| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 50, |
| "global_step": 114, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.13333333333333333, |
| "grad_norm": 21.264778795863695, |
| "learning_rate": 1.9999999999999998e-05, |
| "loss": 3.1597023010253906, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.26666666666666666, |
| "grad_norm": 11.239746271438309, |
| "learning_rate": 2.9942920471376185e-05, |
| "loss": 0.903958511352539, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 4.168820207602787, |
| "learning_rate": 2.959567305869736e-05, |
| "loss": 0.33176212310791015, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.5333333333333333, |
| "grad_norm": 4.124967682779128, |
| "learning_rate": 2.8940212863402334e-05, |
| "loss": 0.1942599296569824, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.6666666666666666, |
| "grad_norm": 3.3997801822687603, |
| "learning_rate": 2.7990381056766583e-05, |
| "loss": 0.14255876541137696, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.8, |
| "grad_norm": 2.6752472907916256, |
| "learning_rate": 2.6766234973793638e-05, |
| "loss": 0.1059659481048584, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.9333333333333333, |
| "grad_norm": 1.1339089252087269, |
| "learning_rate": 2.5293624568031008e-05, |
| "loss": 0.06367313861846924, |
| "step": 35 |
| }, |
| { |
| "epoch": 1.0533333333333332, |
| "grad_norm": 0.9338182919130207, |
| "learning_rate": 2.360364654526569e-05, |
| "loss": 0.04107189476490021, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.1866666666666668, |
| "grad_norm": 0.67532314749969, |
| "learning_rate": 2.1731987703006933e-05, |
| "loss": 0.025246787071228027, |
| "step": 45 |
| }, |
| { |
| "epoch": 1.32, |
| "grad_norm": 0.22416776213454237, |
| "learning_rate": 1.9718171342274207e-05, |
| "loss": 0.012763623893260957, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.32, |
| "eval_loss": 0.0002999651478603482, |
| "eval_runtime": 2.4095, |
| "eval_samples_per_second": 103.756, |
| "eval_steps_per_second": 6.64, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.4533333333333334, |
| "grad_norm": 0.06292803618252299, |
| "learning_rate": 1.760472266500396e-05, |
| "loss": 0.0002823648974299431, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.5866666666666667, |
| "grad_norm": 0.023065956502653622, |
| "learning_rate": 1.543627078114667e-05, |
| "loss": 0.00017986295279115438, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.72, |
| "grad_norm": 0.018974236015362817, |
| "learning_rate": 1.3258606288121545e-05, |
| "loss": 0.0001341827679425478, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.8533333333333335, |
| "grad_norm": 0.01676317351779676, |
| "learning_rate": 1.1117714323462188e-05, |
| "loss": 0.00011214115656912326, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.9866666666666668, |
| "grad_norm": 0.015787599523459657, |
| "learning_rate": 9.058803509412647e-06, |
| "loss": 9.917097631841897e-05, |
| "step": 75 |
| }, |
| { |
| "epoch": 2.1066666666666665, |
| "grad_norm": 0.015024074736092713, |
| "learning_rate": 7.1253512949815995e-06, |
| "loss": 9.251137380488217e-05, |
| "step": 80 |
| }, |
| { |
| "epoch": 2.24, |
| "grad_norm": 0.014486233516685222, |
| "learning_rate": 5.3581858547019095e-06, |
| "loss": 8.82596941664815e-05, |
| "step": 85 |
| }, |
| { |
| "epoch": 2.3733333333333335, |
| "grad_norm": 0.014267650932010779, |
| "learning_rate": 3.7946239313850626e-06, |
| "loss": 8.51782620884478e-05, |
| "step": 90 |
| }, |
| { |
| "epoch": 2.506666666666667, |
| "grad_norm": 0.04592011870627601, |
| "learning_rate": 2.467682828805956e-06, |
| "loss": 8.472277550026774e-05, |
| "step": 95 |
| }, |
| { |
| "epoch": 2.64, |
| "grad_norm": 0.013865131593844745, |
| "learning_rate": 1.4053831944502509e-06, |
| "loss": 8.298547472804785e-05, |
| "step": 100 |
| }, |
| { |
| "epoch": 2.64, |
| "eval_loss": 8.276944572571665e-05, |
| "eval_runtime": 2.4114, |
| "eval_samples_per_second": 103.676, |
| "eval_steps_per_second": 6.635, |
| "step": 100 |
| }, |
| { |
| "epoch": 2.7733333333333334, |
| "grad_norm": 0.013862171882778414, |
| "learning_rate": 6.301573152676665e-07, |
| "loss": 8.245615754276514e-05, |
| "step": 105 |
| }, |
| { |
| "epoch": 2.9066666666666667, |
| "grad_norm": 0.013879740136511708, |
| "learning_rate": 1.5837542128583294e-07, |
| "loss": 8.253837004303932e-05, |
| "step": 110 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 8.275313302874565e-05, |
| "eval_runtime": 2.2598, |
| "eval_samples_per_second": 110.629, |
| "eval_steps_per_second": 7.08, |
| "step": 114 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 114, |
| "total_flos": 496702901452800.0, |
| "train_loss": 0.2185278557189393, |
| "train_runtime": 417.6369, |
| "train_samples_per_second": 34.121, |
| "train_steps_per_second": 0.273 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 114, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 496702901452800.0, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
