| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 36.0, | |
| "global_step": 14040, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5992906078421846, | |
| "eval_loss": 1.33469557762146, | |
| "eval_runtime": 5.8169, | |
| "eval_samples_per_second": 29.397, | |
| "eval_steps_per_second": 14.785, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 9.644586894586896e-06, | |
| "loss": 1.5408, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6796235478653008, | |
| "eval_loss": 1.0698518753051758, | |
| "eval_runtime": 5.8146, | |
| "eval_samples_per_second": 29.409, | |
| "eval_steps_per_second": 14.79, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 9.28917378917379e-06, | |
| "loss": 1.1283, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7372856151482106, | |
| "eval_loss": 0.8751375675201416, | |
| "eval_runtime": 5.8077, | |
| "eval_samples_per_second": 29.444, | |
| "eval_steps_per_second": 14.808, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 8.933048433048434e-06, | |
| "loss": 0.9078, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7711176956331666, | |
| "eval_loss": 0.7534294128417969, | |
| "eval_runtime": 5.8229, | |
| "eval_samples_per_second": 29.367, | |
| "eval_steps_per_second": 14.769, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.802166420482521, | |
| "eval_loss": 0.6710610389709473, | |
| "eval_runtime": 5.6976, | |
| "eval_samples_per_second": 30.013, | |
| "eval_steps_per_second": 15.094, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 8.576923076923077e-06, | |
| "loss": 0.7705, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8168786692759296, | |
| "eval_loss": 0.6077755689620972, | |
| "eval_runtime": 5.8114, | |
| "eval_samples_per_second": 29.425, | |
| "eval_steps_per_second": 14.798, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 8.220797720797722e-06, | |
| "loss": 0.6863, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.831802051431627, | |
| "eval_loss": 0.5667673945426941, | |
| "eval_runtime": 5.8128, | |
| "eval_samples_per_second": 29.418, | |
| "eval_steps_per_second": 14.795, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 7.864672364672366e-06, | |
| "loss": 0.6277, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8386034912718204, | |
| "eval_loss": 0.546131432056427, | |
| "eval_runtime": 5.8073, | |
| "eval_samples_per_second": 29.446, | |
| "eval_steps_per_second": 14.809, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 7.508547008547009e-06, | |
| "loss": 0.5863, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8513571943310113, | |
| "eval_loss": 0.5142761468887329, | |
| "eval_runtime": 5.8058, | |
| "eval_samples_per_second": 29.453, | |
| "eval_steps_per_second": 14.813, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8522174818930449, | |
| "eval_loss": 0.49920225143432617, | |
| "eval_runtime": 5.8205, | |
| "eval_samples_per_second": 29.379, | |
| "eval_steps_per_second": 14.775, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.26, | |
| "learning_rate": 7.152421652421653e-06, | |
| "loss": 0.5564, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8533209429717304, | |
| "eval_loss": 0.49400192499160767, | |
| "eval_runtime": 5.8142, | |
| "eval_samples_per_second": 29.411, | |
| "eval_steps_per_second": 14.791, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "learning_rate": 6.796296296296296e-06, | |
| "loss": 0.5199, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8632922665875019, | |
| "eval_loss": 0.4726846218109131, | |
| "eval_runtime": 5.8155, | |
| "eval_samples_per_second": 29.404, | |
| "eval_steps_per_second": 14.788, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "learning_rate": 6.440170940170941e-06, | |
| "loss": 0.5025, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8637540927527733, | |
| "eval_loss": 0.4585917592048645, | |
| "eval_runtime": 5.8153, | |
| "eval_samples_per_second": 29.405, | |
| "eval_steps_per_second": 14.789, | |
| "step": 5070 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8673048600883653, | |
| "eval_loss": 0.45485442876815796, | |
| "eval_runtime": 5.8178, | |
| "eval_samples_per_second": 29.393, | |
| "eval_steps_per_second": 14.782, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 6.084045584045585e-06, | |
| "loss": 0.4814, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8698035411108416, | |
| "eval_loss": 0.44424179196357727, | |
| "eval_runtime": 5.7973, | |
| "eval_samples_per_second": 29.496, | |
| "eval_steps_per_second": 14.834, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 5.727920227920228e-06, | |
| "loss": 0.4746, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8749573607523999, | |
| "eval_loss": 0.43056586384773254, | |
| "eval_runtime": 5.6925, | |
| "eval_samples_per_second": 30.04, | |
| "eval_steps_per_second": 15.108, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 5.371794871794872e-06, | |
| "loss": 0.4527, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.874227107665129, | |
| "eval_loss": 0.42905324697494507, | |
| "eval_runtime": 5.8566, | |
| "eval_samples_per_second": 29.198, | |
| "eval_steps_per_second": 14.684, | |
| "step": 6630 | |
| }, | |
| { | |
| "epoch": 17.95, | |
| "learning_rate": 5.016381766381767e-06, | |
| "loss": 0.4382, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8751032706419789, | |
| "eval_loss": 0.4213222861289978, | |
| "eval_runtime": 5.8093, | |
| "eval_samples_per_second": 29.435, | |
| "eval_steps_per_second": 14.804, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8751214299591995, | |
| "eval_loss": 0.41926833987236023, | |
| "eval_runtime": 5.8022, | |
| "eval_samples_per_second": 29.472, | |
| "eval_steps_per_second": 14.822, | |
| "step": 7410 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 4.6602564102564106e-06, | |
| "loss": 0.4328, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8759944995580002, | |
| "eval_loss": 0.41431769728660583, | |
| "eval_runtime": 5.8137, | |
| "eval_samples_per_second": 29.413, | |
| "eval_steps_per_second": 14.793, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 20.51, | |
| "learning_rate": 4.304131054131054e-06, | |
| "loss": 0.4191, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8835775987576434, | |
| "eval_loss": 0.4071265459060669, | |
| "eval_runtime": 5.8164, | |
| "eval_samples_per_second": 29.4, | |
| "eval_steps_per_second": 14.786, | |
| "step": 8190 | |
| }, | |
| { | |
| "epoch": 21.79, | |
| "learning_rate": 3.948717948717949e-06, | |
| "loss": 0.4106, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.881939052795031, | |
| "eval_loss": 0.3980366587638855, | |
| "eval_runtime": 5.875, | |
| "eval_samples_per_second": 29.106, | |
| "eval_steps_per_second": 14.638, | |
| "step": 8580 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8821786026625207, | |
| "eval_loss": 0.39872363209724426, | |
| "eval_runtime": 5.8105, | |
| "eval_samples_per_second": 29.43, | |
| "eval_steps_per_second": 14.801, | |
| "step": 8970 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 3.592592592592593e-06, | |
| "loss": 0.4037, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8819395993222366, | |
| "eval_loss": 0.40265128016471863, | |
| "eval_runtime": 5.8012, | |
| "eval_samples_per_second": 29.477, | |
| "eval_steps_per_second": 14.824, | |
| "step": 9360 | |
| }, | |
| { | |
| "epoch": 24.36, | |
| "learning_rate": 3.2364672364672365e-06, | |
| "loss": 0.3893, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8892501819946614, | |
| "eval_loss": 0.3867790400981903, | |
| "eval_runtime": 5.8209, | |
| "eval_samples_per_second": 29.377, | |
| "eval_steps_per_second": 14.774, | |
| "step": 9750 | |
| }, | |
| { | |
| "epoch": 25.64, | |
| "learning_rate": 2.8803418803418804e-06, | |
| "loss": 0.3991, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8846003326484688, | |
| "eval_loss": 0.3882477581501007, | |
| "eval_runtime": 5.8184, | |
| "eval_samples_per_second": 29.389, | |
| "eval_steps_per_second": 14.781, | |
| "step": 10140 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 2.5242165242165246e-06, | |
| "loss": 0.3786, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8858714334822964, | |
| "eval_loss": 0.3939129710197449, | |
| "eval_runtime": 5.8222, | |
| "eval_samples_per_second": 29.37, | |
| "eval_steps_per_second": 14.771, | |
| "step": 10530 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8847989764283254, | |
| "eval_loss": 0.39587706327438354, | |
| "eval_runtime": 5.8063, | |
| "eval_samples_per_second": 29.451, | |
| "eval_steps_per_second": 14.812, | |
| "step": 10920 | |
| }, | |
| { | |
| "epoch": 28.21, | |
| "learning_rate": 2.168803418803419e-06, | |
| "loss": 0.38, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8849687976020835, | |
| "eval_loss": 0.3949810862541199, | |
| "eval_runtime": 5.8138, | |
| "eval_samples_per_second": 29.413, | |
| "eval_steps_per_second": 14.792, | |
| "step": 11310 | |
| }, | |
| { | |
| "epoch": 29.49, | |
| "learning_rate": 1.8126780626780629e-06, | |
| "loss": 0.3764, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8893153879792042, | |
| "eval_loss": 0.3783101439476013, | |
| "eval_runtime": 5.8039, | |
| "eval_samples_per_second": 29.463, | |
| "eval_steps_per_second": 14.818, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 1.4565527065527065e-06, | |
| "loss": 0.3708, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8890792500856626, | |
| "eval_loss": 0.3798995912075043, | |
| "eval_runtime": 5.7021, | |
| "eval_samples_per_second": 29.989, | |
| "eval_steps_per_second": 15.082, | |
| "step": 12090 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8867172306495527, | |
| "eval_loss": 0.39150363206863403, | |
| "eval_runtime": 5.8791, | |
| "eval_samples_per_second": 29.086, | |
| "eval_steps_per_second": 14.628, | |
| "step": 12480 | |
| }, | |
| { | |
| "epoch": 32.05, | |
| "learning_rate": 1.1004273504273506e-06, | |
| "loss": 0.3656, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8902530694061639, | |
| "eval_loss": 0.3780055344104767, | |
| "eval_runtime": 5.8784, | |
| "eval_samples_per_second": 29.09, | |
| "eval_steps_per_second": 14.63, | |
| "step": 12870 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 7.443019943019944e-07, | |
| "loss": 0.3617, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8873723487824038, | |
| "eval_loss": 0.38049712777137756, | |
| "eval_runtime": 5.8779, | |
| "eval_samples_per_second": 29.092, | |
| "eval_steps_per_second": 14.631, | |
| "step": 13260 | |
| }, | |
| { | |
| "epoch": 34.62, | |
| "learning_rate": 3.8817663817663825e-07, | |
| "loss": 0.361, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8919778767559101, | |
| "eval_loss": 0.3775971233844757, | |
| "eval_runtime": 5.8063, | |
| "eval_samples_per_second": 29.451, | |
| "eval_steps_per_second": 14.812, | |
| "step": 13650 | |
| }, | |
| { | |
| "epoch": 35.9, | |
| "learning_rate": 3.205128205128205e-08, | |
| "loss": 0.3595, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8888084202394747, | |
| "eval_loss": 0.3711872100830078, | |
| "eval_runtime": 5.8621, | |
| "eval_samples_per_second": 29.171, | |
| "eval_steps_per_second": 14.671, | |
| "step": 14040 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "step": 14040, | |
| "total_flos": 3.509780816886497e+17, | |
| "train_loss": 0.5379996796958467, | |
| "train_runtime": 13714.1823, | |
| "train_samples_per_second": 8.19, | |
| "train_steps_per_second": 1.024 | |
| } | |
| ], | |
| "max_steps": 14040, | |
| "num_train_epochs": 36, | |
| "total_flos": 3.509780816886497e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |