| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.056338028169014, | |
| "eval_steps": 500, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0880281690140845, | |
| "grad_norm": 1525.1461181640625, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 134.6262, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.176056338028169, | |
| "grad_norm": 2871.249755859375, | |
| "learning_rate": 4.62962962962963e-06, | |
| "loss": 106.7172, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2640845070422535, | |
| "grad_norm": 492.1271667480469, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 102.1769, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.352112676056338, | |
| "grad_norm": 1749.6070556640625, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 106.4214, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.44014084507042256, | |
| "grad_norm": 853.5518798828125, | |
| "learning_rate": 3.240740740740741e-06, | |
| "loss": 92.0699, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.528169014084507, | |
| "grad_norm": 2185.8486328125, | |
| "learning_rate": 2.7777777777777783e-06, | |
| "loss": 90.0462, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6161971830985915, | |
| "grad_norm": 411.7020263671875, | |
| "learning_rate": 2.314814814814815e-06, | |
| "loss": 93.0463, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.704225352112676, | |
| "grad_norm": 764.4384155273438, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 86.6424, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7922535211267606, | |
| "grad_norm": 3891.708251953125, | |
| "learning_rate": 1.3888888888888892e-06, | |
| "loss": 78.9075, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.8802816901408451, | |
| "grad_norm": 1514.9061279296875, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 79.3982, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9683098591549296, | |
| "grad_norm": 1521.738037109375, | |
| "learning_rate": 4.6296296296296297e-07, | |
| "loss": 82.1088, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 197.31683349609375, | |
| "eval_runtime": 13.0547, | |
| "eval_samples_per_second": 77.367, | |
| "eval_steps_per_second": 9.728, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 1.056338028169014, | |
| "grad_norm": 260.0885925292969, | |
| "learning_rate": 0.0, | |
| "loss": 75.2076, | |
| "step": 1200 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 600, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |