| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 4.444444444444445, |
| "eval_steps": 500, |
| "global_step": 1000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2222222222222222, |
| "grad_norm": 0.44146928191185, |
| "learning_rate": 2.45e-05, |
| "loss": 0.3662, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.4444444444444444, |
| "grad_norm": 0.31828737258911133, |
| "learning_rate": 4.9500000000000004e-05, |
| "loss": 0.261, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 0.19953744113445282, |
| "learning_rate": 4.727777777777778e-05, |
| "loss": 0.1323, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.8888888888888888, |
| "grad_norm": 0.16124850511550903, |
| "learning_rate": 4.4500000000000004e-05, |
| "loss": 0.0823, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.17199474573135376, |
| "learning_rate": 4.172222222222222e-05, |
| "loss": 0.0626, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.3333333333333333, |
| "grad_norm": 0.17989133298397064, |
| "learning_rate": 3.894444444444444e-05, |
| "loss": 0.0558, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.5555555555555556, |
| "grad_norm": 0.14189372956752777, |
| "learning_rate": 3.6166666666666674e-05, |
| "loss": 0.0475, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.7777777777777777, |
| "grad_norm": 0.1783151626586914, |
| "learning_rate": 3.338888888888889e-05, |
| "loss": 0.0398, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.16954894363880157, |
| "learning_rate": 3.061111111111111e-05, |
| "loss": 0.0378, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.2222222222222223, |
| "grad_norm": 0.13010431826114655, |
| "learning_rate": 2.7833333333333333e-05, |
| "loss": 0.037, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.4444444444444446, |
| "grad_norm": 0.1510184407234192, |
| "learning_rate": 2.5055555555555555e-05, |
| "loss": 0.0333, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.6666666666666665, |
| "grad_norm": 0.14560167491436005, |
| "learning_rate": 2.2277777777777778e-05, |
| "loss": 0.032, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.888888888888889, |
| "grad_norm": 0.12952196598052979, |
| "learning_rate": 1.9500000000000003e-05, |
| "loss": 0.0311, |
| "step": 650 |
| }, |
| { |
| "epoch": 3.111111111111111, |
| "grad_norm": 0.1540798395872116, |
| "learning_rate": 1.6722222222222222e-05, |
| "loss": 0.0296, |
| "step": 700 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.09700635820627213, |
| "learning_rate": 1.3944444444444446e-05, |
| "loss": 0.0305, |
| "step": 750 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "grad_norm": 0.3841066360473633, |
| "learning_rate": 1.1166666666666668e-05, |
| "loss": 0.0283, |
| "step": 800 |
| }, |
| { |
| "epoch": 3.7777777777777777, |
| "grad_norm": 0.10915391892194748, |
| "learning_rate": 8.38888888888889e-06, |
| "loss": 0.0267, |
| "step": 850 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 0.11399481445550919, |
| "learning_rate": 5.611111111111112e-06, |
| "loss": 0.0282, |
| "step": 900 |
| }, |
| { |
| "epoch": 4.222222222222222, |
| "grad_norm": 0.11289925873279572, |
| "learning_rate": 2.8333333333333335e-06, |
| "loss": 0.0274, |
| "step": 950 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "grad_norm": 0.12091820687055588, |
| "learning_rate": 5.555555555555556e-08, |
| "loss": 0.0267, |
| "step": 1000 |
| }, |
| { |
| "epoch": 4.444444444444445, |
| "step": 1000, |
| "total_flos": 1.099960418304e+16, |
| "train_loss": 0.07079731237888336, |
| "train_runtime": 1216.7022, |
| "train_samples_per_second": 13.15, |
| "train_steps_per_second": 0.822 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 1000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.099960418304e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
