Commit ·
e41802d
1
Parent(s): 4be1171
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": 0.9097127318382263,
|
| 13 |
+
"eval_loss": 0.2761473059654236,
|
| 14 |
+
"eval_runtime": 4.9719,
|
| 15 |
+
"eval_samples_per_second": 147.026,
|
| 16 |
+
"eval_steps_per_second": 18.504,
|
| 17 |
+
"step": 195
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 2.0,
|
| 21 |
+
"eval_accuracy": 0.9220246076583862,
|
| 22 |
+
"eval_loss": 0.310268372297287,
|
| 23 |
+
"eval_runtime": 4.9488,
|
| 24 |
+
"eval_samples_per_second": 147.712,
|
| 25 |
+
"eval_steps_per_second": 18.59,
|
| 26 |
+
"step": 390
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.56,
|
| 30 |
+
"learning_rate": 4.087454207611332e-05,
|
| 31 |
+
"loss": 0.2255,
|
| 32 |
+
"step": 500
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 3.0,
|
| 36 |
+
"eval_accuracy": 0.9124487042427063,
|
| 37 |
+
"eval_loss": 0.29651716351509094,
|
| 38 |
+
"eval_runtime": 4.9678,
|
| 39 |
+
"eval_samples_per_second": 147.148,
|
| 40 |
+
"eval_steps_per_second": 18.519,
|
| 41 |
+
"step": 585
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 4.0,
|
| 45 |
+
"eval_accuracy": 0.9233925938606262,
|
| 46 |
+
"eval_loss": 0.3387739956378937,
|
| 47 |
+
"eval_runtime": 4.9332,
|
| 48 |
+
"eval_samples_per_second": 148.181,
|
| 49 |
+
"eval_steps_per_second": 18.649,
|
| 50 |
+
"step": 780
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 5.0,
|
| 54 |
+
"eval_accuracy": 0.9206566214561462,
|
| 55 |
+
"eval_loss": 0.42386311292648315,
|
| 56 |
+
"eval_runtime": 5.0121,
|
| 57 |
+
"eval_samples_per_second": 145.847,
|
| 58 |
+
"eval_steps_per_second": 18.356,
|
| 59 |
+
"step": 975
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 5.13,
|
| 63 |
+
"learning_rate": 3.486358000609665e-05,
|
| 64 |
+
"loss": 0.1137,
|
| 65 |
+
"step": 1000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 6.0,
|
| 69 |
+
"eval_accuracy": 0.9179206490516663,
|
| 70 |
+
"eval_loss": 0.43168818950653076,
|
| 71 |
+
"eval_runtime": 4.956,
|
| 72 |
+
"eval_samples_per_second": 147.497,
|
| 73 |
+
"eval_steps_per_second": 18.563,
|
| 74 |
+
"step": 1170
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 7.0,
|
| 78 |
+
"eval_accuracy": 0.9206566214561462,
|
| 79 |
+
"eval_loss": 0.5392778515815735,
|
| 80 |
+
"eval_runtime": 5.1536,
|
| 81 |
+
"eval_samples_per_second": 141.844,
|
| 82 |
+
"eval_steps_per_second": 17.852,
|
| 83 |
+
"step": 1365
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 7.69,
|
| 87 |
+
"learning_rate": 2.885261793607999e-05,
|
| 88 |
+
"loss": 0.0469,
|
| 89 |
+
"step": 1500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 8.0,
|
| 93 |
+
"eval_accuracy": 0.9151846766471863,
|
| 94 |
+
"eval_loss": 0.46178123354911804,
|
| 95 |
+
"eval_runtime": 4.9372,
|
| 96 |
+
"eval_samples_per_second": 148.059,
|
| 97 |
+
"eval_steps_per_second": 18.634,
|
| 98 |
+
"step": 1560
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 9.0,
|
| 102 |
+
"eval_accuracy": 0.9138166904449463,
|
| 103 |
+
"eval_loss": 0.6065310835838318,
|
| 104 |
+
"eval_runtime": 4.9633,
|
| 105 |
+
"eval_samples_per_second": 147.282,
|
| 106 |
+
"eval_steps_per_second": 18.536,
|
| 107 |
+
"step": 1755
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 10.0,
|
| 111 |
+
"eval_accuracy": 0.9165526628494263,
|
| 112 |
+
"eval_loss": 0.5557906031608582,
|
| 113 |
+
"eval_runtime": 5.0575,
|
| 114 |
+
"eval_samples_per_second": 144.538,
|
| 115 |
+
"eval_steps_per_second": 18.191,
|
| 116 |
+
"step": 1950
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 10.26,
|
| 120 |
+
"learning_rate": 2.2841655866063323e-05,
|
| 121 |
+
"loss": 0.0283,
|
| 122 |
+
"step": 2000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 11.0,
|
| 126 |
+
"eval_accuracy": 0.9247605800628662,
|
| 127 |
+
"eval_loss": 0.5483647584915161,
|
| 128 |
+
"eval_runtime": 4.8959,
|
| 129 |
+
"eval_samples_per_second": 149.308,
|
| 130 |
+
"eval_steps_per_second": 18.791,
|
| 131 |
+
"step": 2145
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 12.0,
|
| 135 |
+
"eval_accuracy": 0.9151846766471863,
|
| 136 |
+
"eval_loss": 0.7020018100738525,
|
| 137 |
+
"eval_runtime": 5.0449,
|
| 138 |
+
"eval_samples_per_second": 144.899,
|
| 139 |
+
"eval_steps_per_second": 18.236,
|
| 140 |
+
"step": 2340
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 12.82,
|
| 144 |
+
"learning_rate": 1.6830693796046662e-05,
|
| 145 |
+
"loss": 0.015,
|
| 146 |
+
"step": 2500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 13.0,
|
| 150 |
+
"eval_accuracy": 0.9192886352539062,
|
| 151 |
+
"eval_loss": 0.7031099796295166,
|
| 152 |
+
"eval_runtime": 4.8879,
|
| 153 |
+
"eval_samples_per_second": 149.552,
|
| 154 |
+
"eval_steps_per_second": 18.822,
|
| 155 |
+
"step": 2535
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 14.0,
|
| 159 |
+
"eval_accuracy": 0.9151846766471863,
|
| 160 |
+
"eval_loss": 0.7629717588424683,
|
| 161 |
+
"eval_runtime": 5.0202,
|
| 162 |
+
"eval_samples_per_second": 145.613,
|
| 163 |
+
"eval_steps_per_second": 18.326,
|
| 164 |
+
"step": 2730
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 15.0,
|
| 168 |
+
"eval_accuracy": 0.9165526628494263,
|
| 169 |
+
"eval_loss": 0.7813730835914612,
|
| 170 |
+
"eval_runtime": 4.9102,
|
| 171 |
+
"eval_samples_per_second": 148.874,
|
| 172 |
+
"eval_steps_per_second": 18.737,
|
| 173 |
+
"step": 2925
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 15.38,
|
| 177 |
+
"learning_rate": 1.0819731726029997e-05,
|
| 178 |
+
"loss": 0.0043,
|
| 179 |
+
"step": 3000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 16.0,
|
| 183 |
+
"eval_accuracy": 0.9179206490516663,
|
| 184 |
+
"eval_loss": 0.7715274691581726,
|
| 185 |
+
"eval_runtime": 5.0959,
|
| 186 |
+
"eval_samples_per_second": 143.449,
|
| 187 |
+
"eval_steps_per_second": 18.054,
|
| 188 |
+
"step": 3120
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 17.0,
|
| 192 |
+
"eval_accuracy": 0.9192886352539062,
|
| 193 |
+
"eval_loss": 0.7615724802017212,
|
| 194 |
+
"eval_runtime": 4.852,
|
| 195 |
+
"eval_samples_per_second": 150.659,
|
| 196 |
+
"eval_steps_per_second": 18.961,
|
| 197 |
+
"step": 3315
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 17.95,
|
| 201 |
+
"learning_rate": 4.808769656013331e-06,
|
| 202 |
+
"loss": 0.0031,
|
| 203 |
+
"step": 3500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 18.0,
|
| 207 |
+
"eval_accuracy": 0.9220246076583862,
|
| 208 |
+
"eval_loss": 0.7719457745552063,
|
| 209 |
+
"eval_runtime": 5.0863,
|
| 210 |
+
"eval_samples_per_second": 143.72,
|
| 211 |
+
"eval_steps_per_second": 18.088,
|
| 212 |
+
"step": 3510
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 19.0,
|
| 216 |
+
"eval_accuracy": 0.9192886352539062,
|
| 217 |
+
"eval_loss": 0.779067873954773,
|
| 218 |
+
"eval_runtime": 4.9894,
|
| 219 |
+
"eval_samples_per_second": 146.509,
|
| 220 |
+
"eval_steps_per_second": 18.439,
|
| 221 |
+
"step": 3705
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 20.0,
|
| 225 |
+
"eval_accuracy": 0.9192886352539062,
|
| 226 |
+
"eval_loss": 0.7817478775978088,
|
| 227 |
+
"eval_runtime": 4.8884,
|
| 228 |
+
"eval_samples_per_second": 149.538,
|
| 229 |
+
"eval_steps_per_second": 18.82,
|
| 230 |
+
"step": 3900
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 20.0,
|
| 234 |
+
"step": 3900,
|
| 235 |
+
"total_flos": 4114604473061376.0,
|
| 236 |
+
"train_loss": 0.056036626369907305,
|
| 237 |
+
"train_runtime": 3116.2149,
|
| 238 |
+
"train_samples_per_second": 20.076,
|
| 239 |
+
"train_steps_per_second": 1.252
|
| 240 |
+
}
|
| 241 |
+
],
|
| 242 |
+
"max_steps": 3900,
|
| 243 |
+
"num_train_epochs": 20,
|
| 244 |
+
"total_flos": 4114604473061376.0,
|
| 245 |
+
"trial_name": null,
|
| 246 |
+
"trial_params": null
|
| 247 |
+
}
|