| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.21234240212342403, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010617120106171201, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010617120106171201, | |
| "eval_loss": NaN, | |
| "eval_runtime": 162.145, | |
| "eval_samples_per_second": 0.981, | |
| "eval_steps_per_second": 0.493, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.021234240212342402, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03185136031851361, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9471, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.042468480424684804, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0355, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05308560053085601, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05308560053085601, | |
| "eval_loss": NaN, | |
| "eval_runtime": 6.4168, | |
| "eval_samples_per_second": 24.779, | |
| "eval_steps_per_second": 12.467, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06370272063702721, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0743198407431984, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.08493696084936961, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09555408095554081, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 404.891, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10617120106171202, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10617120106171202, | |
| "eval_loss": NaN, | |
| "eval_runtime": 6.4168, | |
| "eval_samples_per_second": 24.779, | |
| "eval_steps_per_second": 12.467, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11678832116788321, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.12740544127405443, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.13802256138022562, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.389, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1486396814863968, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 289.0406, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.15925680159256803, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.15925680159256803, | |
| "eval_loss": NaN, | |
| "eval_runtime": 6.4173, | |
| "eval_samples_per_second": 24.777, | |
| "eval_steps_per_second": 12.466, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.16987392169873922, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.1804910418049104, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.19110816191108163, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.20172528201725282, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.21234240212342403, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21234240212342403, | |
| "eval_loss": NaN, | |
| "eval_runtime": 6.4176, | |
| "eval_samples_per_second": 24.776, | |
| "eval_steps_per_second": 12.466, | |
| "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": 2.418692624429875e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |