| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.788732394366197, | |
| "eval_steps": 54, | |
| "global_step": 594, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.6778770685195923, | |
| "learning_rate": 1.267605633802817e-05, | |
| "loss": 2.0809, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 2.0533597469329834, | |
| "eval_runtime": 65.0241, | |
| "eval_samples_per_second": 3.46, | |
| "eval_steps_per_second": 0.877, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 2.9120090007781982, | |
| "learning_rate": 2.535211267605634e-05, | |
| "loss": 1.8984, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.723413109779358, | |
| "eval_runtime": 65.0617, | |
| "eval_samples_per_second": 3.458, | |
| "eval_steps_per_second": 0.876, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 2.7429661750793457, | |
| "learning_rate": 3.802816901408451e-05, | |
| "loss": 1.6513, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 1.6305742263793945, | |
| "eval_runtime": 65.0654, | |
| "eval_samples_per_second": 3.458, | |
| "eval_steps_per_second": 0.876, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 2.668692111968994, | |
| "learning_rate": 4.992175273865415e-05, | |
| "loss": 1.6082, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 1.5966345071792603, | |
| "eval_runtime": 65.0716, | |
| "eval_samples_per_second": 3.458, | |
| "eval_steps_per_second": 0.876, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 2.2179903984069824, | |
| "learning_rate": 4.85133020344288e-05, | |
| "loss": 1.5274, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_loss": 1.5803054571151733, | |
| "eval_runtime": 65.0649, | |
| "eval_samples_per_second": 3.458, | |
| "eval_steps_per_second": 0.876, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 2.964496374130249, | |
| "learning_rate": 4.710485133020345e-05, | |
| "loss": 1.5474, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 1.5681735277175903, | |
| "eval_runtime": 65.0649, | |
| "eval_samples_per_second": 3.458, | |
| "eval_steps_per_second": 0.876, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 2.6218090057373047, | |
| "learning_rate": 4.569640062597809e-05, | |
| "loss": 1.55, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_loss": 1.5585384368896484, | |
| "eval_runtime": 65.0599, | |
| "eval_samples_per_second": 3.458, | |
| "eval_steps_per_second": 0.876, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 2.9666621685028076, | |
| "learning_rate": 4.428794992175274e-05, | |
| "loss": 1.5002, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_loss": 1.5494067668914795, | |
| "eval_runtime": 65.0722, | |
| "eval_samples_per_second": 3.458, | |
| "eval_steps_per_second": 0.876, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 2.445936918258667, | |
| "learning_rate": 4.287949921752739e-05, | |
| "loss": 1.4529, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_loss": 1.553564429283142, | |
| "eval_runtime": 65.0624, | |
| "eval_samples_per_second": 3.458, | |
| "eval_steps_per_second": 0.876, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 3.49705171585083, | |
| "learning_rate": 4.1471048513302035e-05, | |
| "loss": 1.4639, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "eval_loss": 1.5489095449447632, | |
| "eval_runtime": 65.0549, | |
| "eval_samples_per_second": 3.459, | |
| "eval_steps_per_second": 0.876, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 2.301144599914551, | |
| "learning_rate": 4.0062597809076686e-05, | |
| "loss": 1.463, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_loss": 1.5437979698181152, | |
| "eval_runtime": 65.0526, | |
| "eval_samples_per_second": 3.459, | |
| "eval_steps_per_second": 0.876, | |
| "step": 594 | |
| } | |
| ], | |
| "logging_steps": 54, | |
| "max_steps": 2130, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 54, | |
| "total_flos": 5.290146072576e+16, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |