End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +405 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.873417721518987,
|
| 3 |
+
"total_flos": 1.9321077770606223e+18,
|
| 4 |
+
"train_loss": 0.4781974340096498,
|
| 5 |
+
"train_runtime": 382.4209,
|
| 6 |
+
"train_samples_per_second": 66.027,
|
| 7 |
+
"train_steps_per_second": 1.02
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.873417721518987,
|
| 3 |
+
"total_flos": 1.9321077770606223e+18,
|
| 4 |
+
"train_loss": 0.4781974340096498,
|
| 5 |
+
"train_runtime": 382.4209,
|
| 6 |
+
"train_samples_per_second": 66.027,
|
| 7 |
+
"train_steps_per_second": 1.02
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,405 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7816455696202531,
|
| 3 |
+
"best_model_checkpoint": "tsec_vit_model/checkpoint-276",
|
| 4 |
+
"epoch": 9.873417721518987,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 390,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.25316455696202533,
|
| 13 |
+
"grad_norm": 1.1729109287261963,
|
| 14 |
+
"learning_rate": 1.282051282051282e-05,
|
| 15 |
+
"loss": 0.5143,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5063291139240507,
|
| 20 |
+
"grad_norm": 1.1752610206604004,
|
| 21 |
+
"learning_rate": 2.564102564102564e-05,
|
| 22 |
+
"loss": 0.559,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.759493670886076,
|
| 27 |
+
"grad_norm": 1.3603609800338745,
|
| 28 |
+
"learning_rate": 3.846153846153846e-05,
|
| 29 |
+
"loss": 0.4963,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9873417721518988,
|
| 34 |
+
"eval_accuracy": 0.6724683544303798,
|
| 35 |
+
"eval_loss": 0.5699456334114075,
|
| 36 |
+
"eval_runtime": 4.299,
|
| 37 |
+
"eval_samples_per_second": 147.012,
|
| 38 |
+
"eval_steps_per_second": 9.305,
|
| 39 |
+
"step": 39
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0126582278481013,
|
| 43 |
+
"grad_norm": 2.1996357440948486,
|
| 44 |
+
"learning_rate": 4.985754985754986e-05,
|
| 45 |
+
"loss": 0.5522,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.2658227848101267,
|
| 50 |
+
"grad_norm": 1.0542365312576294,
|
| 51 |
+
"learning_rate": 4.8433048433048433e-05,
|
| 52 |
+
"loss": 0.4911,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.518987341772152,
|
| 57 |
+
"grad_norm": 1.4736473560333252,
|
| 58 |
+
"learning_rate": 4.700854700854701e-05,
|
| 59 |
+
"loss": 0.5448,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.7721518987341773,
|
| 64 |
+
"grad_norm": 1.3524911403656006,
|
| 65 |
+
"learning_rate": 4.558404558404559e-05,
|
| 66 |
+
"loss": 0.4959,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.0,
|
| 71 |
+
"eval_accuracy": 0.7151898734177216,
|
| 72 |
+
"eval_loss": 0.5484605431556702,
|
| 73 |
+
"eval_runtime": 4.1719,
|
| 74 |
+
"eval_samples_per_second": 151.489,
|
| 75 |
+
"eval_steps_per_second": 9.588,
|
| 76 |
+
"step": 79
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.0253164556962027,
|
| 80 |
+
"grad_norm": 1.6714565753936768,
|
| 81 |
+
"learning_rate": 4.415954415954416e-05,
|
| 82 |
+
"loss": 0.4969,
|
| 83 |
+
"step": 80
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.278481012658228,
|
| 87 |
+
"grad_norm": 1.8887039422988892,
|
| 88 |
+
"learning_rate": 4.2735042735042735e-05,
|
| 89 |
+
"loss": 0.4863,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.5316455696202533,
|
| 94 |
+
"grad_norm": 2.640259027481079,
|
| 95 |
+
"learning_rate": 4.131054131054131e-05,
|
| 96 |
+
"loss": 0.5075,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.7848101265822782,
|
| 101 |
+
"grad_norm": 1.167457103729248,
|
| 102 |
+
"learning_rate": 3.988603988603989e-05,
|
| 103 |
+
"loss": 0.4879,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.9873417721518987,
|
| 108 |
+
"eval_accuracy": 0.7689873417721519,
|
| 109 |
+
"eval_loss": 0.48862743377685547,
|
| 110 |
+
"eval_runtime": 4.3228,
|
| 111 |
+
"eval_samples_per_second": 146.201,
|
| 112 |
+
"eval_steps_per_second": 9.253,
|
| 113 |
+
"step": 118
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.037974683544304,
|
| 117 |
+
"grad_norm": 1.0525858402252197,
|
| 118 |
+
"learning_rate": 3.846153846153846e-05,
|
| 119 |
+
"loss": 0.4946,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.291139240506329,
|
| 124 |
+
"grad_norm": 2.056792736053467,
|
| 125 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 126 |
+
"loss": 0.4884,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.5443037974683547,
|
| 131 |
+
"grad_norm": 1.7059816122055054,
|
| 132 |
+
"learning_rate": 3.561253561253561e-05,
|
| 133 |
+
"loss": 0.4926,
|
| 134 |
+
"step": 140
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.7974683544303796,
|
| 138 |
+
"grad_norm": 1.2786283493041992,
|
| 139 |
+
"learning_rate": 3.418803418803419e-05,
|
| 140 |
+
"loss": 0.5243,
|
| 141 |
+
"step": 150
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 4.0,
|
| 145 |
+
"eval_accuracy": 0.7468354430379747,
|
| 146 |
+
"eval_loss": 0.5133278965950012,
|
| 147 |
+
"eval_runtime": 4.3787,
|
| 148 |
+
"eval_samples_per_second": 144.335,
|
| 149 |
+
"eval_steps_per_second": 9.135,
|
| 150 |
+
"step": 158
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 4.050632911392405,
|
| 154 |
+
"grad_norm": 2.1460752487182617,
|
| 155 |
+
"learning_rate": 3.2763532763532764e-05,
|
| 156 |
+
"loss": 0.4983,
|
| 157 |
+
"step": 160
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.30379746835443,
|
| 161 |
+
"grad_norm": 1.596248984336853,
|
| 162 |
+
"learning_rate": 3.133903133903134e-05,
|
| 163 |
+
"loss": 0.4936,
|
| 164 |
+
"step": 170
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.556962025316456,
|
| 168 |
+
"grad_norm": 1.6535227298736572,
|
| 169 |
+
"learning_rate": 2.9914529914529915e-05,
|
| 170 |
+
"loss": 0.5049,
|
| 171 |
+
"step": 180
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.810126582278481,
|
| 175 |
+
"grad_norm": 1.3324358463287354,
|
| 176 |
+
"learning_rate": 2.8490028490028492e-05,
|
| 177 |
+
"loss": 0.4654,
|
| 178 |
+
"step": 190
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.987341772151899,
|
| 182 |
+
"eval_accuracy": 0.7515822784810127,
|
| 183 |
+
"eval_loss": 0.49274376034736633,
|
| 184 |
+
"eval_runtime": 4.3657,
|
| 185 |
+
"eval_samples_per_second": 144.763,
|
| 186 |
+
"eval_steps_per_second": 9.162,
|
| 187 |
+
"step": 197
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 5.063291139240507,
|
| 191 |
+
"grad_norm": 1.212802529335022,
|
| 192 |
+
"learning_rate": 2.706552706552707e-05,
|
| 193 |
+
"loss": 0.4828,
|
| 194 |
+
"step": 200
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.3164556962025316,
|
| 198 |
+
"grad_norm": 1.189599633216858,
|
| 199 |
+
"learning_rate": 2.564102564102564e-05,
|
| 200 |
+
"loss": 0.4816,
|
| 201 |
+
"step": 210
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.569620253164557,
|
| 205 |
+
"grad_norm": 1.3671377897262573,
|
| 206 |
+
"learning_rate": 2.4216524216524217e-05,
|
| 207 |
+
"loss": 0.4709,
|
| 208 |
+
"step": 220
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.822784810126582,
|
| 212 |
+
"grad_norm": 1.5688464641571045,
|
| 213 |
+
"learning_rate": 2.2792022792022794e-05,
|
| 214 |
+
"loss": 0.4776,
|
| 215 |
+
"step": 230
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 6.0,
|
| 219 |
+
"eval_accuracy": 0.7642405063291139,
|
| 220 |
+
"eval_loss": 0.4901277422904968,
|
| 221 |
+
"eval_runtime": 4.2593,
|
| 222 |
+
"eval_samples_per_second": 148.382,
|
| 223 |
+
"eval_steps_per_second": 9.391,
|
| 224 |
+
"step": 237
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 6.075949367088608,
|
| 228 |
+
"grad_norm": 1.6727008819580078,
|
| 229 |
+
"learning_rate": 2.1367521367521368e-05,
|
| 230 |
+
"loss": 0.4878,
|
| 231 |
+
"step": 240
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.329113924050633,
|
| 235 |
+
"grad_norm": 1.2271509170532227,
|
| 236 |
+
"learning_rate": 1.9943019943019945e-05,
|
| 237 |
+
"loss": 0.4725,
|
| 238 |
+
"step": 250
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.582278481012658,
|
| 242 |
+
"grad_norm": 1.5308541059494019,
|
| 243 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 244 |
+
"loss": 0.444,
|
| 245 |
+
"step": 260
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.8354430379746836,
|
| 249 |
+
"grad_norm": 1.6207118034362793,
|
| 250 |
+
"learning_rate": 1.7094017094017095e-05,
|
| 251 |
+
"loss": 0.4767,
|
| 252 |
+
"step": 270
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 6.987341772151899,
|
| 256 |
+
"eval_accuracy": 0.7816455696202531,
|
| 257 |
+
"eval_loss": 0.46520036458969116,
|
| 258 |
+
"eval_runtime": 4.1881,
|
| 259 |
+
"eval_samples_per_second": 150.905,
|
| 260 |
+
"eval_steps_per_second": 9.551,
|
| 261 |
+
"step": 276
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.0886075949367084,
|
| 265 |
+
"grad_norm": 1.3883824348449707,
|
| 266 |
+
"learning_rate": 1.566951566951567e-05,
|
| 267 |
+
"loss": 0.4649,
|
| 268 |
+
"step": 280
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.341772151898734,
|
| 272 |
+
"grad_norm": 2.2651216983795166,
|
| 273 |
+
"learning_rate": 1.4245014245014246e-05,
|
| 274 |
+
"loss": 0.4644,
|
| 275 |
+
"step": 290
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.594936708860759,
|
| 279 |
+
"grad_norm": 1.2660713195800781,
|
| 280 |
+
"learning_rate": 1.282051282051282e-05,
|
| 281 |
+
"loss": 0.4448,
|
| 282 |
+
"step": 300
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.848101265822785,
|
| 286 |
+
"grad_norm": 1.8906290531158447,
|
| 287 |
+
"learning_rate": 1.1396011396011397e-05,
|
| 288 |
+
"loss": 0.4465,
|
| 289 |
+
"step": 310
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 8.0,
|
| 293 |
+
"eval_accuracy": 0.7642405063291139,
|
| 294 |
+
"eval_loss": 0.4795072674751282,
|
| 295 |
+
"eval_runtime": 4.3585,
|
| 296 |
+
"eval_samples_per_second": 145.004,
|
| 297 |
+
"eval_steps_per_second": 9.177,
|
| 298 |
+
"step": 316
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 8.10126582278481,
|
| 302 |
+
"grad_norm": 1.7846542596817017,
|
| 303 |
+
"learning_rate": 9.971509971509972e-06,
|
| 304 |
+
"loss": 0.4552,
|
| 305 |
+
"step": 320
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 8.354430379746836,
|
| 309 |
+
"grad_norm": 1.2796516418457031,
|
| 310 |
+
"learning_rate": 8.547008547008548e-06,
|
| 311 |
+
"loss": 0.4284,
|
| 312 |
+
"step": 330
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 8.60759493670886,
|
| 316 |
+
"grad_norm": 1.181015133857727,
|
| 317 |
+
"learning_rate": 7.122507122507123e-06,
|
| 318 |
+
"loss": 0.3985,
|
| 319 |
+
"step": 340
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.860759493670885,
|
| 323 |
+
"grad_norm": 1.352704644203186,
|
| 324 |
+
"learning_rate": 5.6980056980056985e-06,
|
| 325 |
+
"loss": 0.467,
|
| 326 |
+
"step": 350
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.987341772151899,
|
| 330 |
+
"eval_accuracy": 0.7484177215189873,
|
| 331 |
+
"eval_loss": 0.4690525233745575,
|
| 332 |
+
"eval_runtime": 4.2658,
|
| 333 |
+
"eval_samples_per_second": 148.154,
|
| 334 |
+
"eval_steps_per_second": 9.377,
|
| 335 |
+
"step": 355
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 9.113924050632912,
|
| 339 |
+
"grad_norm": 1.3692480325698853,
|
| 340 |
+
"learning_rate": 4.273504273504274e-06,
|
| 341 |
+
"loss": 0.437,
|
| 342 |
+
"step": 360
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 9.367088607594937,
|
| 346 |
+
"grad_norm": 2.6086838245391846,
|
| 347 |
+
"learning_rate": 2.8490028490028492e-06,
|
| 348 |
+
"loss": 0.4062,
|
| 349 |
+
"step": 370
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 9.620253164556962,
|
| 353 |
+
"grad_norm": 1.352748155593872,
|
| 354 |
+
"learning_rate": 1.4245014245014246e-06,
|
| 355 |
+
"loss": 0.4365,
|
| 356 |
+
"step": 380
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 9.873417721518987,
|
| 360 |
+
"grad_norm": 1.9576495885849,
|
| 361 |
+
"learning_rate": 0.0,
|
| 362 |
+
"loss": 0.4121,
|
| 363 |
+
"step": 390
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 9.873417721518987,
|
| 367 |
+
"eval_accuracy": 0.7689873417721519,
|
| 368 |
+
"eval_loss": 0.482129842042923,
|
| 369 |
+
"eval_runtime": 4.3754,
|
| 370 |
+
"eval_samples_per_second": 144.443,
|
| 371 |
+
"eval_steps_per_second": 9.142,
|
| 372 |
+
"step": 390
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 9.873417721518987,
|
| 376 |
+
"step": 390,
|
| 377 |
+
"total_flos": 1.9321077770606223e+18,
|
| 378 |
+
"train_loss": 0.4781974340096498,
|
| 379 |
+
"train_runtime": 382.4209,
|
| 380 |
+
"train_samples_per_second": 66.027,
|
| 381 |
+
"train_steps_per_second": 1.02
|
| 382 |
+
}
|
| 383 |
+
],
|
| 384 |
+
"logging_steps": 10,
|
| 385 |
+
"max_steps": 390,
|
| 386 |
+
"num_input_tokens_seen": 0,
|
| 387 |
+
"num_train_epochs": 10,
|
| 388 |
+
"save_steps": 500,
|
| 389 |
+
"stateful_callbacks": {
|
| 390 |
+
"TrainerControl": {
|
| 391 |
+
"args": {
|
| 392 |
+
"should_epoch_stop": false,
|
| 393 |
+
"should_evaluate": false,
|
| 394 |
+
"should_log": false,
|
| 395 |
+
"should_save": true,
|
| 396 |
+
"should_training_stop": true
|
| 397 |
+
},
|
| 398 |
+
"attributes": {}
|
| 399 |
+
}
|
| 400 |
+
},
|
| 401 |
+
"total_flos": 1.9321077770606223e+18,
|
| 402 |
+
"train_batch_size": 16,
|
| 403 |
+
"trial_name": null,
|
| 404 |
+
"trial_params": null
|
| 405 |
+
}
|