| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.976, | |
| "eval_steps": 500, | |
| "global_step": 93, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.7531044483184814, | |
| "learning_rate": 0.00019857697953148037, | |
| "loss": 2.1283, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 2.2025675773620605, | |
| "learning_rate": 0.00019434841787099803, | |
| "loss": 1.8, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.9267807602882385, | |
| "learning_rate": 0.00018743466161445823, | |
| "loss": 1.5595, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.6875874996185303, | |
| "learning_rate": 0.0001780324790952092, | |
| "loss": 1.6097, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.7420657873153687, | |
| "learning_rate": 0.00016640946027672392, | |
| "loss": 1.2942, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.6181728839874268, | |
| "learning_rate": 0.00015289640103269625, | |
| "loss": 1.462, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.8008699417114258, | |
| "learning_rate": 0.0001378778885610576, | |
| "loss": 1.4432, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.0314457416534424, | |
| "learning_rate": 0.00012178135587488515, | |
| "loss": 1.3561, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.47113487124443054, | |
| "learning_rate": 0.00010506491688387127, | |
| "loss": 1.3299, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.41678062081336975, | |
| "learning_rate": 8.820432828491542e-05, | |
| "loss": 1.2208, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.4888109862804413, | |
| "learning_rate": 7.16794493317696e-05, | |
| "loss": 1.4574, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.3696904480457306, | |
| "learning_rate": 5.596058484423656e-05, | |
| "loss": 1.1423, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.6482197046279907, | |
| "learning_rate": 4.149510014046922e-05, | |
| "loss": 1.3134, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.7164911031723022, | |
| "learning_rate": 2.869468883687798e-05, | |
| "loss": 1.3384, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.3950205147266388, | |
| "learning_rate": 1.7923655879272393e-05, | |
| "loss": 1.3873, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.387712687253952, | |
| "learning_rate": 9.488549274967872e-06, | |
| "loss": 1.2214, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.2879863679409027, | |
| "learning_rate": 3.6294356110059157e-06, | |
| "loss": 1.3198, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.9349786043167114, | |
| "learning_rate": 5.130676608104845e-07, | |
| "loss": 1.3832, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "step": 93, | |
| "total_flos": 1234235324694528.0, | |
| "train_loss": 1.4291518221619308, | |
| "train_runtime": 192.8949, | |
| "train_samples_per_second": 7.776, | |
| "train_steps_per_second": 0.482 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 93, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 1234235324694528.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |