Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +630 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,630 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 1505,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.013289036544850499,
|
| 13 |
+
"grad_norm": 0.10525072365999222,
|
| 14 |
+
"learning_rate": 4.9978216198586135e-05,
|
| 15 |
+
"loss": 0.6155,
|
| 16 |
+
"step": 20
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.026578073089700997,
|
| 20 |
+
"grad_norm": 0.08554615080356598,
|
| 21 |
+
"learning_rate": 4.991290275706486e-05,
|
| 22 |
+
"loss": 0.5694,
|
| 23 |
+
"step": 40
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.03986710963455149,
|
| 27 |
+
"grad_norm": 0.08361516892910004,
|
| 28 |
+
"learning_rate": 4.980417349743936e-05,
|
| 29 |
+
"loss": 0.557,
|
| 30 |
+
"step": 60
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.053156146179401995,
|
| 34 |
+
"grad_norm": 0.08680060505867004,
|
| 35 |
+
"learning_rate": 4.9652217902637596e-05,
|
| 36 |
+
"loss": 0.548,
|
| 37 |
+
"step": 80
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.0664451827242525,
|
| 41 |
+
"grad_norm": 0.08960291743278503,
|
| 42 |
+
"learning_rate": 4.945730078629964e-05,
|
| 43 |
+
"loss": 0.5427,
|
| 44 |
+
"step": 100
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07973421926910298,
|
| 48 |
+
"grad_norm": 0.09262242168188095,
|
| 49 |
+
"learning_rate": 4.921976183128585e-05,
|
| 50 |
+
"loss": 0.5384,
|
| 51 |
+
"step": 120
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.09302325581395349,
|
| 55 |
+
"grad_norm": 0.08780515193939209,
|
| 56 |
+
"learning_rate": 4.894001499771015e-05,
|
| 57 |
+
"loss": 0.5362,
|
| 58 |
+
"step": 140
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.10631229235880399,
|
| 62 |
+
"grad_norm": 0.09249912202358246,
|
| 63 |
+
"learning_rate": 4.861854780153004e-05,
|
| 64 |
+
"loss": 0.5324,
|
| 65 |
+
"step": 160
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11960132890365449,
|
| 69 |
+
"grad_norm": 0.09562400728464127,
|
| 70 |
+
"learning_rate": 4.825592046495054e-05,
|
| 71 |
+
"loss": 0.5311,
|
| 72 |
+
"step": 180
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.132890365448505,
|
| 76 |
+
"grad_norm": 0.09372778236865997,
|
| 77 |
+
"learning_rate": 4.785276494012263e-05,
|
| 78 |
+
"loss": 0.5278,
|
| 79 |
+
"step": 200
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.132890365448505,
|
| 83 |
+
"eval_accuracy": 0.19452303794312395,
|
| 84 |
+
"eval_loss": 0.5592088103294373,
|
| 85 |
+
"eval_runtime": 19.5284,
|
| 86 |
+
"eval_samples_per_second": 93.914,
|
| 87 |
+
"eval_steps_per_second": 0.41,
|
| 88 |
+
"step": 200
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.1461794019933555,
|
| 92 |
+
"grad_norm": 0.08762918412685394,
|
| 93 |
+
"learning_rate": 4.740978380783765e-05,
|
| 94 |
+
"loss": 0.5253,
|
| 95 |
+
"step": 220
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.15946843853820597,
|
| 99 |
+
"grad_norm": 0.08518578112125397,
|
| 100 |
+
"learning_rate": 4.6927749053136866e-05,
|
| 101 |
+
"loss": 0.5192,
|
| 102 |
+
"step": 240
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.17275747508305647,
|
| 106 |
+
"grad_norm": 0.09664598107337952,
|
| 107 |
+
"learning_rate": 4.640750071996995e-05,
|
| 108 |
+
"loss": 0.5217,
|
| 109 |
+
"step": 260
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.18604651162790697,
|
| 113 |
+
"grad_norm": 0.08245342969894409,
|
| 114 |
+
"learning_rate": 4.584994544724695e-05,
|
| 115 |
+
"loss": 0.5172,
|
| 116 |
+
"step": 280
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.19933554817275748,
|
| 120 |
+
"grad_norm": 0.08551981300115585,
|
| 121 |
+
"learning_rate": 4.5256054888834934e-05,
|
| 122 |
+
"loss": 0.5152,
|
| 123 |
+
"step": 300
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.21262458471760798,
|
| 127 |
+
"grad_norm": 0.09647104889154434,
|
| 128 |
+
"learning_rate": 4.4626864020252774e-05,
|
| 129 |
+
"loss": 0.5139,
|
| 130 |
+
"step": 320
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.22591362126245848,
|
| 134 |
+
"grad_norm": 0.09810427576303482,
|
| 135 |
+
"learning_rate": 4.3963469335015085e-05,
|
| 136 |
+
"loss": 0.5129,
|
| 137 |
+
"step": 340
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.23920265780730898,
|
| 141 |
+
"grad_norm": 0.08342389762401581,
|
| 142 |
+
"learning_rate": 4.326702693376844e-05,
|
| 143 |
+
"loss": 0.5119,
|
| 144 |
+
"step": 360
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.25249169435215946,
|
| 148 |
+
"grad_norm": 0.08738644421100616,
|
| 149 |
+
"learning_rate": 4.2538750509550054e-05,
|
| 150 |
+
"loss": 0.511,
|
| 151 |
+
"step": 380
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.26578073089701,
|
| 155 |
+
"grad_norm": 0.08475251495838165,
|
| 156 |
+
"learning_rate": 4.177990923267986e-05,
|
| 157 |
+
"loss": 0.5117,
|
| 158 |
+
"step": 400
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.26578073089701,
|
| 162 |
+
"eval_accuracy": 0.1953402564276045,
|
| 163 |
+
"eval_loss": 0.5438870787620544,
|
| 164 |
+
"eval_runtime": 15.5302,
|
| 165 |
+
"eval_samples_per_second": 118.093,
|
| 166 |
+
"eval_steps_per_second": 0.515,
|
| 167 |
+
"step": 400
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.27906976744186046,
|
| 171 |
+
"grad_norm": 0.07873477786779404,
|
| 172 |
+
"learning_rate": 4.099182553897229e-05,
|
| 173 |
+
"loss": 0.5084,
|
| 174 |
+
"step": 420
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.292358803986711,
|
| 178 |
+
"grad_norm": 0.09158772230148315,
|
| 179 |
+
"learning_rate": 4.017587282512181e-05,
|
| 180 |
+
"loss": 0.5065,
|
| 181 |
+
"step": 440
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.30564784053156147,
|
| 185 |
+
"grad_norm": 0.07729614526033401,
|
| 186 |
+
"learning_rate": 3.933347305527898e-05,
|
| 187 |
+
"loss": 0.5047,
|
| 188 |
+
"step": 460
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.31893687707641194,
|
| 192 |
+
"grad_norm": 0.08530613034963608,
|
| 193 |
+
"learning_rate": 3.846609428298757e-05,
|
| 194 |
+
"loss": 0.5049,
|
| 195 |
+
"step": 480
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.33222591362126247,
|
| 199 |
+
"grad_norm": 0.07760792225599289,
|
| 200 |
+
"learning_rate": 3.7575248092801686e-05,
|
| 201 |
+
"loss": 0.5035,
|
| 202 |
+
"step": 500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.34551495016611294,
|
| 206 |
+
"grad_norm": 0.08521712571382523,
|
| 207 |
+
"learning_rate": 3.66624869660411e-05,
|
| 208 |
+
"loss": 0.5042,
|
| 209 |
+
"step": 520
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.3588039867109635,
|
| 213 |
+
"grad_norm": 0.08439727872610092,
|
| 214 |
+
"learning_rate": 3.572940157527572e-05,
|
| 215 |
+
"loss": 0.5021,
|
| 216 |
+
"step": 540
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 0.37209302325581395,
|
| 220 |
+
"grad_norm": 0.09042590111494064,
|
| 221 |
+
"learning_rate": 3.47776180122539e-05,
|
| 222 |
+
"loss": 0.5019,
|
| 223 |
+
"step": 560
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 0.3853820598006645,
|
| 227 |
+
"grad_norm": 0.08219762146472931,
|
| 228 |
+
"learning_rate": 3.3808794954105716e-05,
|
| 229 |
+
"loss": 0.501,
|
| 230 |
+
"step": 580
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.39867109634551495,
|
| 234 |
+
"grad_norm": 0.08426713198423386,
|
| 235 |
+
"learning_rate": 3.282462077275947e-05,
|
| 236 |
+
"loss": 0.5013,
|
| 237 |
+
"step": 600
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.39867109634551495,
|
| 241 |
+
"eval_accuracy": 0.19588631180347973,
|
| 242 |
+
"eval_loss": 0.5341373682022095,
|
| 243 |
+
"eval_runtime": 16.1072,
|
| 244 |
+
"eval_samples_per_second": 113.862,
|
| 245 |
+
"eval_steps_per_second": 0.497,
|
| 246 |
+
"step": 600
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 0.4119601328903654,
|
| 250 |
+
"grad_norm": 0.08020314574241638,
|
| 251 |
+
"learning_rate": 3.1826810592609036e-05,
|
| 252 |
+
"loss": 0.4968,
|
| 253 |
+
"step": 620
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 0.42524916943521596,
|
| 257 |
+
"grad_norm": 0.07975760847330093,
|
| 258 |
+
"learning_rate": 3.081710330155942e-05,
|
| 259 |
+
"loss": 0.4997,
|
| 260 |
+
"step": 640
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.43853820598006643,
|
| 264 |
+
"grad_norm": 0.08056964725255966,
|
| 265 |
+
"learning_rate": 2.979725852065981e-05,
|
| 266 |
+
"loss": 0.4968,
|
| 267 |
+
"step": 660
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.45182724252491696,
|
| 271 |
+
"grad_norm": 0.08022565394639969,
|
| 272 |
+
"learning_rate": 2.876905353760459e-05,
|
| 273 |
+
"loss": 0.4976,
|
| 274 |
+
"step": 680
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.46511627906976744,
|
| 278 |
+
"grad_norm": 0.08131925761699677,
|
| 279 |
+
"learning_rate": 2.7734280209446865e-05,
|
| 280 |
+
"loss": 0.4973,
|
| 281 |
+
"step": 700
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.47840531561461797,
|
| 285 |
+
"grad_norm": 0.07562076300382614,
|
| 286 |
+
"learning_rate": 2.6694741839921732e-05,
|
| 287 |
+
"loss": 0.4956,
|
| 288 |
+
"step": 720
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.49169435215946844,
|
| 292 |
+
"grad_norm": 0.07877329736948013,
|
| 293 |
+
"learning_rate": 2.5652250036821523e-05,
|
| 294 |
+
"loss": 0.4966,
|
| 295 |
+
"step": 740
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.5049833887043189,
|
| 299 |
+
"grad_norm": 0.08014395087957382,
|
| 300 |
+
"learning_rate": 2.4608621554899362e-05,
|
| 301 |
+
"loss": 0.4934,
|
| 302 |
+
"step": 760
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.5182724252491694,
|
| 306 |
+
"grad_norm": 0.07770328223705292,
|
| 307 |
+
"learning_rate": 2.356567512980326e-05,
|
| 308 |
+
"loss": 0.4934,
|
| 309 |
+
"step": 780
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.53156146179402,
|
| 313 |
+
"grad_norm": 0.07732851803302765,
|
| 314 |
+
"learning_rate": 2.252522830855798e-05,
|
| 315 |
+
"loss": 0.4951,
|
| 316 |
+
"step": 800
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 0.53156146179402,
|
| 320 |
+
"eval_accuracy": 0.19623978277118043,
|
| 321 |
+
"eval_loss": 0.5274041295051575,
|
| 322 |
+
"eval_runtime": 16.4552,
|
| 323 |
+
"eval_samples_per_second": 111.454,
|
| 324 |
+
"eval_steps_per_second": 0.486,
|
| 325 |
+
"step": 800
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.5448504983388704,
|
| 329 |
+
"grad_norm": 0.07608461380004883,
|
| 330 |
+
"learning_rate": 2.1489094282118395e-05,
|
| 331 |
+
"loss": 0.4896,
|
| 332 |
+
"step": 820
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.5581395348837209,
|
| 336 |
+
"grad_norm": 0.07657533138990402,
|
| 337 |
+
"learning_rate": 2.0459078725514092e-05,
|
| 338 |
+
"loss": 0.4918,
|
| 339 |
+
"step": 840
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.5714285714285714,
|
| 343 |
+
"grad_norm": 0.07983728498220444,
|
| 344 |
+
"learning_rate": 1.9436976651092144e-05,
|
| 345 |
+
"loss": 0.4927,
|
| 346 |
+
"step": 860
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.584717607973422,
|
| 350 |
+
"grad_norm": 0.07355430722236633,
|
| 351 |
+
"learning_rate": 1.8424569280341653e-05,
|
| 352 |
+
"loss": 0.493,
|
| 353 |
+
"step": 880
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.5980066445182725,
|
| 357 |
+
"grad_norm": 0.08014149218797684,
|
| 358 |
+
"learning_rate": 1.7423620939751788e-05,
|
| 359 |
+
"loss": 0.4922,
|
| 360 |
+
"step": 900
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.6112956810631229,
|
| 364 |
+
"grad_norm": 0.07500924915075302,
|
| 365 |
+
"learning_rate": 1.6435875986112685e-05,
|
| 366 |
+
"loss": 0.491,
|
| 367 |
+
"step": 920
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.6245847176079734,
|
| 371 |
+
"grad_norm": 0.07356715947389603,
|
| 372 |
+
"learning_rate": 1.546305576661776e-05,
|
| 373 |
+
"loss": 0.4909,
|
| 374 |
+
"step": 940
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.6378737541528239,
|
| 378 |
+
"grad_norm": 0.07140863686800003,
|
| 379 |
+
"learning_rate": 1.4506855619064846e-05,
|
| 380 |
+
"loss": 0.489,
|
| 381 |
+
"step": 960
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.6511627906976745,
|
| 385 |
+
"grad_norm": 0.07692987471818924,
|
| 386 |
+
"learning_rate": 1.3568941917384036e-05,
|
| 387 |
+
"loss": 0.4902,
|
| 388 |
+
"step": 980
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.6644518272425249,
|
| 392 |
+
"grad_norm": 0.07356040179729462,
|
| 393 |
+
"learning_rate": 1.2650949167640997e-05,
|
| 394 |
+
"loss": 0.4894,
|
| 395 |
+
"step": 1000
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.6644518272425249,
|
| 399 |
+
"eval_accuracy": 0.19652373156663552,
|
| 400 |
+
"eval_loss": 0.5229406952857971,
|
| 401 |
+
"eval_runtime": 15.6791,
|
| 402 |
+
"eval_samples_per_second": 116.971,
|
| 403 |
+
"eval_steps_per_second": 0.51,
|
| 404 |
+
"step": 1000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.6777408637873754,
|
| 408 |
+
"grad_norm": 0.0691773071885109,
|
| 409 |
+
"learning_rate": 1.1754477159576499e-05,
|
| 410 |
+
"loss": 0.4869,
|
| 411 |
+
"step": 1020
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 0.6910299003322259,
|
| 415 |
+
"grad_norm": 0.07505939155817032,
|
| 416 |
+
"learning_rate": 1.088108817864629e-05,
|
| 417 |
+
"loss": 0.4865,
|
| 418 |
+
"step": 1040
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.7043189368770764,
|
| 422 |
+
"grad_norm": 0.06973451375961304,
|
| 423 |
+
"learning_rate": 1.003230428341979e-05,
|
| 424 |
+
"loss": 0.4888,
|
| 425 |
+
"step": 1060
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.717607973421927,
|
| 429 |
+
"grad_norm": 0.07225219160318375,
|
| 430 |
+
"learning_rate": 9.209604653082326e-06,
|
| 431 |
+
"loss": 0.4858,
|
| 432 |
+
"step": 1080
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.7308970099667774,
|
| 436 |
+
"grad_norm": 0.07558443397283554,
|
| 437 |
+
"learning_rate": 8.414423009663563e-06,
|
| 438 |
+
"loss": 0.4891,
|
| 439 |
+
"step": 1100
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 0.7441860465116279,
|
| 443 |
+
"grad_norm": 0.0698658898472786,
|
| 444 |
+
"learning_rate": 7.648145119484152e-06,
|
| 445 |
+
"loss": 0.4871,
|
| 446 |
+
"step": 1120
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.7574750830564784,
|
| 450 |
+
"grad_norm": 0.06963298469781876,
|
| 451 |
+
"learning_rate": 6.912106378175098e-06,
|
| 452 |
+
"loss": 0.4884,
|
| 453 |
+
"step": 1140
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.770764119601329,
|
| 457 |
+
"grad_norm": 0.0692787617444992,
|
| 458 |
+
"learning_rate": 6.207589483478266e-06,
|
| 459 |
+
"loss": 0.4877,
|
| 460 |
+
"step": 1160
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 0.7840531561461794,
|
| 464 |
+
"grad_norm": 0.07016126066446304,
|
| 465 |
+
"learning_rate": 5.53582219988382e-06,
|
| 466 |
+
"loss": 0.4856,
|
| 467 |
+
"step": 1180
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 0.7973421926910299,
|
| 471 |
+
"grad_norm": 0.06945677101612091,
|
| 472 |
+
"learning_rate": 4.897975218999926e-06,
|
| 473 |
+
"loss": 0.4868,
|
| 474 |
+
"step": 1200
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.7973421926910299,
|
| 478 |
+
"eval_accuracy": 0.19665158843513314,
|
| 479 |
+
"eval_loss": 0.5205041170120239,
|
| 480 |
+
"eval_runtime": 14.8321,
|
| 481 |
+
"eval_samples_per_second": 123.651,
|
| 482 |
+
"eval_steps_per_second": 0.539,
|
| 483 |
+
"step": 1200
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.8106312292358804,
|
| 487 |
+
"grad_norm": 0.07045505195856094,
|
| 488 |
+
"learning_rate": 4.295160119383712e-06,
|
| 489 |
+
"loss": 0.4859,
|
| 490 |
+
"step": 1220
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.8239202657807309,
|
| 494 |
+
"grad_norm": 0.06839559227228165,
|
| 495 |
+
"learning_rate": 3.728427429388709e-06,
|
| 496 |
+
"loss": 0.4863,
|
| 497 |
+
"step": 1240
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 0.8372093023255814,
|
| 501 |
+
"grad_norm": 0.06684821844100952,
|
| 502 |
+
"learning_rate": 3.198764796404807e-06,
|
| 503 |
+
"loss": 0.4856,
|
| 504 |
+
"step": 1260
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 0.8504983388704319,
|
| 508 |
+
"grad_norm": 0.06731660664081573,
|
| 509 |
+
"learning_rate": 2.707095265681081e-06,
|
| 510 |
+
"loss": 0.4854,
|
| 511 |
+
"step": 1280
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 0.8637873754152824,
|
| 515 |
+
"grad_norm": 0.06780705600976944,
|
| 516 |
+
"learning_rate": 2.254275671731007e-06,
|
| 517 |
+
"loss": 0.4868,
|
| 518 |
+
"step": 1300
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.8770764119601329,
|
| 522 |
+
"grad_norm": 0.06815515458583832,
|
| 523 |
+
"learning_rate": 1.8410951451234533e-06,
|
| 524 |
+
"loss": 0.4854,
|
| 525 |
+
"step": 1320
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.8903654485049833,
|
| 529 |
+
"grad_norm": 0.0670180469751358,
|
| 530 |
+
"learning_rate": 1.4682737372615967e-06,
|
| 531 |
+
"loss": 0.485,
|
| 532 |
+
"step": 1340
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 0.9036544850498339,
|
| 536 |
+
"grad_norm": 0.06649608910083771,
|
| 537 |
+
"learning_rate": 1.1364611655463736e-06,
|
| 538 |
+
"loss": 0.4867,
|
| 539 |
+
"step": 1360
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 0.9169435215946844,
|
| 543 |
+
"grad_norm": 0.0674930214881897,
|
| 544 |
+
"learning_rate": 8.462356811112987e-07,
|
| 545 |
+
"loss": 0.4865,
|
| 546 |
+
"step": 1380
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.9302325581395349,
|
| 550 |
+
"grad_norm": 0.06808231770992279,
|
| 551 |
+
"learning_rate": 5.981030611018234e-07,
|
| 552 |
+
"loss": 0.4864,
|
| 553 |
+
"step": 1400
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.9302325581395349,
|
| 557 |
+
"eval_accuracy": 0.19667556159797644,
|
| 558 |
+
"eval_loss": 0.519675612449646,
|
| 559 |
+
"eval_runtime": 14.9507,
|
| 560 |
+
"eval_samples_per_second": 122.67,
|
| 561 |
+
"eval_steps_per_second": 0.535,
|
| 562 |
+
"step": 1400
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.9435215946843853,
|
| 566 |
+
"grad_norm": 0.06696037203073502,
|
| 567 |
+
"learning_rate": 3.9249572725543196e-07,
|
| 568 |
+
"loss": 0.4852,
|
| 569 |
+
"step": 1420
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.9568106312292359,
|
| 573 |
+
"grad_norm": 0.06675516068935394,
|
| 574 |
+
"learning_rate": 2.297719923185032e-07,
|
| 575 |
+
"loss": 0.4875,
|
| 576 |
+
"step": 1440
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.9700996677740864,
|
| 580 |
+
"grad_norm": 0.06678403913974762,
|
| 581 |
+
"learning_rate": 1.1021543561322012e-07,
|
| 582 |
+
"loss": 0.4852,
|
| 583 |
+
"step": 1460
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.9833887043189369,
|
| 587 |
+
"grad_norm": 0.0660882443189621,
|
| 588 |
+
"learning_rate": 3.403440884269526e-08,
|
| 589 |
+
"loss": 0.4848,
|
| 590 |
+
"step": 1480
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.9966777408637874,
|
| 594 |
+
"grad_norm": 0.06698651611804962,
|
| 595 |
+
"learning_rate": 1.3616729956228425e-09,
|
| 596 |
+
"loss": 0.4847,
|
| 597 |
+
"step": 1500
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.0,
|
| 601 |
+
"step": 1505,
|
| 602 |
+
"total_flos": 2.786803439690685e+19,
|
| 603 |
+
"train_loss": 0.0,
|
| 604 |
+
"train_runtime": 4.5361,
|
| 605 |
+
"train_samples_per_second": 339673.082,
|
| 606 |
+
"train_steps_per_second": 331.781
|
| 607 |
+
}
|
| 608 |
+
],
|
| 609 |
+
"logging_steps": 20,
|
| 610 |
+
"max_steps": 1505,
|
| 611 |
+
"num_input_tokens_seen": 0,
|
| 612 |
+
"num_train_epochs": 1,
|
| 613 |
+
"save_steps": 200,
|
| 614 |
+
"stateful_callbacks": {
|
| 615 |
+
"TrainerControl": {
|
| 616 |
+
"args": {
|
| 617 |
+
"should_epoch_stop": false,
|
| 618 |
+
"should_evaluate": false,
|
| 619 |
+
"should_log": false,
|
| 620 |
+
"should_save": true,
|
| 621 |
+
"should_training_stop": true
|
| 622 |
+
},
|
| 623 |
+
"attributes": {}
|
| 624 |
+
}
|
| 625 |
+
},
|
| 626 |
+
"total_flos": 2.786803439690685e+19,
|
| 627 |
+
"train_batch_size": 32,
|
| 628 |
+
"trial_name": null,
|
| 629 |
+
"trial_params": null
|
| 630 |
+
}
|