| { | |
| "best_metric": 3.5847387313842773, | |
| "best_model_checkpoint": "./models/22_12_13_luther_blocks_larger_fp16_20ep/checkpoint-400", | |
| "epoch": 19.984126984126984, | |
| "global_step": 620, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.21560707501060514, | |
| "eval_loss": 4.621785640716553, | |
| "eval_runtime": 1.1004, | |
| "eval_samples_per_second": 48.163, | |
| "eval_steps_per_second": 6.361, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 4.2338709677419356e-05, | |
| "loss": 8.1175, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_accuracy": 0.263284088603626, | |
| "eval_loss": 4.0403876304626465, | |
| "eval_runtime": 1.1008, | |
| "eval_samples_per_second": 48.145, | |
| "eval_steps_per_second": 6.359, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "eval_accuracy": 0.28707648610265774, | |
| "eval_loss": 3.8119544982910156, | |
| "eval_runtime": 1.1015, | |
| "eval_samples_per_second": 48.117, | |
| "eval_steps_per_second": 6.355, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 3.427419354838709e-05, | |
| "loss": 3.734, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "eval_accuracy": 0.299655102454859, | |
| "eval_loss": 3.7062137126922607, | |
| "eval_runtime": 1.1015, | |
| "eval_samples_per_second": 48.116, | |
| "eval_steps_per_second": 6.355, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_accuracy": 0.3081576569099393, | |
| "eval_loss": 3.638169765472412, | |
| "eval_runtime": 1.1008, | |
| "eval_samples_per_second": 48.148, | |
| "eval_steps_per_second": 6.359, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 2.620967741935484e-05, | |
| "loss": 3.3639, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "eval_accuracy": 0.3127685866578137, | |
| "eval_loss": 3.610761880874634, | |
| "eval_runtime": 1.1014, | |
| "eval_samples_per_second": 48.121, | |
| "eval_steps_per_second": 6.356, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "eval_accuracy": 0.31479739574687843, | |
| "eval_loss": 3.6012349128723145, | |
| "eval_runtime": 1.1022, | |
| "eval_samples_per_second": 48.084, | |
| "eval_steps_per_second": 6.351, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "learning_rate": 1.8145161290322583e-05, | |
| "loss": 3.1363, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "eval_accuracy": 0.31680776111695164, | |
| "eval_loss": 3.5847387313842773, | |
| "eval_runtime": 1.1008, | |
| "eval_samples_per_second": 48.147, | |
| "eval_steps_per_second": 6.359, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 14.51, | |
| "eval_accuracy": 0.31802504657039044, | |
| "eval_loss": 3.5913662910461426, | |
| "eval_runtime": 1.0976, | |
| "eval_samples_per_second": 48.286, | |
| "eval_steps_per_second": 6.377, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "learning_rate": 1.0080645161290323e-05, | |
| "loss": 2.9884, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 16.13, | |
| "eval_accuracy": 0.317711503347535, | |
| "eval_loss": 3.5954136848449707, | |
| "eval_runtime": 1.0979, | |
| "eval_samples_per_second": 48.272, | |
| "eval_steps_per_second": 6.376, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 17.73, | |
| "eval_accuracy": 0.317563953595603, | |
| "eval_loss": 3.600076675415039, | |
| "eval_runtime": 1.1024, | |
| "eval_samples_per_second": 48.075, | |
| "eval_steps_per_second": 6.35, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 2.0161290322580646e-06, | |
| "loss": 2.8748, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_accuracy": 0.31879968276803333, | |
| "eval_loss": 3.604796886444092, | |
| "eval_runtime": 1.1023, | |
| "eval_samples_per_second": 48.08, | |
| "eval_steps_per_second": 6.35, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.98, | |
| "step": 620, | |
| "total_flos": 5223750303744000.0, | |
| "train_loss": 3.996271416448778, | |
| "train_runtime": 495.8533, | |
| "train_samples_per_second": 20.167, | |
| "train_steps_per_second": 1.25 | |
| } | |
| ], | |
| "max_steps": 620, | |
| "num_train_epochs": 20, | |
| "total_flos": 5223750303744000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |