| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.004554655870445344, |
| "eval_steps": 9, |
| "global_step": 81, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 5.623031938821413e-05, |
| "eval_loss": 1.2098013162612915, |
| "eval_runtime": 289.309, |
| "eval_samples_per_second": 25.882, |
| "eval_steps_per_second": 12.941, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.0005060728744939271, |
| "eval_loss": 0.9007603526115417, |
| "eval_runtime": 289.7759, |
| "eval_samples_per_second": 25.841, |
| "eval_steps_per_second": 12.92, |
| "step": 9 |
| }, |
| { |
| "epoch": 0.0005623031938821412, |
| "grad_norm": 5.226573944091797, |
| "learning_rate": 0.0002, |
| "loss": 4.2859, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.0010121457489878543, |
| "eval_loss": 0.8225377798080444, |
| "eval_runtime": 428.5277, |
| "eval_samples_per_second": 17.474, |
| "eval_steps_per_second": 8.737, |
| "step": 18 |
| }, |
| { |
| "epoch": 0.0011246063877642825, |
| "grad_norm": 5.386185646057129, |
| "learning_rate": 0.00019396926207859084, |
| "loss": 3.3328, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.0015182186234817814, |
| "eval_loss": 0.8000039458274841, |
| "eval_runtime": 664.0082, |
| "eval_samples_per_second": 11.277, |
| "eval_steps_per_second": 5.638, |
| "step": 27 |
| }, |
| { |
| "epoch": 0.0016869095816464238, |
| "grad_norm": 4.547457218170166, |
| "learning_rate": 0.0001766044443118978, |
| "loss": 3.1939, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.0020242914979757085, |
| "eval_loss": 0.7851625084877014, |
| "eval_runtime": 660.146, |
| "eval_samples_per_second": 11.343, |
| "eval_steps_per_second": 5.671, |
| "step": 36 |
| }, |
| { |
| "epoch": 0.002249212775528565, |
| "grad_norm": 5.285128593444824, |
| "learning_rate": 0.00015000000000000001, |
| "loss": 3.1844, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.0025303643724696357, |
| "eval_loss": 0.7743021845817566, |
| "eval_runtime": 575.5468, |
| "eval_samples_per_second": 13.01, |
| "eval_steps_per_second": 6.505, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.0028115159694107063, |
| "grad_norm": 4.259403705596924, |
| "learning_rate": 0.00011736481776669306, |
| "loss": 3.1686, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.003036437246963563, |
| "eval_loss": 0.7657517790794373, |
| "eval_runtime": 287.3701, |
| "eval_samples_per_second": 26.057, |
| "eval_steps_per_second": 13.028, |
| "step": 54 |
| }, |
| { |
| "epoch": 0.0033738191632928477, |
| "grad_norm": 4.84288215637207, |
| "learning_rate": 8.263518223330697e-05, |
| "loss": 3.0761, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.00354251012145749, |
| "eval_loss": 0.7605909109115601, |
| "eval_runtime": 289.3206, |
| "eval_samples_per_second": 25.881, |
| "eval_steps_per_second": 12.941, |
| "step": 63 |
| }, |
| { |
| "epoch": 0.003936122357174989, |
| "grad_norm": 3.9699506759643555, |
| "learning_rate": 5.000000000000002e-05, |
| "loss": 2.9889, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.004048582995951417, |
| "eval_loss": 0.7541989088058472, |
| "eval_runtime": 288.1317, |
| "eval_samples_per_second": 25.988, |
| "eval_steps_per_second": 12.994, |
| "step": 72 |
| }, |
| { |
| "epoch": 0.00449842555105713, |
| "grad_norm": 3.68632173538208, |
| "learning_rate": 2.339555568810221e-05, |
| "loss": 2.8912, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.004554655870445344, |
| "eval_loss": 0.7495855093002319, |
| "eval_runtime": 288.8772, |
| "eval_samples_per_second": 25.921, |
| "eval_steps_per_second": 12.961, |
| "step": 81 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 100, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 9, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.5468848970989568e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
