| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.008131734092295182, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016263468184590365, | |
| "eval_loss": 0.9036540389060974, | |
| "eval_runtime": 85.5431, | |
| "eval_samples_per_second": 30.265, | |
| "eval_steps_per_second": 15.139, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008131734092295182, | |
| "grad_norm": 0.7166550159454346, | |
| "learning_rate": 5e-05, | |
| "loss": 0.858, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0016263468184590363, | |
| "grad_norm": 0.6799307465553284, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8541, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0016263468184590363, | |
| "eval_loss": 0.8901762366294861, | |
| "eval_runtime": 85.6285, | |
| "eval_samples_per_second": 30.235, | |
| "eval_steps_per_second": 15.123, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.002439520227688555, | |
| "grad_norm": 0.6755326986312866, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.867, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0032526936369180726, | |
| "grad_norm": 0.9362283945083618, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.749, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0032526936369180726, | |
| "eval_loss": 0.8581121563911438, | |
| "eval_runtime": 86.0267, | |
| "eval_samples_per_second": 30.095, | |
| "eval_steps_per_second": 15.053, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.004065867046147591, | |
| "grad_norm": 0.9754035472869873, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.8638, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00487904045537711, | |
| "grad_norm": 0.7521759271621704, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8099, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00487904045537711, | |
| "eval_loss": 0.8457998633384705, | |
| "eval_runtime": 85.7126, | |
| "eval_samples_per_second": 30.206, | |
| "eval_steps_per_second": 15.109, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.005692213864606627, | |
| "grad_norm": 0.8753585815429688, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.7782, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.006505387273836145, | |
| "grad_norm": 0.7789082527160645, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.8728, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.006505387273836145, | |
| "eval_loss": 0.8393402099609375, | |
| "eval_runtime": 85.7455, | |
| "eval_samples_per_second": 30.194, | |
| "eval_steps_per_second": 15.103, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.007318560683065664, | |
| "grad_norm": 0.7507452964782715, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.8237, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.008131734092295182, | |
| "grad_norm": 0.8261967897415161, | |
| "learning_rate": 0.0, | |
| "loss": 0.7639, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.008131734092295182, | |
| "eval_loss": 0.83784019947052, | |
| "eval_runtime": 85.6513, | |
| "eval_samples_per_second": 30.227, | |
| "eval_steps_per_second": 15.119, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1990514140250112.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |