| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6298815822625347, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.000975, | |
| "loss": 0.9774, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.8144938945770264, | |
| "eval_runtime": 14.6573, | |
| "eval_samples_per_second": 68.225, | |
| "eval_steps_per_second": 34.113, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 0.00095, | |
| "loss": 0.9617, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 0.8077166080474854, | |
| "eval_runtime": 15.1765, | |
| "eval_samples_per_second": 65.891, | |
| "eval_steps_per_second": 32.946, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000925, | |
| "loss": 0.911, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 0.8064053058624268, | |
| "eval_runtime": 15.3268, | |
| "eval_samples_per_second": 65.245, | |
| "eval_steps_per_second": 32.623, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.8954, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.8163686990737915, | |
| "eval_runtime": 14.5944, | |
| "eval_samples_per_second": 68.519, | |
| "eval_steps_per_second": 34.26, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.000875, | |
| "loss": 0.883, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 0.8068735003471375, | |
| "eval_runtime": 14.8614, | |
| "eval_samples_per_second": 67.288, | |
| "eval_steps_per_second": 33.644, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00085, | |
| "loss": 0.8867, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 0.7819482088088989, | |
| "eval_runtime": 14.6896, | |
| "eval_samples_per_second": 68.076, | |
| "eval_steps_per_second": 34.038, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.000825, | |
| "loss": 0.8688, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 0.8062307238578796, | |
| "eval_runtime": 14.6856, | |
| "eval_samples_per_second": 68.094, | |
| "eval_steps_per_second": 34.047, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0008, | |
| "loss": 0.8446, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.7707250714302063, | |
| "eval_runtime": 15.0517, | |
| "eval_samples_per_second": 66.438, | |
| "eval_steps_per_second": 33.219, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 0.0007750000000000001, | |
| "loss": 0.8617, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 0.7528353333473206, | |
| "eval_runtime": 14.7368, | |
| "eval_samples_per_second": 67.857, | |
| "eval_steps_per_second": 33.929, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00075, | |
| "loss": 0.8158, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 0.7551385760307312, | |
| "eval_runtime": 15.3736, | |
| "eval_samples_per_second": 65.047, | |
| "eval_steps_per_second": 32.523, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20000, | |
| "num_train_epochs": 3, | |
| "save_steps": 5000, | |
| "total_flos": 3.1313397362098176e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |