| { | |
| "best_metric": 1.3609592914581299, | |
| "best_model_checkpoint": "./output/20_epochs_sep/checkpoint-780", | |
| "epoch": 20.0, | |
| "global_step": 7800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 5.0549, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.402883529663086, | |
| "eval_runtime": 121.8218, | |
| "eval_samples_per_second": 8.57, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.787234042553192e-05, | |
| "loss": 1.2729, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.3609592914581299, | |
| "eval_runtime": 121.6763, | |
| "eval_samples_per_second": 8.58, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.5212765957446815e-05, | |
| "loss": 0.8899, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.4757730960845947, | |
| "eval_runtime": 121.8679, | |
| "eval_samples_per_second": 8.567, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.2553191489361704e-05, | |
| "loss": 0.4533, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.7082480192184448, | |
| "eval_runtime": 121.9437, | |
| "eval_samples_per_second": 8.561, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.9893617021276594e-05, | |
| "loss": 0.182, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.9131618738174438, | |
| "eval_runtime": 121.9168, | |
| "eval_samples_per_second": 8.563, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.723404255319149e-05, | |
| "loss": 0.0838, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.0302553176879883, | |
| "eval_runtime": 121.9432, | |
| "eval_samples_per_second": 8.561, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.4574468085106386e-05, | |
| "loss": 0.054, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.1023058891296387, | |
| "eval_runtime": 121.8944, | |
| "eval_samples_per_second": 8.565, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.191489361702128e-05, | |
| "loss": 0.0396, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.1586802005767822, | |
| "eval_runtime": 121.8358, | |
| "eval_samples_per_second": 8.569, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.925531914893617e-05, | |
| "loss": 0.0317, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.2018632888793945, | |
| "eval_runtime": 121.8177, | |
| "eval_samples_per_second": 8.57, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.6595744680851064e-05, | |
| "loss": 0.0264, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.2312045097351074, | |
| "eval_runtime": 121.887, | |
| "eval_samples_per_second": 8.565, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.393617021276596e-05, | |
| "loss": 0.0225, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 2.2828288078308105, | |
| "eval_runtime": 121.91, | |
| "eval_samples_per_second": 8.564, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2.1276595744680852e-05, | |
| "loss": 0.0209, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 2.3354954719543457, | |
| "eval_runtime": 121.9104, | |
| "eval_samples_per_second": 8.564, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.8617021276595745e-05, | |
| "loss": 0.0203, | |
| "step": 5070 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 2.3210549354553223, | |
| "eval_runtime": 121.9222, | |
| "eval_samples_per_second": 8.563, | |
| "step": 5070 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 1.595744680851064e-05, | |
| "loss": 0.0195, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 2.3695647716522217, | |
| "eval_runtime": 121.8891, | |
| "eval_samples_per_second": 8.565, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.3297872340425532e-05, | |
| "loss": 0.0189, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 2.402658224105835, | |
| "eval_runtime": 121.8955, | |
| "eval_samples_per_second": 8.565, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 1.0638297872340426e-05, | |
| "loss": 0.0184, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 2.4124085903167725, | |
| "eval_runtime": 121.8966, | |
| "eval_samples_per_second": 8.565, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 7.97872340425532e-06, | |
| "loss": 0.0181, | |
| "step": 6630 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 2.4452691078186035, | |
| "eval_runtime": 121.8236, | |
| "eval_samples_per_second": 8.57, | |
| "step": 6630 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 5.319148936170213e-06, | |
| "loss": 0.0176, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 2.449826717376709, | |
| "eval_runtime": 121.8997, | |
| "eval_samples_per_second": 8.564, | |
| "step": 7020 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 2.6595744680851065e-06, | |
| "loss": 0.0172, | |
| "step": 7410 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 2.4811935424804688, | |
| "eval_runtime": 121.9376, | |
| "eval_samples_per_second": 8.562, | |
| "step": 7410 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0167, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 2.5037059783935547, | |
| "eval_runtime": 121.9361, | |
| "eval_samples_per_second": 8.562, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 7800, | |
| "total_flos": 0, | |
| "train_runtime": 22284.2589, | |
| "train_samples_per_second": 0.35 | |
| } | |
| ], | |
| "max_steps": 7800, | |
| "num_train_epochs": 20, | |
| "total_flos": 0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |