| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.0, | |
| "eval_steps": 500, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 166.0, | |
| "learning_rate": 0.0001, | |
| "loss": 33.9156, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 16.560213088989258, | |
| "eval_runtime": 0.2548, | |
| "eval_samples_per_second": 39.24, | |
| "eval_steps_per_second": 3.924, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 22.625, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 27.435, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 13.68699836730957, | |
| "eval_runtime": 0.2314, | |
| "eval_samples_per_second": 43.206, | |
| "eval_steps_per_second": 4.321, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 12.075172424316406, | |
| "eval_runtime": 0.2412, | |
| "eval_samples_per_second": 41.466, | |
| "eval_steps_per_second": 4.147, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 10.3125, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 17.6281, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 10.12792682647705, | |
| "eval_runtime": 0.2331, | |
| "eval_samples_per_second": 42.906, | |
| "eval_steps_per_second": 4.291, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 9.297090530395508, | |
| "eval_runtime": 0.2318, | |
| "eval_samples_per_second": 43.135, | |
| "eval_steps_per_second": 4.314, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 15.107, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 8.732256889343262, | |
| "eval_runtime": 0.2336, | |
| "eval_samples_per_second": 42.814, | |
| "eval_steps_per_second": 4.281, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_loss": 8.606382369995117, | |
| "eval_runtime": 0.2466, | |
| "eval_samples_per_second": 40.544, | |
| "eval_steps_per_second": 4.054, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.984375, | |
| "learning_rate": 0.0, | |
| "loss": 14.5856, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 8.568929672241211, | |
| "eval_runtime": 0.2329, | |
| "eval_samples_per_second": 42.94, | |
| "eval_steps_per_second": 4.294, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 20, | |
| "total_flos": 6.098355666236211e+16, | |
| "train_loss": 19.012953186035155, | |
| "train_runtime": 50.5903, | |
| "train_samples_per_second": 30.045, | |
| "train_steps_per_second": 0.395 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 6.098355666236211e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |