Commit
•
0f5fba5
1
Parent(s):
cd5988f
Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +247 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,247 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.997442455242968,
|
| 5 |
+
"global_step": 3900,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"eval_accuracy": 0.9110807180404663,
|
| 13 |
+
"eval_loss": 0.2603251338005066,
|
| 14 |
+
"eval_runtime": 3.7587,
|
| 15 |
+
"eval_samples_per_second": 194.484,
|
| 16 |
+
"eval_steps_per_second": 24.477,
|
| 17 |
+
"step": 195
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 2.0,
|
| 21 |
+
"eval_accuracy": 0.9179206490516663,
|
| 22 |
+
"eval_loss": 0.2944748103618622,
|
| 23 |
+
"eval_runtime": 3.8108,
|
| 24 |
+
"eval_samples_per_second": 191.821,
|
| 25 |
+
"eval_steps_per_second": 24.142,
|
| 26 |
+
"step": 390
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.56,
|
| 30 |
+
"learning_rate": 3.282996735800617e-05,
|
| 31 |
+
"loss": 0.1807,
|
| 32 |
+
"step": 500
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.0,
|
| 36 |
+
"eval_accuracy": 0.9083447456359863,
|
| 37 |
+
"eval_loss": 0.43401873111724854,
|
| 38 |
+
"eval_runtime": 3.7691,
|
| 39 |
+
"eval_samples_per_second": 193.947,
|
| 40 |
+
"eval_steps_per_second": 24.409,
|
| 41 |
+
"step": 585
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 4.0,
|
| 45 |
+
"eval_accuracy": 0.9124487042427063,
|
| 46 |
+
"eval_loss": 0.39779114723205566,
|
| 47 |
+
"eval_runtime": 3.7531,
|
| 48 |
+
"eval_samples_per_second": 194.771,
|
| 49 |
+
"eval_steps_per_second": 24.513,
|
| 50 |
+
"step": 780
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 5.0,
|
| 54 |
+
"eval_accuracy": 0.9042407870292664,
|
| 55 |
+
"eval_loss": 0.6419674754142761,
|
| 56 |
+
"eval_runtime": 3.6686,
|
| 57 |
+
"eval_samples_per_second": 199.26,
|
| 58 |
+
"eval_steps_per_second": 25.078,
|
| 59 |
+
"step": 975
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 5.13,
|
| 63 |
+
"learning_rate": 2.800203098182879e-05,
|
| 64 |
+
"loss": 0.0523,
|
| 65 |
+
"step": 1000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 6.0,
|
| 69 |
+
"eval_accuracy": 0.9069767594337463,
|
| 70 |
+
"eval_loss": 0.6490611433982849,
|
| 71 |
+
"eval_runtime": 3.596,
|
| 72 |
+
"eval_samples_per_second": 203.28,
|
| 73 |
+
"eval_steps_per_second": 25.584,
|
| 74 |
+
"step": 1170
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 7.0,
|
| 78 |
+
"eval_accuracy": 0.9124487042427063,
|
| 79 |
+
"eval_loss": 0.6302383542060852,
|
| 80 |
+
"eval_runtime": 3.7728,
|
| 81 |
+
"eval_samples_per_second": 193.754,
|
| 82 |
+
"eval_steps_per_second": 24.385,
|
| 83 |
+
"step": 1365
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 7.69,
|
| 87 |
+
"learning_rate": 2.3174094605651413e-05,
|
| 88 |
+
"loss": 0.0146,
|
| 89 |
+
"step": 1500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 8.0,
|
| 93 |
+
"eval_accuracy": 0.9097127318382263,
|
| 94 |
+
"eval_loss": 0.728517472743988,
|
| 95 |
+
"eval_runtime": 3.574,
|
| 96 |
+
"eval_samples_per_second": 204.533,
|
| 97 |
+
"eval_steps_per_second": 25.742,
|
| 98 |
+
"step": 1560
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 9.0,
|
| 102 |
+
"eval_accuracy": 0.9206566214561462,
|
| 103 |
+
"eval_loss": 0.6614168882369995,
|
| 104 |
+
"eval_runtime": 3.5326,
|
| 105 |
+
"eval_samples_per_second": 206.93,
|
| 106 |
+
"eval_steps_per_second": 26.043,
|
| 107 |
+
"step": 1755
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 10.0,
|
| 111 |
+
"eval_accuracy": 0.9151846766471863,
|
| 112 |
+
"eval_loss": 0.7393134236335754,
|
| 113 |
+
"eval_runtime": 3.6341,
|
| 114 |
+
"eval_samples_per_second": 201.15,
|
| 115 |
+
"eval_steps_per_second": 25.316,
|
| 116 |
+
"step": 1950
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 10.26,
|
| 120 |
+
"learning_rate": 1.8346158229474035e-05,
|
| 121 |
+
"loss": 0.0063,
|
| 122 |
+
"step": 2000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 11.0,
|
| 126 |
+
"eval_accuracy": 0.9165526628494263,
|
| 127 |
+
"eval_loss": 0.7556606531143188,
|
| 128 |
+
"eval_runtime": 3.6289,
|
| 129 |
+
"eval_samples_per_second": 201.437,
|
| 130 |
+
"eval_steps_per_second": 25.352,
|
| 131 |
+
"step": 2145
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 12.0,
|
| 135 |
+
"eval_accuracy": 0.9151846766471863,
|
| 136 |
+
"eval_loss": 0.7709615230560303,
|
| 137 |
+
"eval_runtime": 3.5866,
|
| 138 |
+
"eval_samples_per_second": 203.815,
|
| 139 |
+
"eval_steps_per_second": 25.651,
|
| 140 |
+
"step": 2340
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 12.82,
|
| 144 |
+
"learning_rate": 1.3518221853296657e-05,
|
| 145 |
+
"loss": 0.0037,
|
| 146 |
+
"step": 2500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 13.0,
|
| 150 |
+
"eval_accuracy": 0.9192886352539062,
|
| 151 |
+
"eval_loss": 0.7879533171653748,
|
| 152 |
+
"eval_runtime": 3.5996,
|
| 153 |
+
"eval_samples_per_second": 203.079,
|
| 154 |
+
"eval_steps_per_second": 25.558,
|
| 155 |
+
"step": 2535
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 14.0,
|
| 159 |
+
"eval_accuracy": 0.9124487042427063,
|
| 160 |
+
"eval_loss": 0.8315412402153015,
|
| 161 |
+
"eval_runtime": 3.763,
|
| 162 |
+
"eval_samples_per_second": 194.257,
|
| 163 |
+
"eval_steps_per_second": 24.448,
|
| 164 |
+
"step": 2730
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 15.0,
|
| 168 |
+
"eval_accuracy": 0.9179206490516663,
|
| 169 |
+
"eval_loss": 0.8191958069801331,
|
| 170 |
+
"eval_runtime": 3.6714,
|
| 171 |
+
"eval_samples_per_second": 199.108,
|
| 172 |
+
"eval_steps_per_second": 25.059,
|
| 173 |
+
"step": 2925
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 15.38,
|
| 177 |
+
"learning_rate": 8.690285477119281e-06,
|
| 178 |
+
"loss": 0.0001,
|
| 179 |
+
"step": 3000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 16.0,
|
| 183 |
+
"eval_accuracy": 0.9192886352539062,
|
| 184 |
+
"eval_loss": 0.8122023940086365,
|
| 185 |
+
"eval_runtime": 3.7565,
|
| 186 |
+
"eval_samples_per_second": 194.598,
|
| 187 |
+
"eval_steps_per_second": 24.491,
|
| 188 |
+
"step": 3120
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 17.0,
|
| 192 |
+
"eval_accuracy": 0.9179206490516663,
|
| 193 |
+
"eval_loss": 0.8226170539855957,
|
| 194 |
+
"eval_runtime": 3.6589,
|
| 195 |
+
"eval_samples_per_second": 199.788,
|
| 196 |
+
"eval_steps_per_second": 25.144,
|
| 197 |
+
"step": 3315
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 17.95,
|
| 201 |
+
"learning_rate": 3.862349100941902e-06,
|
| 202 |
+
"loss": 0.0001,
|
| 203 |
+
"step": 3500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 18.0,
|
| 207 |
+
"eval_accuracy": 0.9179206490516663,
|
| 208 |
+
"eval_loss": 0.8273718953132629,
|
| 209 |
+
"eval_runtime": 3.7025,
|
| 210 |
+
"eval_samples_per_second": 197.433,
|
| 211 |
+
"eval_steps_per_second": 24.848,
|
| 212 |
+
"step": 3510
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 19.0,
|
| 216 |
+
"eval_accuracy": 0.9179206490516663,
|
| 217 |
+
"eval_loss": 0.8303136825561523,
|
| 218 |
+
"eval_runtime": 3.6587,
|
| 219 |
+
"eval_samples_per_second": 199.797,
|
| 220 |
+
"eval_steps_per_second": 25.145,
|
| 221 |
+
"step": 3705
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 20.0,
|
| 225 |
+
"eval_accuracy": 0.9179206490516663,
|
| 226 |
+
"eval_loss": 0.8313552737236023,
|
| 227 |
+
"eval_runtime": 3.6572,
|
| 228 |
+
"eval_samples_per_second": 199.879,
|
| 229 |
+
"eval_steps_per_second": 25.156,
|
| 230 |
+
"step": 3900
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 20.0,
|
| 234 |
+
"step": 3900,
|
| 235 |
+
"total_flos": 4114530683719680.0,
|
| 236 |
+
"train_loss": 0.033037299305582656,
|
| 237 |
+
"train_runtime": 1549.9583,
|
| 238 |
+
"train_samples_per_second": 40.362,
|
| 239 |
+
"train_steps_per_second": 2.516
|
| 240 |
+
}
|
| 241 |
+
],
|
| 242 |
+
"max_steps": 3900,
|
| 243 |
+
"num_train_epochs": 20,
|
| 244 |
+
"total_flos": 4114530683719680.0,
|
| 245 |
+
"trial_name": null,
|
| 246 |
+
"trial_params": null
|
| 247 |
+
}
|