| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03865201111245319, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0019326005556226597, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 33.8977, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0019326005556226597, | |
| "eval_loss": NaN, | |
| "eval_runtime": 122.0931, | |
| "eval_samples_per_second": 7.142, | |
| "eval_steps_per_second": 3.571, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0038652011112453193, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005797801666867979, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 8.6521, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.007730402222490639, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 295.1216, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.009663002778113298, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.009663002778113298, | |
| "eval_loss": NaN, | |
| "eval_runtime": 36.0981, | |
| "eval_samples_per_second": 24.156, | |
| "eval_steps_per_second": 12.078, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.011595603333735958, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 150.8499, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.013528203889358619, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 7.0981, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.015460804444981277, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.017393405000603938, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0338, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.019326005556226596, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.019326005556226596, | |
| "eval_loss": NaN, | |
| "eval_runtime": 35.922, | |
| "eval_samples_per_second": 24.275, | |
| "eval_steps_per_second": 12.137, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02125860611184926, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.023191206667471917, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 10.0715, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.025123807223094576, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 4.8381, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.027056407778717238, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.3283, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.028989008334339896, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.028989008334339896, | |
| "eval_loss": NaN, | |
| "eval_runtime": 35.9118, | |
| "eval_samples_per_second": 24.282, | |
| "eval_steps_per_second": 12.141, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.030921608889962555, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03285420944558522, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 179.9758, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.034786810001207875, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.036719410556830534, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03865201111245319, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 178.6408, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03865201111245319, | |
| "eval_loss": NaN, | |
| "eval_runtime": 35.8846, | |
| "eval_samples_per_second": 24.3, | |
| "eval_steps_per_second": 12.15, | |
| "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": 2.4668841405579264e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |