| { | |
| "best_metric": 2.2219202518463135, | |
| "best_model_checkpoint": "output/shadowraze/checkpoint-81", | |
| "epoch": 9.0, | |
| "global_step": 81, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 5.668773501204858e-05, | |
| "loss": 2.801, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.403702974319458, | |
| "eval_runtime": 0.5725, | |
| "eval_samples_per_second": 22.709, | |
| "eval_steps_per_second": 3.494, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.137086214086682e-06, | |
| "loss": 2.5593, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00010290000000000001, | |
| "loss": 2.5034, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.3624541759490967, | |
| "eval_runtime": 0.5663, | |
| "eval_samples_per_second": 22.957, | |
| "eval_steps_per_second": 3.532, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0001211506487979619, | |
| "loss": 2.4263, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.6049351202038163e-05, | |
| "loss": 2.3294, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.2799673080444336, | |
| "eval_runtime": 0.5719, | |
| "eval_samples_per_second": 22.73, | |
| "eval_steps_per_second": 3.497, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.4300000000000014e-05, | |
| "loss": 2.3824, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 0.00013306291378591332, | |
| "loss": 2.2699, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.2671029567718506, | |
| "eval_runtime": 0.5765, | |
| "eval_samples_per_second": 22.55, | |
| "eval_steps_per_second": 3.469, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 8.051226498795145e-05, | |
| "loss": 2.1985, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 2.1072, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.232370615005493, | |
| "eval_runtime": 0.5566, | |
| "eval_samples_per_second": 23.356, | |
| "eval_steps_per_second": 3.593, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 8.051226498795124e-05, | |
| "loss": 2.1033, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.2706549167633057, | |
| "eval_runtime": 0.5754, | |
| "eval_samples_per_second": 22.595, | |
| "eval_steps_per_second": 3.476, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.00013306291378591335, | |
| "loss": 2.0514, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 3.429999999999998e-05, | |
| "loss": 2.0029, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.2322123050689697, | |
| "eval_runtime": 0.5836, | |
| "eval_samples_per_second": 22.275, | |
| "eval_steps_per_second": 3.427, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "learning_rate": 1.604935120203803e-05, | |
| "loss": 2.0429, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 0.00012115064879796188, | |
| "loss": 1.9633, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.246337652206421, | |
| "eval_runtime": 0.582, | |
| "eval_samples_per_second": 22.338, | |
| "eval_steps_per_second": 3.437, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 0.00010289999999999993, | |
| "loss": 1.9794, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 4.137086214086705e-06, | |
| "loss": 1.8826, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.2219202518463135, | |
| "eval_runtime": 0.5809, | |
| "eval_samples_per_second": 22.378, | |
| "eval_steps_per_second": 3.443, | |
| "step": 81 | |
| } | |
| ], | |
| "max_steps": 90, | |
| "num_train_epochs": 10, | |
| "total_flos": 82306990080000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |