| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.84, | |
| "eval_steps": 500, | |
| "global_step": 36, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1556, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0001995184726672197, | |
| "loss": 1.8767, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.4225959777832031, | |
| "eval_runtime": 8.0339, | |
| "eval_samples_per_second": 12.447, | |
| "eval_steps_per_second": 1.618, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 0.00018314696123025454, | |
| "loss": 0.685, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0001471396736825998, | |
| "loss": 0.3349, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_loss": 0.30958235263824463, | |
| "eval_runtime": 4.9729, | |
| "eval_samples_per_second": 20.109, | |
| "eval_steps_per_second": 2.614, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2955, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 5.286032631740023e-05, | |
| "loss": 0.2678, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_loss": 0.2913205325603485, | |
| "eval_runtime": 8.0251, | |
| "eval_samples_per_second": 12.461, | |
| "eval_steps_per_second": 1.62, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 0.2584, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 4.815273327803182e-07, | |
| "loss": 0.2538, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "eval_loss": 0.2900236248970032, | |
| "eval_runtime": 5.7265, | |
| "eval_samples_per_second": 17.463, | |
| "eval_steps_per_second": 2.27, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "step": 36, | |
| "total_flos": 55410626723840.0, | |
| "train_loss": 0.5664321829875311, | |
| "train_runtime": 739.8338, | |
| "train_samples_per_second": 1.622, | |
| "train_steps_per_second": 0.049 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 36, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 55410626723840.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |