| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0014389041306141243, | |
| "eval_steps": 25, | |
| "global_step": 150, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 6.853318691253662, | |
| "learning_rate": 2.3797595190380762e-05, | |
| "loss": 4.0257, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "eval_loss": 3.5309877395629883, | |
| "eval_runtime": 335.3635, | |
| "eval_samples_per_second": 69.077, | |
| "eval_steps_per_second": 8.635, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 5.111692905426025, | |
| "learning_rate": 2.2545090180360722e-05, | |
| "loss": 3.3251, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "eval_loss": 3.3726470470428467, | |
| "eval_runtime": 336.3014, | |
| "eval_samples_per_second": 68.885, | |
| "eval_steps_per_second": 8.611, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 5.772947788238525, | |
| "learning_rate": 2.1292585170340683e-05, | |
| "loss": 3.2615, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "eval_loss": 3.2642982006073, | |
| "eval_runtime": 362.3653, | |
| "eval_samples_per_second": 63.93, | |
| "eval_steps_per_second": 7.992, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 5.64016056060791, | |
| "learning_rate": 2.0040080160320643e-05, | |
| "loss": 3.4283, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "eval_loss": 3.2154319286346436, | |
| "eval_runtime": 342.7157, | |
| "eval_samples_per_second": 67.595, | |
| "eval_steps_per_second": 8.45, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 7.610867977142334, | |
| "learning_rate": 1.87875751503006e-05, | |
| "loss": 3.1832, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "eval_loss": 3.1817972660064697, | |
| "eval_runtime": 339.5184, | |
| "eval_samples_per_second": 68.232, | |
| "eval_steps_per_second": 8.53, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 5.710243225097656, | |
| "learning_rate": 1.7535070140280564e-05, | |
| "loss": 3.1834, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0, | |
| "eval_loss": 3.162160873413086, | |
| "eval_runtime": 340.5303, | |
| "eval_samples_per_second": 68.029, | |
| "eval_steps_per_second": 8.504, | |
| "step": 150 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "total_flos": 182642872320000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |