| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.571428571428571, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 187.0, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 49.7965, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "eval_loss": 15.486898422241211, | |
| "eval_runtime": 0.2527, | |
| "eval_samples_per_second": 39.572, | |
| "eval_steps_per_second": 3.957, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 52.5, | |
| "learning_rate": 0.00019730448705798239, | |
| "loss": 39.0892, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 12.533833503723145, | |
| "eval_runtime": 0.2329, | |
| "eval_samples_per_second": 42.936, | |
| "eval_steps_per_second": 4.294, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 11.8125, | |
| "learning_rate": 0.0001686241637868734, | |
| "loss": 23.1465, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "eval_loss": 10.118247985839844, | |
| "eval_runtime": 0.2382, | |
| "eval_samples_per_second": 41.973, | |
| "eval_steps_per_second": 4.197, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 8.45867919921875, | |
| "eval_runtime": 0.2351, | |
| "eval_samples_per_second": 42.528, | |
| "eval_steps_per_second": 4.253, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 19.7449, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "eval_loss": 8.061217308044434, | |
| "eval_runtime": 0.2464, | |
| "eval_samples_per_second": 40.588, | |
| "eval_steps_per_second": 4.059, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 4.21875, | |
| "learning_rate": 6.039202339608432e-05, | |
| "loss": 18.8693, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 7.827044486999512, | |
| "eval_runtime": 0.2351, | |
| "eval_samples_per_second": 42.54, | |
| "eval_steps_per_second": 4.254, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "eval_loss": 7.732919216156006, | |
| "eval_runtime": 0.245, | |
| "eval_samples_per_second": 40.809, | |
| "eval_steps_per_second": 4.081, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 7.142857142857143, | |
| "grad_norm": 4.65625, | |
| "learning_rate": 1.6451218858706374e-05, | |
| "loss": 18.4335, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 7.698564052581787, | |
| "eval_runtime": 0.2338, | |
| "eval_samples_per_second": 42.772, | |
| "eval_steps_per_second": 4.277, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 4.6875, | |
| "learning_rate": 0.0, | |
| "loss": 18.2424, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "eval_loss": 7.704453945159912, | |
| "eval_runtime": 0.2338, | |
| "eval_samples_per_second": 42.771, | |
| "eval_steps_per_second": 4.277, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "step": 30, | |
| "total_flos": 9.147533553041408e+16, | |
| "train_loss": 23.27788518269857, | |
| "train_runtime": 75.0535, | |
| "train_samples_per_second": 29.046, | |
| "train_steps_per_second": 0.4 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 9.147533553041408e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |