| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.013043655484449516, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006521827742224758, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 163.8659, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006521827742224758, | |
| "eval_loss": NaN, | |
| "eval_runtime": 186.4675, | |
| "eval_samples_per_second": 13.852, | |
| "eval_steps_per_second": 6.929, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013043655484449516, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0019565483226674277, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 1926.5157, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002608731096889903, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 187.0539, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003260913871112379, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.003260913871112379, | |
| "eval_loss": NaN, | |
| "eval_runtime": 84.7214, | |
| "eval_samples_per_second": 30.488, | |
| "eval_steps_per_second": 15.25, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0039130966453348555, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004565279419557331, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.005217462193779806, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005869644968002283, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006521827742224758, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 1536.9054, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006521827742224758, | |
| "eval_loss": NaN, | |
| "eval_runtime": 128.7296, | |
| "eval_samples_per_second": 20.065, | |
| "eval_steps_per_second": 10.037, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0071740105164472346, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 10.8312, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.007826193290669711, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008478376064892186, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.3673, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.009130558839114662, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.009782741613337137, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 237.8173, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.009782741613337137, | |
| "eval_loss": NaN, | |
| "eval_runtime": 92.3039, | |
| "eval_samples_per_second": 27.984, | |
| "eval_steps_per_second": 13.997, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.010434924387559613, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 40.8634, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01108710716178209, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.011739289936004565, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012391472710227041, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.013043655484449516, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013043655484449516, | |
| "eval_loss": NaN, | |
| "eval_runtime": 74.3386, | |
| "eval_samples_per_second": 34.746, | |
| "eval_steps_per_second": 17.38, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5553097342451712.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |