 5 |
+
"eval_runtime": 7.7423,
|
| 6 |
+
"eval_samples_per_second": 5.296,
|
| 7 |
+
"eval_steps_per_second": 1.421
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "MAE-CT-CPC-Dicotomized-v4-early-stop/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -11,257 +11,947 @@
|
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
"learning_rate": 6.944444444444446e-07,
|
| 14 |
-
"loss": 0.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.01,
|
| 19 |
"learning_rate": 1.3888888888888892e-06,
|
| 20 |
-
"loss": 0.
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.02,
|
| 25 |
"learning_rate": 2.0833333333333334e-06,
|
| 26 |
-
"loss": 0.
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.03,
|
| 31 |
"learning_rate": 2.7777777777777783e-06,
|
| 32 |
-
"loss": 0.
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.03,
|
| 37 |
"learning_rate": 3.4722222222222224e-06,
|
| 38 |
-
"loss": 0.
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.04,
|
| 43 |
"learning_rate": 4.166666666666667e-06,
|
| 44 |
-
"loss": 0.
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.05,
|
| 49 |
"learning_rate": 4.861111111111111e-06,
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.06,
|
| 55 |
"learning_rate": 5.555555555555557e-06,
|
| 56 |
-
"loss": 0.
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.06,
|
| 61 |
"eval_accuracy": 0.6829268292682927,
|
| 62 |
-
"eval_loss": 0.
|
| 63 |
-
"eval_runtime": 8.
|
| 64 |
-
"eval_samples_per_second":
|
| 65 |
-
"eval_steps_per_second": 1.
|
| 66 |
"step": 81
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 1.01,
|
| 70 |
"learning_rate": 6.25e-06,
|
| 71 |
-
"loss": 0.
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.01,
|
| 76 |
"learning_rate": 6.944444444444445e-06,
|
| 77 |
-
"loss": 0.
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
"epoch": 1.02,
|
| 82 |
"learning_rate": 7.638888888888888e-06,
|
| 83 |
-
"loss": 0.
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 1.03,
|
| 88 |
"learning_rate": 8.333333333333334e-06,
|
| 89 |
-
"loss": 0.
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 1.03,
|
| 94 |
"learning_rate": 9.027777777777779e-06,
|
| 95 |
-
"loss": 0.
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 1.04,
|
| 100 |
"learning_rate": 9.722222222222223e-06,
|
| 101 |
-
"loss": 0.
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 1.05,
|
| 106 |
"learning_rate": 9.953703703703704e-06,
|
| 107 |
-
"loss": 0.
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 1.05,
|
| 112 |
"learning_rate": 9.876543209876543e-06,
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 1.06,
|
| 118 |
"eval_accuracy": 0.6829268292682927,
|
| 119 |
-
"eval_loss": 0.
|
| 120 |
-
"eval_runtime": 8.
|
| 121 |
-
"eval_samples_per_second": 5.
|
| 122 |
-
"eval_steps_per_second": 1.
|
| 123 |
"step": 162
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 2.01,
|
| 127 |
"learning_rate": 9.799382716049384e-06,
|
| 128 |
-
"loss": 0.
|
| 129 |
"step": 170
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 2.01,
|
| 133 |
"learning_rate": 9.722222222222223e-06,
|
| 134 |
-
"loss": 0.
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 2.02,
|
| 139 |
"learning_rate": 9.645061728395062e-06,
|
| 140 |
-
"loss": 0.
|
| 141 |
"step": 190
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 2.03,
|
| 145 |
"learning_rate": 9.567901234567902e-06,
|
| 146 |
-
"loss": 0.
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 2.03,
|
| 151 |
"learning_rate": 9.490740740740741e-06,
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 2.04,
|
| 157 |
"learning_rate": 9.413580246913581e-06,
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 2.05,
|
| 163 |
"learning_rate": 9.33641975308642e-06,
|
| 164 |
-
"loss": 0.
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 2.05,
|
| 169 |
"learning_rate": 9.25925925925926e-06,
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 2.06,
|
| 175 |
-
"eval_accuracy": 0.
|
| 176 |
-
"eval_loss": 0.
|
| 177 |
-
"eval_runtime":
|
| 178 |
-
"eval_samples_per_second":
|
| 179 |
-
"eval_steps_per_second": 1.
|
| 180 |
"step": 243
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 3.0,
|
| 184 |
"learning_rate": 9.1820987654321e-06,
|
| 185 |
-
"loss": 0.
|
| 186 |
"step": 250
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 3.01,
|
| 190 |
"learning_rate": 9.10493827160494e-06,
|
| 191 |
-
"loss": 0.
|
| 192 |
"step": 260
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 3.02,
|
| 196 |
"learning_rate": 9.027777777777779e-06,
|
| 197 |
-
"loss": 0.
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 3.03,
|
| 202 |
"learning_rate": 8.950617283950618e-06,
|
| 203 |
-
"loss": 0.
|
| 204 |
"step": 280
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 3.03,
|
| 208 |
"learning_rate": 8.873456790123458e-06,
|
| 209 |
-
"loss": 0.
|
| 210 |
"step": 290
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 3.04,
|
| 214 |
"learning_rate": 8.796296296296297e-06,
|
| 215 |
-
"loss": 0.
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 3.05,
|
| 220 |
"learning_rate": 8.719135802469136e-06,
|
| 221 |
-
"loss": 0.
|
| 222 |
"step": 310
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 3.05,
|
| 226 |
"learning_rate": 8.641975308641975e-06,
|
| 227 |
-
"loss": 0.
|
| 228 |
"step": 320
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 3.06,
|
| 232 |
-
"eval_accuracy": 0.
|
| 233 |
-
"eval_loss": 0.
|
| 234 |
-
"eval_runtime":
|
| 235 |
-
"eval_samples_per_second":
|
| 236 |
-
"eval_steps_per_second": 1.
|
| 237 |
"step": 324
|
| 238 |
},
|
| 239 |
{
|
| 240 |
-
"epoch":
|
| 241 |
-
"
|
| 242 |
-
"
|
| 243 |
-
"
|
| 244 |
-
"train_runtime": 569.9574,
|
| 245 |
-
"train_samples_per_second": 10.106,
|
| 246 |
-
"train_steps_per_second": 2.527
|
| 247 |
},
|
| 248 |
{
|
| 249 |
-
"epoch":
|
| 250 |
-
"
|
| 251 |
-
"
|
| 252 |
-
"
|
| 253 |
-
"eval_samples_per_second": 4.665,
|
| 254 |
-
"eval_steps_per_second": 1.252,
|
| 255 |
-
"step": 324
|
| 256 |
},
|
| 257 |
{
|
| 258 |
-
"epoch":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 259 |
"eval_accuracy": 0.6829268292682927,
|
| 260 |
-
"eval_loss": 0.
|
| 261 |
-
"eval_runtime": 8.
|
| 262 |
-
"eval_samples_per_second":
|
| 263 |
-
"eval_steps_per_second": 1.
|
| 264 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 265 |
}
|
| 266 |
],
|
| 267 |
"logging_steps": 10,
|
|
@@ -269,7 +959,7 @@
|
|
| 269 |
"num_input_tokens_seen": 0,
|
| 270 |
"num_train_epochs": 9223372036854775807,
|
| 271 |
"save_steps": 500,
|
| 272 |
-
"total_flos":
|
| 273 |
"train_batch_size": 4,
|
| 274 |
"trial_name": null,
|
| 275 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8048780487804879,
|
| 3 |
+
"best_model_checkpoint": "MAE-CT-CPC-Dicotomized-v4-early-stop/checkpoint-648",
|
| 4 |
+
"epoch": 15.05625,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1296,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 11 |
{
|
| 12 |
"epoch": 0.01,
|
| 13 |
"learning_rate": 6.944444444444446e-07,
|
| 14 |
+
"loss": 0.5724,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.01,
|
| 19 |
"learning_rate": 1.3888888888888892e-06,
|
| 20 |
+
"loss": 0.633,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.02,
|
| 25 |
"learning_rate": 2.0833333333333334e-06,
|
| 26 |
+
"loss": 0.5686,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.03,
|
| 31 |
"learning_rate": 2.7777777777777783e-06,
|
| 32 |
+
"loss": 0.605,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.03,
|
| 37 |
"learning_rate": 3.4722222222222224e-06,
|
| 38 |
+
"loss": 0.7115,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.04,
|
| 43 |
"learning_rate": 4.166666666666667e-06,
|
| 44 |
+
"loss": 0.6218,
|
| 45 |
"step": 60
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.05,
|
| 49 |
"learning_rate": 4.861111111111111e-06,
|
| 50 |
+
"loss": 0.6258,
|
| 51 |
"step": 70
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.06,
|
| 55 |
"learning_rate": 5.555555555555557e-06,
|
| 56 |
+
"loss": 0.5342,
|
| 57 |
"step": 80
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.06,
|
| 61 |
"eval_accuracy": 0.6829268292682927,
|
| 62 |
+
"eval_loss": 0.6496201753616333,
|
| 63 |
+
"eval_runtime": 8.1028,
|
| 64 |
+
"eval_samples_per_second": 5.06,
|
| 65 |
+
"eval_steps_per_second": 1.358,
|
| 66 |
"step": 81
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 1.01,
|
| 70 |
"learning_rate": 6.25e-06,
|
| 71 |
+
"loss": 0.6638,
|
| 72 |
"step": 90
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 1.01,
|
| 76 |
"learning_rate": 6.944444444444445e-06,
|
| 77 |
+
"loss": 0.5117,
|
| 78 |
"step": 100
|
| 79 |
},
|
| 80 |
{
|
| 81 |
"epoch": 1.02,
|
| 82 |
"learning_rate": 7.638888888888888e-06,
|
| 83 |
+
"loss": 0.6552,
|
| 84 |
"step": 110
|
| 85 |
},
|
| 86 |
{
|
| 87 |
"epoch": 1.03,
|
| 88 |
"learning_rate": 8.333333333333334e-06,
|
| 89 |
+
"loss": 0.5231,
|
| 90 |
"step": 120
|
| 91 |
},
|
| 92 |
{
|
| 93 |
"epoch": 1.03,
|
| 94 |
"learning_rate": 9.027777777777779e-06,
|
| 95 |
+
"loss": 0.6073,
|
| 96 |
"step": 130
|
| 97 |
},
|
| 98 |
{
|
| 99 |
"epoch": 1.04,
|
| 100 |
"learning_rate": 9.722222222222223e-06,
|
| 101 |
+
"loss": 0.6471,
|
| 102 |
"step": 140
|
| 103 |
},
|
| 104 |
{
|
| 105 |
"epoch": 1.05,
|
| 106 |
"learning_rate": 9.953703703703704e-06,
|
| 107 |
+
"loss": 0.6375,
|
| 108 |
"step": 150
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 1.05,
|
| 112 |
"learning_rate": 9.876543209876543e-06,
|
| 113 |
+
"loss": 0.6266,
|
| 114 |
"step": 160
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 1.06,
|
| 118 |
"eval_accuracy": 0.6829268292682927,
|
| 119 |
+
"eval_loss": 0.5853927135467529,
|
| 120 |
+
"eval_runtime": 8.0931,
|
| 121 |
+
"eval_samples_per_second": 5.066,
|
| 122 |
+
"eval_steps_per_second": 1.359,
|
| 123 |
"step": 162
|
| 124 |
},
|
| 125 |
{
|
| 126 |
"epoch": 2.01,
|
| 127 |
"learning_rate": 9.799382716049384e-06,
|
| 128 |
+
"loss": 0.5042,
|
| 129 |
"step": 170
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"epoch": 2.01,
|
| 133 |
"learning_rate": 9.722222222222223e-06,
|
| 134 |
+
"loss": 0.4928,
|
| 135 |
"step": 180
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"epoch": 2.02,
|
| 139 |
"learning_rate": 9.645061728395062e-06,
|
| 140 |
+
"loss": 0.5231,
|
| 141 |
"step": 190
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 2.03,
|
| 145 |
"learning_rate": 9.567901234567902e-06,
|
| 146 |
+
"loss": 0.5299,
|
| 147 |
"step": 200
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 2.03,
|
| 151 |
"learning_rate": 9.490740740740741e-06,
|
| 152 |
+
"loss": 0.6126,
|
| 153 |
"step": 210
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 2.04,
|
| 157 |
"learning_rate": 9.413580246913581e-06,
|
| 158 |
+
"loss": 0.4196,
|
| 159 |
"step": 220
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 2.05,
|
| 163 |
"learning_rate": 9.33641975308642e-06,
|
| 164 |
+
"loss": 0.9175,
|
| 165 |
"step": 230
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 2.05,
|
| 169 |
"learning_rate": 9.25925925925926e-06,
|
| 170 |
+
"loss": 0.6599,
|
| 171 |
"step": 240
|
| 172 |
},
|
| 173 |
{
|
| 174 |
"epoch": 2.06,
|
| 175 |
+
"eval_accuracy": 0.6829268292682927,
|
| 176 |
+
"eval_loss": 0.5205796957015991,
|
| 177 |
+
"eval_runtime": 8.4167,
|
| 178 |
+
"eval_samples_per_second": 4.871,
|
| 179 |
+
"eval_steps_per_second": 1.307,
|
| 180 |
"step": 243
|
| 181 |
},
|
| 182 |
{
|
| 183 |
"epoch": 3.0,
|
| 184 |
"learning_rate": 9.1820987654321e-06,
|
| 185 |
+
"loss": 0.5163,
|
| 186 |
"step": 250
|
| 187 |
},
|
| 188 |
{
|
| 189 |
"epoch": 3.01,
|
| 190 |
"learning_rate": 9.10493827160494e-06,
|
| 191 |
+
"loss": 0.6428,
|
| 192 |
"step": 260
|
| 193 |
},
|
| 194 |
{
|
| 195 |
"epoch": 3.02,
|
| 196 |
"learning_rate": 9.027777777777779e-06,
|
| 197 |
+
"loss": 0.5963,
|
| 198 |
"step": 270
|
| 199 |
},
|
| 200 |
{
|
| 201 |
"epoch": 3.03,
|
| 202 |
"learning_rate": 8.950617283950618e-06,
|
| 203 |
+
"loss": 0.5579,
|
| 204 |
"step": 280
|
| 205 |
},
|
| 206 |
{
|
| 207 |
"epoch": 3.03,
|
| 208 |
"learning_rate": 8.873456790123458e-06,
|
| 209 |
+
"loss": 0.6436,
|
| 210 |
"step": 290
|
| 211 |
},
|
| 212 |
{
|
| 213 |
"epoch": 3.04,
|
| 214 |
"learning_rate": 8.796296296296297e-06,
|
| 215 |
+
"loss": 0.6368,
|
| 216 |
"step": 300
|
| 217 |
},
|
| 218 |
{
|
| 219 |
"epoch": 3.05,
|
| 220 |
"learning_rate": 8.719135802469136e-06,
|
| 221 |
+
"loss": 0.5194,
|
| 222 |
"step": 310
|
| 223 |
},
|
| 224 |
{
|
| 225 |
"epoch": 3.05,
|
| 226 |
"learning_rate": 8.641975308641975e-06,
|
| 227 |
+
"loss": 0.877,
|
| 228 |
"step": 320
|
| 229 |
},
|
| 230 |
{
|
| 231 |
"epoch": 3.06,
|
| 232 |
+
"eval_accuracy": 0.6097560975609756,
|
| 233 |
+
"eval_loss": 0.5995270013809204,
|
| 234 |
+
"eval_runtime": 8.3891,
|
| 235 |
+
"eval_samples_per_second": 4.887,
|
| 236 |
+
"eval_steps_per_second": 1.311,
|
| 237 |
"step": 324
|
| 238 |
},
|
| 239 |
{
|
| 240 |
+
"epoch": 4.0,
|
| 241 |
+
"learning_rate": 8.564814814814816e-06,
|
| 242 |
+
"loss": 0.5644,
|
| 243 |
+
"step": 330
|
|
|
|
|
|
|
|
|
|
| 244 |
},
|
| 245 |
{
|
| 246 |
+
"epoch": 4.01,
|
| 247 |
+
"learning_rate": 8.487654320987654e-06,
|
| 248 |
+
"loss": 0.4776,
|
| 249 |
+
"step": 340
|
|
|
|
|
|
|
|
|
|
| 250 |
},
|
| 251 |
{
|
| 252 |
+
"epoch": 4.02,
|
| 253 |
+
"learning_rate": 8.410493827160495e-06,
|
| 254 |
+
"loss": 0.5117,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.03,
|
| 259 |
+
"learning_rate": 8.333333333333334e-06,
|
| 260 |
+
"loss": 0.4388,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.03,
|
| 265 |
+
"learning_rate": 8.256172839506174e-06,
|
| 266 |
+
"loss": 0.5065,
|
| 267 |
+
"step": 370
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 4.04,
|
| 271 |
+
"learning_rate": 8.179012345679013e-06,
|
| 272 |
+
"loss": 0.5147,
|
| 273 |
+
"step": 380
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 4.05,
|
| 277 |
+
"learning_rate": 8.101851851851854e-06,
|
| 278 |
+
"loss": 0.6924,
|
| 279 |
+
"step": 390
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.05,
|
| 283 |
+
"learning_rate": 8.024691358024692e-06,
|
| 284 |
+
"loss": 0.653,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.06,
|
| 289 |
+
"eval_accuracy": 0.7560975609756098,
|
| 290 |
+
"eval_loss": 0.4907689392566681,
|
| 291 |
+
"eval_runtime": 7.7719,
|
| 292 |
+
"eval_samples_per_second": 5.275,
|
| 293 |
+
"eval_steps_per_second": 1.415,
|
| 294 |
+
"step": 405
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.0,
|
| 298 |
+
"learning_rate": 7.947530864197531e-06,
|
| 299 |
+
"loss": 0.536,
|
| 300 |
+
"step": 410
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 5.01,
|
| 304 |
+
"learning_rate": 7.870370370370372e-06,
|
| 305 |
+
"loss": 0.4254,
|
| 306 |
+
"step": 420
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 5.02,
|
| 310 |
+
"learning_rate": 7.79320987654321e-06,
|
| 311 |
+
"loss": 0.4454,
|
| 312 |
+
"step": 430
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 5.02,
|
| 316 |
+
"learning_rate": 7.71604938271605e-06,
|
| 317 |
+
"loss": 0.4594,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 5.03,
|
| 322 |
+
"learning_rate": 7.638888888888888e-06,
|
| 323 |
+
"loss": 0.4811,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 5.04,
|
| 328 |
+
"learning_rate": 7.561728395061729e-06,
|
| 329 |
+
"loss": 0.4024,
|
| 330 |
+
"step": 460
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 5.05,
|
| 334 |
+
"learning_rate": 7.484567901234569e-06,
|
| 335 |
+
"loss": 0.3778,
|
| 336 |
+
"step": 470
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 5.05,
|
| 340 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 341 |
+
"loss": 0.7604,
|
| 342 |
+
"step": 480
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 5.06,
|
| 346 |
+
"eval_accuracy": 0.7804878048780488,
|
| 347 |
+
"eval_loss": 0.49357670545578003,
|
| 348 |
+
"eval_runtime": 8.1369,
|
| 349 |
+
"eval_samples_per_second": 5.039,
|
| 350 |
+
"eval_steps_per_second": 1.352,
|
| 351 |
+
"step": 486
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 6.0,
|
| 355 |
+
"learning_rate": 7.330246913580248e-06,
|
| 356 |
+
"loss": 0.5151,
|
| 357 |
+
"step": 490
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 6.01,
|
| 361 |
+
"learning_rate": 7.253086419753087e-06,
|
| 362 |
+
"loss": 0.3392,
|
| 363 |
+
"step": 500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 6.02,
|
| 367 |
+
"learning_rate": 7.1759259259259266e-06,
|
| 368 |
+
"loss": 0.5844,
|
| 369 |
+
"step": 510
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 6.02,
|
| 373 |
+
"learning_rate": 7.098765432098766e-06,
|
| 374 |
+
"loss": 0.5688,
|
| 375 |
+
"step": 520
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 6.03,
|
| 379 |
+
"learning_rate": 7.021604938271606e-06,
|
| 380 |
+
"loss": 0.247,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 6.04,
|
| 385 |
+
"learning_rate": 6.944444444444445e-06,
|
| 386 |
+
"loss": 0.7517,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 6.04,
|
| 391 |
+
"learning_rate": 6.867283950617285e-06,
|
| 392 |
+
"loss": 0.3608,
|
| 393 |
+
"step": 550
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 6.05,
|
| 397 |
+
"learning_rate": 6.790123456790124e-06,
|
| 398 |
+
"loss": 0.4795,
|
| 399 |
+
"step": 560
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 6.06,
|
| 403 |
"eval_accuracy": 0.6829268292682927,
|
| 404 |
+
"eval_loss": 0.9527755379676819,
|
| 405 |
+
"eval_runtime": 8.3829,
|
| 406 |
+
"eval_samples_per_second": 4.891,
|
| 407 |
+
"eval_steps_per_second": 1.312,
|
| 408 |
+
"step": 567
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.0,
|
| 412 |
+
"learning_rate": 6.712962962962963e-06,
|
| 413 |
+
"loss": 0.3914,
|
| 414 |
+
"step": 570
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 7.01,
|
| 418 |
+
"learning_rate": 6.635802469135803e-06,
|
| 419 |
+
"loss": 0.4056,
|
| 420 |
+
"step": 580
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 7.02,
|
| 424 |
+
"learning_rate": 6.558641975308642e-06,
|
| 425 |
+
"loss": 0.4884,
|
| 426 |
+
"step": 590
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 7.02,
|
| 430 |
+
"learning_rate": 6.481481481481482e-06,
|
| 431 |
+
"loss": 0.3237,
|
| 432 |
+
"step": 600
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 7.03,
|
| 436 |
+
"learning_rate": 6.404320987654321e-06,
|
| 437 |
+
"loss": 0.4742,
|
| 438 |
+
"step": 610
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 7.04,
|
| 442 |
+
"learning_rate": 6.3271604938271615e-06,
|
| 443 |
+
"loss": 0.4839,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 7.04,
|
| 448 |
+
"learning_rate": 6.25e-06,
|
| 449 |
+
"loss": 0.265,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 7.05,
|
| 454 |
+
"learning_rate": 6.17283950617284e-06,
|
| 455 |
+
"loss": 0.278,
|
| 456 |
+
"step": 640
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 7.06,
|
| 460 |
+
"eval_accuracy": 0.8048780487804879,
|
| 461 |
+
"eval_loss": 0.5564919710159302,
|
| 462 |
+
"eval_runtime": 8.4164,
|
| 463 |
+
"eval_samples_per_second": 4.871,
|
| 464 |
+
"eval_steps_per_second": 1.307,
|
| 465 |
+
"step": 648
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 8.0,
|
| 469 |
+
"learning_rate": 6.09567901234568e-06,
|
| 470 |
+
"loss": 0.1885,
|
| 471 |
+
"step": 650
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 8.01,
|
| 475 |
+
"learning_rate": 6.018518518518519e-06,
|
| 476 |
+
"loss": 0.2307,
|
| 477 |
+
"step": 660
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 8.02,
|
| 481 |
+
"learning_rate": 5.941358024691358e-06,
|
| 482 |
+
"loss": 0.4786,
|
| 483 |
+
"step": 670
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 8.02,
|
| 487 |
+
"learning_rate": 5.864197530864199e-06,
|
| 488 |
+
"loss": 0.3708,
|
| 489 |
+
"step": 680
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 8.03,
|
| 493 |
+
"learning_rate": 5.787037037037038e-06,
|
| 494 |
+
"loss": 0.5062,
|
| 495 |
+
"step": 690
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 8.04,
|
| 499 |
+
"learning_rate": 5.7098765432098764e-06,
|
| 500 |
+
"loss": 0.7825,
|
| 501 |
+
"step": 700
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 8.04,
|
| 505 |
+
"learning_rate": 5.632716049382716e-06,
|
| 506 |
+
"loss": 0.1902,
|
| 507 |
+
"step": 710
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 8.05,
|
| 511 |
+
"learning_rate": 5.555555555555557e-06,
|
| 512 |
+
"loss": 0.3548,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 8.06,
|
| 517 |
+
"eval_accuracy": 0.7560975609756098,
|
| 518 |
+
"eval_loss": 0.5855191946029663,
|
| 519 |
+
"eval_runtime": 7.7883,
|
| 520 |
+
"eval_samples_per_second": 5.264,
|
| 521 |
+
"eval_steps_per_second": 1.412,
|
| 522 |
+
"step": 729
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 9.0,
|
| 526 |
+
"learning_rate": 5.4783950617283955e-06,
|
| 527 |
+
"loss": 0.2654,
|
| 528 |
+
"step": 730
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 9.01,
|
| 532 |
+
"learning_rate": 5.401234567901234e-06,
|
| 533 |
+
"loss": 0.2532,
|
| 534 |
+
"step": 740
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 9.01,
|
| 538 |
+
"learning_rate": 5.324074074074075e-06,
|
| 539 |
+
"loss": 0.243,
|
| 540 |
+
"step": 750
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 9.02,
|
| 544 |
+
"learning_rate": 5.246913580246914e-06,
|
| 545 |
+
"loss": 0.4346,
|
| 546 |
+
"step": 760
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 9.03,
|
| 550 |
+
"learning_rate": 5.1697530864197534e-06,
|
| 551 |
+
"loss": 0.1327,
|
| 552 |
+
"step": 770
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 9.04,
|
| 556 |
+
"learning_rate": 5.092592592592593e-06,
|
| 557 |
+
"loss": 0.3484,
|
| 558 |
+
"step": 780
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 9.04,
|
| 562 |
+
"learning_rate": 5.015432098765433e-06,
|
| 563 |
+
"loss": 0.4213,
|
| 564 |
+
"step": 790
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 9.05,
|
| 568 |
+
"learning_rate": 4.938271604938272e-06,
|
| 569 |
+
"loss": 0.3673,
|
| 570 |
+
"step": 800
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 9.06,
|
| 574 |
+
"learning_rate": 4.861111111111111e-06,
|
| 575 |
+
"loss": 0.4386,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 9.06,
|
| 580 |
+
"eval_accuracy": 0.7560975609756098,
|
| 581 |
+
"eval_loss": 0.6577650904655457,
|
| 582 |
+
"eval_runtime": 8.1212,
|
| 583 |
+
"eval_samples_per_second": 5.049,
|
| 584 |
+
"eval_steps_per_second": 1.354,
|
| 585 |
+
"step": 810
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 10.01,
|
| 589 |
+
"learning_rate": 4.783950617283951e-06,
|
| 590 |
+
"loss": 0.5757,
|
| 591 |
+
"step": 820
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 10.01,
|
| 595 |
+
"learning_rate": 4.706790123456791e-06,
|
| 596 |
+
"loss": 0.2701,
|
| 597 |
+
"step": 830
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 10.02,
|
| 601 |
+
"learning_rate": 4.62962962962963e-06,
|
| 602 |
+
"loss": 0.1286,
|
| 603 |
+
"step": 840
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 10.03,
|
| 607 |
+
"learning_rate": 4.55246913580247e-06,
|
| 608 |
+
"loss": 0.2345,
|
| 609 |
+
"step": 850
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 10.03,
|
| 613 |
+
"learning_rate": 4.475308641975309e-06,
|
| 614 |
+
"loss": 0.1286,
|
| 615 |
+
"step": 860
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 10.04,
|
| 619 |
+
"learning_rate": 4.398148148148149e-06,
|
| 620 |
+
"loss": 0.3579,
|
| 621 |
+
"step": 870
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 10.05,
|
| 625 |
+
"learning_rate": 4.3209876543209875e-06,
|
| 626 |
+
"loss": 0.408,
|
| 627 |
+
"step": 880
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 10.06,
|
| 631 |
+
"learning_rate": 4.243827160493827e-06,
|
| 632 |
+
"loss": 0.3007,
|
| 633 |
+
"step": 890
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 10.06,
|
| 637 |
+
"eval_accuracy": 0.7804878048780488,
|
| 638 |
+
"eval_loss": 0.6622430682182312,
|
| 639 |
+
"eval_runtime": 8.0725,
|
| 640 |
+
"eval_samples_per_second": 5.079,
|
| 641 |
+
"eval_steps_per_second": 1.363,
|
| 642 |
+
"step": 891
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 11.01,
|
| 646 |
+
"learning_rate": 4.166666666666667e-06,
|
| 647 |
+
"loss": 0.4887,
|
| 648 |
+
"step": 900
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 11.01,
|
| 652 |
+
"learning_rate": 4.0895061728395066e-06,
|
| 653 |
+
"loss": 0.1813,
|
| 654 |
+
"step": 910
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 11.02,
|
| 658 |
+
"learning_rate": 4.012345679012346e-06,
|
| 659 |
+
"loss": 0.1648,
|
| 660 |
+
"step": 920
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 11.03,
|
| 664 |
+
"learning_rate": 3.935185185185186e-06,
|
| 665 |
+
"loss": 0.4133,
|
| 666 |
+
"step": 930
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 11.03,
|
| 670 |
+
"learning_rate": 3.858024691358025e-06,
|
| 671 |
+
"loss": 0.1691,
|
| 672 |
+
"step": 940
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 11.04,
|
| 676 |
+
"learning_rate": 3.7808641975308645e-06,
|
| 677 |
+
"loss": 0.1626,
|
| 678 |
+
"step": 950
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 11.05,
|
| 682 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 683 |
+
"loss": 0.2215,
|
| 684 |
+
"step": 960
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 11.05,
|
| 688 |
+
"learning_rate": 3.6265432098765434e-06,
|
| 689 |
+
"loss": 0.313,
|
| 690 |
+
"step": 970
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 11.06,
|
| 694 |
+
"eval_accuracy": 0.7560975609756098,
|
| 695 |
+
"eval_loss": 0.8349580764770508,
|
| 696 |
+
"eval_runtime": 8.0234,
|
| 697 |
+
"eval_samples_per_second": 5.11,
|
| 698 |
+
"eval_steps_per_second": 1.371,
|
| 699 |
+
"step": 972
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 12.01,
|
| 703 |
+
"learning_rate": 3.549382716049383e-06,
|
| 704 |
+
"loss": 0.4007,
|
| 705 |
+
"step": 980
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 12.01,
|
| 709 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 710 |
+
"loss": 0.2461,
|
| 711 |
+
"step": 990
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 12.02,
|
| 715 |
+
"learning_rate": 3.395061728395062e-06,
|
| 716 |
+
"loss": 0.2541,
|
| 717 |
+
"step": 1000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 12.03,
|
| 721 |
+
"learning_rate": 3.3179012345679013e-06,
|
| 722 |
+
"loss": 0.3941,
|
| 723 |
+
"step": 1010
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 12.03,
|
| 727 |
+
"learning_rate": 3.240740740740741e-06,
|
| 728 |
+
"loss": 0.0845,
|
| 729 |
+
"step": 1020
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 12.04,
|
| 733 |
+
"learning_rate": 3.1635802469135807e-06,
|
| 734 |
+
"loss": 0.2944,
|
| 735 |
+
"step": 1030
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 12.05,
|
| 739 |
+
"learning_rate": 3.08641975308642e-06,
|
| 740 |
+
"loss": 0.2391,
|
| 741 |
+
"step": 1040
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 12.05,
|
| 745 |
+
"learning_rate": 3.0092592592592597e-06,
|
| 746 |
+
"loss": 0.0554,
|
| 747 |
+
"step": 1050
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 12.06,
|
| 751 |
+
"eval_accuracy": 0.7073170731707317,
|
| 752 |
+
"eval_loss": 1.0043153762817383,
|
| 753 |
+
"eval_runtime": 7.5796,
|
| 754 |
+
"eval_samples_per_second": 5.409,
|
| 755 |
+
"eval_steps_per_second": 1.451,
|
| 756 |
+
"step": 1053
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 13.0,
|
| 760 |
+
"learning_rate": 2.9320987654320994e-06,
|
| 761 |
+
"loss": 0.2887,
|
| 762 |
+
"step": 1060
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 13.01,
|
| 766 |
+
"learning_rate": 2.8549382716049382e-06,
|
| 767 |
+
"loss": 0.4346,
|
| 768 |
+
"step": 1070
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 13.02,
|
| 772 |
+
"learning_rate": 2.7777777777777783e-06,
|
| 773 |
+
"loss": 0.3038,
|
| 774 |
+
"step": 1080
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 13.03,
|
| 778 |
+
"learning_rate": 2.700617283950617e-06,
|
| 779 |
+
"loss": 0.1073,
|
| 780 |
+
"step": 1090
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 13.03,
|
| 784 |
+
"learning_rate": 2.623456790123457e-06,
|
| 785 |
+
"loss": 0.2334,
|
| 786 |
+
"step": 1100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 13.04,
|
| 790 |
+
"learning_rate": 2.5462962962962966e-06,
|
| 791 |
+
"loss": 0.34,
|
| 792 |
+
"step": 1110
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 13.05,
|
| 796 |
+
"learning_rate": 2.469135802469136e-06,
|
| 797 |
+
"loss": 0.0705,
|
| 798 |
+
"step": 1120
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 13.05,
|
| 802 |
+
"learning_rate": 2.3919753086419755e-06,
|
| 803 |
+
"loss": 0.2804,
|
| 804 |
+
"step": 1130
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 13.06,
|
| 808 |
+
"eval_accuracy": 0.7073170731707317,
|
| 809 |
+
"eval_loss": 1.0246809720993042,
|
| 810 |
+
"eval_runtime": 7.4342,
|
| 811 |
+
"eval_samples_per_second": 5.515,
|
| 812 |
+
"eval_steps_per_second": 1.48,
|
| 813 |
+
"step": 1134
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 14.0,
|
| 817 |
+
"learning_rate": 2.314814814814815e-06,
|
| 818 |
+
"loss": 0.0367,
|
| 819 |
+
"step": 1140
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 14.01,
|
| 823 |
+
"learning_rate": 2.2376543209876545e-06,
|
| 824 |
+
"loss": 0.2872,
|
| 825 |
+
"step": 1150
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 14.02,
|
| 829 |
+
"learning_rate": 2.1604938271604937e-06,
|
| 830 |
+
"loss": 0.0806,
|
| 831 |
+
"step": 1160
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 14.03,
|
| 835 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 836 |
+
"loss": 0.1376,
|
| 837 |
+
"step": 1170
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 14.03,
|
| 841 |
+
"learning_rate": 2.006172839506173e-06,
|
| 842 |
+
"loss": 0.466,
|
| 843 |
+
"step": 1180
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 14.04,
|
| 847 |
+
"learning_rate": 1.9290123456790124e-06,
|
| 848 |
+
"loss": 0.2337,
|
| 849 |
+
"step": 1190
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 14.05,
|
| 853 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 854 |
+
"loss": 0.3175,
|
| 855 |
+
"step": 1200
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 14.05,
|
| 859 |
+
"learning_rate": 1.7746913580246916e-06,
|
| 860 |
+
"loss": 0.1424,
|
| 861 |
+
"step": 1210
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 14.06,
|
| 865 |
+
"eval_accuracy": 0.7804878048780488,
|
| 866 |
+
"eval_loss": 0.8541743755340576,
|
| 867 |
+
"eval_runtime": 7.7108,
|
| 868 |
+
"eval_samples_per_second": 5.317,
|
| 869 |
+
"eval_steps_per_second": 1.427,
|
| 870 |
+
"step": 1215
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 15.0,
|
| 874 |
+
"learning_rate": 1.697530864197531e-06,
|
| 875 |
+
"loss": 0.1911,
|
| 876 |
+
"step": 1220
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 15.01,
|
| 880 |
+
"learning_rate": 1.6203703703703705e-06,
|
| 881 |
+
"loss": 0.0041,
|
| 882 |
+
"step": 1230
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 15.02,
|
| 886 |
+
"learning_rate": 1.54320987654321e-06,
|
| 887 |
+
"loss": 0.3306,
|
| 888 |
+
"step": 1240
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 15.02,
|
| 892 |
+
"learning_rate": 1.4660493827160497e-06,
|
| 893 |
+
"loss": 0.1108,
|
| 894 |
+
"step": 1250
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 15.03,
|
| 898 |
+
"learning_rate": 1.3888888888888892e-06,
|
| 899 |
+
"loss": 0.1808,
|
| 900 |
+
"step": 1260
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 15.04,
|
| 904 |
+
"learning_rate": 1.3117283950617284e-06,
|
| 905 |
+
"loss": 0.1984,
|
| 906 |
+
"step": 1270
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 15.05,
|
| 910 |
+
"learning_rate": 1.234567901234568e-06,
|
| 911 |
+
"loss": 0.0157,
|
| 912 |
+
"step": 1280
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 15.05,
|
| 916 |
+
"learning_rate": 1.1574074074074076e-06,
|
| 917 |
+
"loss": 0.4692,
|
| 918 |
+
"step": 1290
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 15.06,
|
| 922 |
+
"eval_accuracy": 0.7317073170731707,
|
| 923 |
+
"eval_loss": 1.0264424085617065,
|
| 924 |
+
"eval_runtime": 7.57,
|
| 925 |
+
"eval_samples_per_second": 5.416,
|
| 926 |
+
"eval_steps_per_second": 1.453,
|
| 927 |
+
"step": 1296
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 15.06,
|
| 931 |
+
"step": 1296,
|
| 932 |
+
"total_flos": 2.262256757640895e+19,
|
| 933 |
+
"train_loss": 0.4041450611419148,
|
| 934 |
+
"train_runtime": 2319.3656,
|
| 935 |
+
"train_samples_per_second": 2.483,
|
| 936 |
+
"train_steps_per_second": 0.621
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 15.06,
|
| 940 |
+
"eval_accuracy": 0.8292682926829268,
|
| 941 |
+
"eval_loss": 0.4281896650791168,
|
| 942 |
+
"eval_runtime": 7.8902,
|
| 943 |
+
"eval_samples_per_second": 5.196,
|
| 944 |
+
"eval_steps_per_second": 1.394,
|
| 945 |
+
"step": 1296
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 15.06,
|
| 949 |
+
"eval_accuracy": 0.8292682926829268,
|
| 950 |
+
"eval_loss": 0.4281897246837616,
|
| 951 |
+
"eval_runtime": 7.7423,
|
| 952 |
+
"eval_samples_per_second": 5.296,
|
| 953 |
+
"eval_steps_per_second": 1.421,
|
| 954 |
+
"step": 1296
|
| 955 |
}
|
| 956 |
],
|
| 957 |
"logging_steps": 10,
|
|
|
|
| 959 |
"num_input_tokens_seen": 0,
|
| 960 |
"num_train_epochs": 9223372036854775807,
|
| 961 |
"save_steps": 500,
|
| 962 |
+
"total_flos": 2.262256757640895e+19,
|
| 963 |
"train_batch_size": 4,
|
| 964 |
"trial_name": null,
|
| 965 |
"trial_params": null
|