End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +457 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.9922182705152e+18,
|
| 4 |
+
"train_loss": 0.11038042506926758,
|
| 5 |
+
"train_runtime": 7101.5946,
|
| 6 |
+
"train_samples_per_second": 56.325,
|
| 7 |
+
"train_steps_per_second": 0.111
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 1.9922182705152e+18,
|
| 4 |
+
"train_loss": 0.11038042506926758,
|
| 5 |
+
"train_runtime": 7101.5946,
|
| 6 |
+
"train_samples_per_second": 56.325,
|
| 7 |
+
"train_steps_per_second": 0.111
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,457 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.05071697756648064,
|
| 3 |
+
"best_model_checkpoint": "./dinov2_Liveness_detection_v2.2.1/checkpoint-512",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 128,
|
| 6 |
+
"global_step": 785,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.10191082802547771,
|
| 13 |
+
"grad_norm": 3.0143625736236572,
|
| 14 |
+
"learning_rate": 4.923566878980892e-05,
|
| 15 |
+
"loss": 1.8573,
|
| 16 |
+
"step": 16
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.20382165605095542,
|
| 20 |
+
"grad_norm": 13.854424476623535,
|
| 21 |
+
"learning_rate": 4.821656050955414e-05,
|
| 22 |
+
"loss": 0.4389,
|
| 23 |
+
"step": 32
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.3057324840764331,
|
| 27 |
+
"grad_norm": 57.04545593261719,
|
| 28 |
+
"learning_rate": 4.7197452229299366e-05,
|
| 29 |
+
"loss": 0.3386,
|
| 30 |
+
"step": 48
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.40764331210191085,
|
| 34 |
+
"grad_norm": 26.68746566772461,
|
| 35 |
+
"learning_rate": 4.617834394904459e-05,
|
| 36 |
+
"loss": 0.3205,
|
| 37 |
+
"step": 64
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5095541401273885,
|
| 41 |
+
"grad_norm": 28.5941219329834,
|
| 42 |
+
"learning_rate": 4.5159235668789814e-05,
|
| 43 |
+
"loss": 0.2216,
|
| 44 |
+
"step": 80
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.6114649681528662,
|
| 48 |
+
"grad_norm": 7.417737007141113,
|
| 49 |
+
"learning_rate": 4.414012738853504e-05,
|
| 50 |
+
"loss": 0.1841,
|
| 51 |
+
"step": 96
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.7133757961783439,
|
| 55 |
+
"grad_norm": 15.305898666381836,
|
| 56 |
+
"learning_rate": 4.312101910828026e-05,
|
| 57 |
+
"loss": 0.2161,
|
| 58 |
+
"step": 112
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.8152866242038217,
|
| 62 |
+
"grad_norm": 65.58370208740234,
|
| 63 |
+
"learning_rate": 4.210191082802548e-05,
|
| 64 |
+
"loss": 0.183,
|
| 65 |
+
"step": 128
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.8152866242038217,
|
| 69 |
+
"eval_accuracy": 0.9016,
|
| 70 |
+
"eval_f1": 0.9039152443229508,
|
| 71 |
+
"eval_loss": 0.24732650816440582,
|
| 72 |
+
"eval_precision": 0.9122675566006839,
|
| 73 |
+
"eval_recall": 0.9016,
|
| 74 |
+
"eval_runtime": 285.4094,
|
| 75 |
+
"eval_samples_per_second": 70.075,
|
| 76 |
+
"eval_steps_per_second": 8.759,
|
| 77 |
+
"step": 128
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.9171974522292994,
|
| 81 |
+
"grad_norm": 13.938675880432129,
|
| 82 |
+
"learning_rate": 4.10828025477707e-05,
|
| 83 |
+
"loss": 0.1348,
|
| 84 |
+
"step": 144
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.019108280254777,
|
| 88 |
+
"grad_norm": 7.33504056930542,
|
| 89 |
+
"learning_rate": 4.0063694267515926e-05,
|
| 90 |
+
"loss": 0.1039,
|
| 91 |
+
"step": 160
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.1210191082802548,
|
| 95 |
+
"grad_norm": 29.384906768798828,
|
| 96 |
+
"learning_rate": 3.904458598726115e-05,
|
| 97 |
+
"loss": 0.1084,
|
| 98 |
+
"step": 176
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.2229299363057324,
|
| 102 |
+
"grad_norm": 14.105592727661133,
|
| 103 |
+
"learning_rate": 3.8025477707006374e-05,
|
| 104 |
+
"loss": 0.0855,
|
| 105 |
+
"step": 192
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 1.3248407643312101,
|
| 109 |
+
"grad_norm": 4.494396209716797,
|
| 110 |
+
"learning_rate": 3.700636942675159e-05,
|
| 111 |
+
"loss": 0.1292,
|
| 112 |
+
"step": 208
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.426751592356688,
|
| 116 |
+
"grad_norm": 15.541342735290527,
|
| 117 |
+
"learning_rate": 3.5987261146496815e-05,
|
| 118 |
+
"loss": 0.1187,
|
| 119 |
+
"step": 224
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.5286624203821657,
|
| 123 |
+
"grad_norm": 37.96527862548828,
|
| 124 |
+
"learning_rate": 3.496815286624204e-05,
|
| 125 |
+
"loss": 0.0875,
|
| 126 |
+
"step": 240
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.6305732484076434,
|
| 130 |
+
"grad_norm": 8.300914764404297,
|
| 131 |
+
"learning_rate": 3.394904458598726e-05,
|
| 132 |
+
"loss": 0.1022,
|
| 133 |
+
"step": 256
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.6305732484076434,
|
| 137 |
+
"eval_accuracy": 0.9729,
|
| 138 |
+
"eval_f1": 0.9727317036154979,
|
| 139 |
+
"eval_loss": 0.07502331584692001,
|
| 140 |
+
"eval_precision": 0.9737473833915566,
|
| 141 |
+
"eval_recall": 0.9729,
|
| 142 |
+
"eval_runtime": 271.7966,
|
| 143 |
+
"eval_samples_per_second": 73.584,
|
| 144 |
+
"eval_steps_per_second": 9.198,
|
| 145 |
+
"step": 256
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 1.732484076433121,
|
| 149 |
+
"grad_norm": 5.855906009674072,
|
| 150 |
+
"learning_rate": 3.2929936305732486e-05,
|
| 151 |
+
"loss": 0.0579,
|
| 152 |
+
"step": 272
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 1.8343949044585988,
|
| 156 |
+
"grad_norm": 6.417220115661621,
|
| 157 |
+
"learning_rate": 3.191082802547771e-05,
|
| 158 |
+
"loss": 0.0656,
|
| 159 |
+
"step": 288
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.9363057324840764,
|
| 163 |
+
"grad_norm": 5.988188743591309,
|
| 164 |
+
"learning_rate": 3.089171974522293e-05,
|
| 165 |
+
"loss": 0.0536,
|
| 166 |
+
"step": 304
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 2.038216560509554,
|
| 170 |
+
"grad_norm": 11.449820518493652,
|
| 171 |
+
"learning_rate": 2.9872611464968155e-05,
|
| 172 |
+
"loss": 0.05,
|
| 173 |
+
"step": 320
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.140127388535032,
|
| 177 |
+
"grad_norm": 8.527064323425293,
|
| 178 |
+
"learning_rate": 2.8853503184713375e-05,
|
| 179 |
+
"loss": 0.0399,
|
| 180 |
+
"step": 336
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 2.2420382165605095,
|
| 184 |
+
"grad_norm": 5.836983680725098,
|
| 185 |
+
"learning_rate": 2.78343949044586e-05,
|
| 186 |
+
"loss": 0.0518,
|
| 187 |
+
"step": 352
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.343949044585987,
|
| 191 |
+
"grad_norm": 26.618896484375,
|
| 192 |
+
"learning_rate": 2.6815286624203823e-05,
|
| 193 |
+
"loss": 0.0372,
|
| 194 |
+
"step": 368
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.445859872611465,
|
| 198 |
+
"grad_norm": 8.455528259277344,
|
| 199 |
+
"learning_rate": 2.5796178343949047e-05,
|
| 200 |
+
"loss": 0.0432,
|
| 201 |
+
"step": 384
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.445859872611465,
|
| 205 |
+
"eval_accuracy": 0.98205,
|
| 206 |
+
"eval_f1": 0.9819928803546507,
|
| 207 |
+
"eval_loss": 0.05750665441155434,
|
| 208 |
+
"eval_precision": 0.9823073850700162,
|
| 209 |
+
"eval_recall": 0.98205,
|
| 210 |
+
"eval_runtime": 276.5019,
|
| 211 |
+
"eval_samples_per_second": 72.332,
|
| 212 |
+
"eval_steps_per_second": 9.042,
|
| 213 |
+
"step": 384
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.5477707006369426,
|
| 217 |
+
"grad_norm": 15.16251277923584,
|
| 218 |
+
"learning_rate": 2.477707006369427e-05,
|
| 219 |
+
"loss": 0.0592,
|
| 220 |
+
"step": 400
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 2.6496815286624202,
|
| 224 |
+
"grad_norm": 5.268926620483398,
|
| 225 |
+
"learning_rate": 2.375796178343949e-05,
|
| 226 |
+
"loss": 0.0323,
|
| 227 |
+
"step": 416
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 2.7515923566878984,
|
| 231 |
+
"grad_norm": 5.213953018188477,
|
| 232 |
+
"learning_rate": 2.2738853503184715e-05,
|
| 233 |
+
"loss": 0.0247,
|
| 234 |
+
"step": 432
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.853503184713376,
|
| 238 |
+
"grad_norm": 4.638315677642822,
|
| 239 |
+
"learning_rate": 2.171974522292994e-05,
|
| 240 |
+
"loss": 0.0306,
|
| 241 |
+
"step": 448
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 2.9554140127388537,
|
| 245 |
+
"grad_norm": 6.414985179901123,
|
| 246 |
+
"learning_rate": 2.0700636942675162e-05,
|
| 247 |
+
"loss": 0.0302,
|
| 248 |
+
"step": 464
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 3.0573248407643314,
|
| 252 |
+
"grad_norm": 4.792835235595703,
|
| 253 |
+
"learning_rate": 1.9681528662420383e-05,
|
| 254 |
+
"loss": 0.0239,
|
| 255 |
+
"step": 480
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 3.159235668789809,
|
| 259 |
+
"grad_norm": 3.82145357131958,
|
| 260 |
+
"learning_rate": 1.8662420382165603e-05,
|
| 261 |
+
"loss": 0.016,
|
| 262 |
+
"step": 496
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 3.261146496815287,
|
| 266 |
+
"grad_norm": 6.392252445220947,
|
| 267 |
+
"learning_rate": 1.7643312101910827e-05,
|
| 268 |
+
"loss": 0.0247,
|
| 269 |
+
"step": 512
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.261146496815287,
|
| 273 |
+
"eval_accuracy": 0.9832,
|
| 274 |
+
"eval_f1": 0.9831797846161753,
|
| 275 |
+
"eval_loss": 0.05071697756648064,
|
| 276 |
+
"eval_precision": 0.983261693079801,
|
| 277 |
+
"eval_recall": 0.9832,
|
| 278 |
+
"eval_runtime": 271.679,
|
| 279 |
+
"eval_samples_per_second": 73.616,
|
| 280 |
+
"eval_steps_per_second": 9.202,
|
| 281 |
+
"step": 512
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.3630573248407645,
|
| 285 |
+
"grad_norm": 7.162614345550537,
|
| 286 |
+
"learning_rate": 1.662420382165605e-05,
|
| 287 |
+
"loss": 0.0135,
|
| 288 |
+
"step": 528
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.464968152866242,
|
| 292 |
+
"grad_norm": 3.0641708374023438,
|
| 293 |
+
"learning_rate": 1.5605095541401275e-05,
|
| 294 |
+
"loss": 0.0172,
|
| 295 |
+
"step": 544
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.56687898089172,
|
| 299 |
+
"grad_norm": 11.316327095031738,
|
| 300 |
+
"learning_rate": 1.4585987261146497e-05,
|
| 301 |
+
"loss": 0.0129,
|
| 302 |
+
"step": 560
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.6687898089171975,
|
| 306 |
+
"grad_norm": 9.987749099731445,
|
| 307 |
+
"learning_rate": 1.356687898089172e-05,
|
| 308 |
+
"loss": 0.0087,
|
| 309 |
+
"step": 576
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.770700636942675,
|
| 313 |
+
"grad_norm": 1.408227562904358,
|
| 314 |
+
"learning_rate": 1.2547770700636943e-05,
|
| 315 |
+
"loss": 0.0145,
|
| 316 |
+
"step": 592
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.872611464968153,
|
| 320 |
+
"grad_norm": 4.08975076675415,
|
| 321 |
+
"learning_rate": 1.1528662420382167e-05,
|
| 322 |
+
"loss": 0.0169,
|
| 323 |
+
"step": 608
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.9745222929936306,
|
| 327 |
+
"grad_norm": 11.745902061462402,
|
| 328 |
+
"learning_rate": 1.0509554140127389e-05,
|
| 329 |
+
"loss": 0.0122,
|
| 330 |
+
"step": 624
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 4.076433121019108,
|
| 334 |
+
"grad_norm": 1.0680924654006958,
|
| 335 |
+
"learning_rate": 9.490445859872613e-06,
|
| 336 |
+
"loss": 0.0115,
|
| 337 |
+
"step": 640
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 4.076433121019108,
|
| 341 |
+
"eval_accuracy": 0.98645,
|
| 342 |
+
"eval_f1": 0.9864145632656558,
|
| 343 |
+
"eval_loss": 0.053590141236782074,
|
| 344 |
+
"eval_precision": 0.9866191209109798,
|
| 345 |
+
"eval_recall": 0.98645,
|
| 346 |
+
"eval_runtime": 270.7038,
|
| 347 |
+
"eval_samples_per_second": 73.881,
|
| 348 |
+
"eval_steps_per_second": 9.235,
|
| 349 |
+
"step": 640
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 4.178343949044586,
|
| 353 |
+
"grad_norm": 7.269558429718018,
|
| 354 |
+
"learning_rate": 8.471337579617835e-06,
|
| 355 |
+
"loss": 0.0092,
|
| 356 |
+
"step": 656
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 4.280254777070064,
|
| 360 |
+
"grad_norm": 4.04400110244751,
|
| 361 |
+
"learning_rate": 7.452229299363057e-06,
|
| 362 |
+
"loss": 0.0065,
|
| 363 |
+
"step": 672
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 4.382165605095541,
|
| 367 |
+
"grad_norm": 0.2636276185512543,
|
| 368 |
+
"learning_rate": 6.433121019108281e-06,
|
| 369 |
+
"loss": 0.0028,
|
| 370 |
+
"step": 688
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 4.484076433121019,
|
| 374 |
+
"grad_norm": 11.116532325744629,
|
| 375 |
+
"learning_rate": 5.414012738853504e-06,
|
| 376 |
+
"loss": 0.0033,
|
| 377 |
+
"step": 704
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 4.585987261146497,
|
| 381 |
+
"grad_norm": 0.07053454965353012,
|
| 382 |
+
"learning_rate": 4.394904458598726e-06,
|
| 383 |
+
"loss": 0.0023,
|
| 384 |
+
"step": 720
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 4.687898089171974,
|
| 388 |
+
"grad_norm": 2.811331033706665,
|
| 389 |
+
"learning_rate": 3.375796178343949e-06,
|
| 390 |
+
"loss": 0.0057,
|
| 391 |
+
"step": 736
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 4.789808917197452,
|
| 395 |
+
"grad_norm": 0.1817609667778015,
|
| 396 |
+
"learning_rate": 2.3566878980891724e-06,
|
| 397 |
+
"loss": 0.0043,
|
| 398 |
+
"step": 752
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.89171974522293,
|
| 402 |
+
"grad_norm": 0.5368472933769226,
|
| 403 |
+
"learning_rate": 1.337579617834395e-06,
|
| 404 |
+
"loss": 0.002,
|
| 405 |
+
"step": 768
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 4.89171974522293,
|
| 409 |
+
"eval_accuracy": 0.98685,
|
| 410 |
+
"eval_f1": 0.9868193085416297,
|
| 411 |
+
"eval_loss": 0.06709764152765274,
|
| 412 |
+
"eval_precision": 0.9869900769633103,
|
| 413 |
+
"eval_recall": 0.98685,
|
| 414 |
+
"eval_runtime": 275.4886,
|
| 415 |
+
"eval_samples_per_second": 72.598,
|
| 416 |
+
"eval_steps_per_second": 9.075,
|
| 417 |
+
"step": 768
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.993630573248407,
|
| 421 |
+
"grad_norm": 2.216209888458252,
|
| 422 |
+
"learning_rate": 3.1847133757961787e-07,
|
| 423 |
+
"loss": 0.0012,
|
| 424 |
+
"step": 784
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 5.0,
|
| 428 |
+
"step": 785,
|
| 429 |
+
"total_flos": 1.9922182705152e+18,
|
| 430 |
+
"train_loss": 0.11038042506926758,
|
| 431 |
+
"train_runtime": 7101.5946,
|
| 432 |
+
"train_samples_per_second": 56.325,
|
| 433 |
+
"train_steps_per_second": 0.111
|
| 434 |
+
}
|
| 435 |
+
],
|
| 436 |
+
"logging_steps": 16,
|
| 437 |
+
"max_steps": 785,
|
| 438 |
+
"num_input_tokens_seen": 0,
|
| 439 |
+
"num_train_epochs": 5,
|
| 440 |
+
"save_steps": 128,
|
| 441 |
+
"stateful_callbacks": {
|
| 442 |
+
"TrainerControl": {
|
| 443 |
+
"args": {
|
| 444 |
+
"should_epoch_stop": false,
|
| 445 |
+
"should_evaluate": false,
|
| 446 |
+
"should_log": false,
|
| 447 |
+
"should_save": true,
|
| 448 |
+
"should_training_stop": true
|
| 449 |
+
},
|
| 450 |
+
"attributes": {}
|
| 451 |
+
}
|
| 452 |
+
},
|
| 453 |
+
"total_flos": 1.9922182705152e+18,
|
| 454 |
+
"train_batch_size": 512,
|
| 455 |
+
"trial_name": null,
|
| 456 |
+
"trial_params": null
|
| 457 |
+
}
|