| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.028070175438596492, | |
| "eval_steps": 3, | |
| "global_step": 6, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004678362573099415, | |
| "grad_norm": 1.4389572143554688, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1689, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004678362573099415, | |
| "eval_loss": 1.1875958442687988, | |
| "eval_runtime": 2.4228, | |
| "eval_samples_per_second": 37.147, | |
| "eval_steps_per_second": 18.574, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00935672514619883, | |
| "grad_norm": 0.7820485234260559, | |
| "learning_rate": 4e-05, | |
| "loss": 0.8861, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.014035087719298246, | |
| "grad_norm": 1.0180163383483887, | |
| "learning_rate": 6e-05, | |
| "loss": 0.8148, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.014035087719298246, | |
| "eval_loss": 1.1761608123779297, | |
| "eval_runtime": 2.4101, | |
| "eval_samples_per_second": 37.342, | |
| "eval_steps_per_second": 18.671, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01871345029239766, | |
| "grad_norm": 1.5755374431610107, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0969, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.023391812865497075, | |
| "grad_norm": 1.0206433534622192, | |
| "learning_rate": 0.0001, | |
| "loss": 0.9503, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.028070175438596492, | |
| "grad_norm": 1.029083251953125, | |
| "learning_rate": 0.00012, | |
| "loss": 0.8455, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.028070175438596492, | |
| "eval_loss": 1.0448113679885864, | |
| "eval_runtime": 2.4564, | |
| "eval_samples_per_second": 36.639, | |
| "eval_steps_per_second": 18.319, | |
| "step": 6 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 312560831692800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |