| { | |
| "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": 2641.419677734375, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 207.5718, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.176056338028169, | |
| "grad_norm": 3060.73486328125, | |
| "learning_rate": 4.62962962962963e-06, | |
| "loss": 126.3771, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2640845070422535, | |
| "grad_norm": 808.4242553710938, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 125.2416, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.352112676056338, | |
| "grad_norm": 1393.006591796875, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 104.9298, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.44014084507042256, | |
| "grad_norm": 1380.7303466796875, | |
| "learning_rate": 3.240740740740741e-06, | |
| "loss": 109.4596, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.528169014084507, | |
| "grad_norm": 2457.56494140625, | |
| "learning_rate": 2.7777777777777783e-06, | |
| "loss": 110.524, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.6161971830985915, | |
| "grad_norm": 2740.54052734375, | |
| "learning_rate": 2.314814814814815e-06, | |
| "loss": 114.8047, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.704225352112676, | |
| "grad_norm": 1358.532470703125, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 105.7686, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7922535211267606, | |
| "grad_norm": 2155.906982421875, | |
| "learning_rate": 1.3888888888888892e-06, | |
| "loss": 93.1012, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.8802816901408451, | |
| "grad_norm": 2438.5126953125, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 108.8455, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9683098591549296, | |
| "grad_norm": 1066.7130126953125, | |
| "learning_rate": 4.6296296296296297e-07, | |
| "loss": 99.7556, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 283.56231689453125, | |
| "eval_runtime": 22.2792, | |
| "eval_samples_per_second": 45.334, | |
| "eval_steps_per_second": 5.7, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 1.056338028169014, | |
| "grad_norm": 2160.288818359375, | |
| "learning_rate": 0.0, | |
| "loss": 106.0304, | |
| "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 | |
| } | |