| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 11.051574012056262, | |
| "global_step": 16500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4176, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3487, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.8214285714285716e-05, | |
| "loss": 0.3403, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.642857142857143e-05, | |
| "loss": 0.3348, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.464285714285715e-05, | |
| "loss": 0.3279, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 0.3278, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.107142857142857e-05, | |
| "loss": 0.3253, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.928571428571429e-05, | |
| "loss": 0.3211, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3209, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.3251, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 0.3245891034603119, | |
| "eval_runtime": 22.7214, | |
| "eval_samples_per_second": 144.269, | |
| "eval_steps_per_second": 36.089, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 3.392857142857143e-05, | |
| "loss": 0.3185, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.2142857142857144e-05, | |
| "loss": 0.3166, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.0357142857142857e-05, | |
| "loss": 0.3176, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.3121, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.6785714285714288e-05, | |
| "loss": 0.3167, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2982, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 2.3214285714285715e-05, | |
| "loss": 0.2894, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 0.289, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.9642857142857145e-05, | |
| "loss": 0.2882, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.286, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 0.3317989408969879, | |
| "eval_runtime": 22.4167, | |
| "eval_samples_per_second": 146.23, | |
| "eval_steps_per_second": 36.58, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.6071428571428572e-05, | |
| "loss": 0.2899, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.2901, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2877, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.0714285714285714e-05, | |
| "loss": 0.2877, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.2878, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 0.2883, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 5.357142857142857e-06, | |
| "loss": 0.2872, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 3.5714285714285714e-06, | |
| "loss": 0.2847, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.7857142857142857e-06, | |
| "loss": 0.2868, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.0, | |
| "loss": 0.284, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 0.317633718252182, | |
| "eval_runtime": 22.7238, | |
| "eval_samples_per_second": 144.254, | |
| "eval_steps_per_second": 36.085, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 1.1842105263157895e-05, | |
| "loss": 0.1046, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 10.72, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.0676, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 9.210526315789474e-06, | |
| "loss": 0.0606, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "eval_loss": 0.35377955436706543, | |
| "eval_runtime": 42.666, | |
| "eval_samples_per_second": 76.829, | |
| "eval_steps_per_second": 19.219, | |
| "step": 16500 | |
| } | |
| ], | |
| "max_steps": 20000, | |
| "num_train_epochs": 14, | |
| "total_flos": 5.680414299577478e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |