{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 11.0,
  "eval_steps": 500,
  "global_step": 3498,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9968553459119497,
      "grad_norm": 0.5557537078857422,
      "learning_rate": 0.00032468465370880226,
      "loss": 0.2366,
      "step": 317
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8729032258064516,
      "eval_loss": 0.06576934456825256,
      "eval_runtime": 2.6516,
      "eval_samples_per_second": 1169.097,
      "eval_steps_per_second": 24.513,
      "step": 318
    },
    {
      "epoch": 1.9937106918238994,
      "grad_norm": 0.34931233525276184,
      "learning_rate": 0.0002923284653322885,
      "loss": 0.0546,
      "step": 634
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9090322580645162,
      "eval_loss": 0.04315788671374321,
      "eval_runtime": 2.6079,
      "eval_samples_per_second": 1188.685,
      "eval_steps_per_second": 24.924,
      "step": 636
    },
    {
      "epoch": 2.990566037735849,
      "grad_norm": 0.35186338424682617,
      "learning_rate": 0.0002599722769557747,
      "loss": 0.034,
      "step": 951
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.922258064516129,
      "eval_loss": 0.038206350058317184,
      "eval_runtime": 2.6441,
      "eval_samples_per_second": 1172.423,
      "eval_steps_per_second": 24.583,
      "step": 954
    },
    {
      "epoch": 3.9874213836477987,
      "grad_norm": 0.393400639295578,
      "learning_rate": 0.00022761608857926093,
      "loss": 0.0277,
      "step": 1268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9296774193548387,
      "eval_loss": 0.03332997485995293,
      "eval_runtime": 2.6838,
      "eval_samples_per_second": 1155.08,
      "eval_steps_per_second": 24.219,
      "step": 1272
    },
    {
      "epoch": 4.984276729559748,
      "grad_norm": 0.08043188601732254,
      "learning_rate": 0.00019525990020274712,
      "loss": 0.0223,
      "step": 1585
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9387096774193548,
      "eval_loss": 0.028788963332772255,
      "eval_runtime": 2.6515,
      "eval_samples_per_second": 1169.143,
      "eval_steps_per_second": 24.514,
      "step": 1590
    },
    {
      "epoch": 5.981132075471698,
      "grad_norm": 0.08290695399045944,
      "learning_rate": 0.00016290371182623337,
      "loss": 0.0192,
      "step": 1902
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9393548387096774,
      "eval_loss": 0.02643631212413311,
      "eval_runtime": 2.63,
      "eval_samples_per_second": 1178.689,
      "eval_steps_per_second": 24.714,
      "step": 1908
    },
    {
      "epoch": 6.977987421383648,
      "grad_norm": 0.08547654002904892,
      "learning_rate": 0.0001305475234497196,
      "loss": 0.0172,
      "step": 2219
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9458064516129032,
      "eval_loss": 0.023828405886888504,
      "eval_runtime": 2.6537,
      "eval_samples_per_second": 1168.168,
      "eval_steps_per_second": 24.494,
      "step": 2226
    },
    {
      "epoch": 7.9748427672955975,
      "grad_norm": 0.0587199442088604,
      "learning_rate": 9.819133507320583e-05,
      "loss": 0.0159,
      "step": 2536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9448387096774193,
      "eval_loss": 0.023019853979349136,
      "eval_runtime": 2.6517,
      "eval_samples_per_second": 1169.073,
      "eval_steps_per_second": 24.513,
      "step": 2544
    },
    {
      "epoch": 8.971698113207546,
      "grad_norm": 0.06312248110771179,
      "learning_rate": 6.583514669669206e-05,
      "loss": 0.0149,
      "step": 2853
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9438709677419355,
      "eval_loss": 0.02213130332529545,
      "eval_runtime": 2.6757,
      "eval_samples_per_second": 1158.577,
      "eval_steps_per_second": 24.293,
      "step": 2862
    },
    {
      "epoch": 9.968553459119496,
      "grad_norm": 0.051413584500551224,
      "learning_rate": 3.3478958320178286e-05,
      "loss": 0.014,
      "step": 3170
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9438709677419355,
      "eval_loss": 0.021344251930713654,
      "eval_runtime": 2.6694,
      "eval_samples_per_second": 1161.296,
      "eval_steps_per_second": 24.35,
      "step": 3180
    },
    {
      "epoch": 10.965408805031446,
      "grad_norm": 0.0515441969037056,
      "learning_rate": 1.1227699436645158e-06,
      "loss": 0.0133,
      "step": 3487
    }
  ],
  "logging_steps": 317,
  "max_steps": 3498,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 11,
  "save_steps": 1000000000.0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 907333363443108.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9860087246286913,
    "fp16": true,
    "learning_rate": 0.0003356061431608116,
    "lr_scheduler": "cosine",
    "num_train_epochs": 11,
    "temperature": 20,
    "warmup_steps": 210,
    "weight_decay": 0.1094679341539599
  }
}