| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6127450980392157, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.3999999999999997e-05, | |
| "loss": 0.8848, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4.7999999999999994e-05, | |
| "loss": 0.7411, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.9070383906364441, | |
| "eval_runtime": 1.3223, | |
| "eval_samples_per_second": 756.258, | |
| "eval_steps_per_second": 12.1, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 7.199999999999999e-05, | |
| "loss": 0.7398, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.599999999999999e-05, | |
| "loss": 0.7395, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.9064154028892517, | |
| "eval_runtime": 1.194, | |
| "eval_samples_per_second": 837.495, | |
| "eval_steps_per_second": 13.4, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 0.7392, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00014399999999999998, | |
| "loss": 0.7387, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 0.9047173857688904, | |
| "eval_runtime": 1.3077, | |
| "eval_samples_per_second": 764.675, | |
| "eval_steps_per_second": 12.235, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.000168, | |
| "loss": 0.7384, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.00019199999999999998, | |
| "loss": 0.7382, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.9014638662338257, | |
| "eval_runtime": 1.2645, | |
| "eval_samples_per_second": 790.849, | |
| "eval_steps_per_second": 12.654, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00021599999999999996, | |
| "loss": 0.7382, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.7381, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.9044438600540161, | |
| "eval_runtime": 1.2394, | |
| "eval_samples_per_second": 806.828, | |
| "eval_steps_per_second": 12.909, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00026399999999999997, | |
| "loss": 0.7379, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00028799999999999995, | |
| "loss": 0.7379, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.9041813015937805, | |
| "eval_runtime": 1.2368, | |
| "eval_samples_per_second": 808.566, | |
| "eval_steps_per_second": 12.937, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.000312, | |
| "loss": 0.7381, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.000336, | |
| "loss": 0.7379, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 0.9053561687469482, | |
| "eval_runtime": 1.2638, | |
| "eval_samples_per_second": 791.247, | |
| "eval_steps_per_second": 12.66, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00035999999999999997, | |
| "loss": 0.7378, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.00038399999999999996, | |
| "loss": 0.7378, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_loss": 0.9035025835037231, | |
| "eval_runtime": 1.2079, | |
| "eval_samples_per_second": 827.911, | |
| "eval_steps_per_second": 13.247, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.000408, | |
| "loss": 0.7378, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00043199999999999993, | |
| "loss": 0.7378, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_loss": 0.9025757312774658, | |
| "eval_runtime": 1.2745, | |
| "eval_samples_per_second": 784.598, | |
| "eval_steps_per_second": 12.554, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00045599999999999997, | |
| "loss": 0.7375, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00047999999999999996, | |
| "loss": 0.7371, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 0.903812050819397, | |
| "eval_runtime": 1.2376, | |
| "eval_samples_per_second": 807.983, | |
| "eval_steps_per_second": 12.928, | |
| "step": 10000 | |
| } | |
| ], | |
| "max_steps": 250000, | |
| "num_train_epochs": 16, | |
| "total_flos": 1.6016800286580408e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |