{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.126984126984127,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12698412698412698,
      "grad_norm": 0.014252515509724617,
      "learning_rate": 4.95e-06,
      "loss": 0.007,
      "on_policy_loss": 0.007,
      "step": 2
    },
    {
      "epoch": 0.25396825396825395,
      "grad_norm": 0.010893095284700394,
      "learning_rate": 4.85e-06,
      "loss": 0.0067,
      "on_policy_loss": 0.0067,
      "step": 4
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 0.011728067882359028,
      "learning_rate": 4.75e-06,
      "loss": 0.0057,
      "on_policy_loss": 0.0057,
      "step": 6
    },
    {
      "epoch": 0.5079365079365079,
      "grad_norm": 0.010753393173217773,
      "learning_rate": 4.65e-06,
      "loss": 0.0062,
      "on_policy_loss": 0.0062,
      "step": 8
    },
    {
      "epoch": 0.6349206349206349,
      "grad_norm": 0.015550932846963406,
      "learning_rate": 4.5500000000000005e-06,
      "loss": 0.0059,
      "on_policy_loss": 0.0059,
      "step": 10
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.010618478059768677,
      "learning_rate": 4.450000000000001e-06,
      "loss": 0.0052,
      "on_policy_loss": 0.0052,
      "step": 12
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.02142387442290783,
      "learning_rate": 4.350000000000001e-06,
      "loss": 0.0053,
      "on_policy_loss": 0.0053,
      "step": 14
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.009728207252919674,
      "learning_rate": 4.25e-06,
      "loss": 0.0057,
      "on_policy_loss": 0.0065,
      "step": 16
    },
    {
      "epoch": 1.126984126984127,
      "grad_norm": 0.013969828374683857,
      "learning_rate": 4.15e-06,
      "loss": 0.006,
      "on_policy_loss": 0.006,
      "step": 18
    },
    {
      "epoch": 1.253968253968254,
      "grad_norm": 0.01231201458722353,
      "learning_rate": 4.05e-06,
      "loss": 0.0043,
      "on_policy_loss": 0.0043,
      "step": 20
    },
    {
      "epoch": 1.380952380952381,
      "grad_norm": 0.01638919860124588,
      "learning_rate": 3.95e-06,
      "loss": 0.0042,
      "on_policy_loss": 0.0042,
      "step": 22
    },
    {
      "epoch": 1.507936507936508,
      "grad_norm": 0.01664210483431816,
      "learning_rate": 3.85e-06,
      "loss": 0.0049,
      "on_policy_loss": 0.0049,
      "step": 24
    },
    {
      "epoch": 1.6349206349206349,
      "grad_norm": 0.0705290138721466,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.0046,
      "on_policy_loss": 0.0046,
      "step": 26
    },
    {
      "epoch": 1.7619047619047619,
      "grad_norm": 0.01711524836719036,
      "learning_rate": 3.65e-06,
      "loss": 0.0031,
      "on_policy_loss": 0.0031,
      "step": 28
    },
    {
      "epoch": 1.8888888888888888,
      "grad_norm": 0.022614775225520134,
      "learning_rate": 3.5500000000000003e-06,
      "loss": -0.0001,
      "on_policy_loss": -0.0001,
      "step": 30
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.03971567749977112,
      "learning_rate": 3.45e-06,
      "loss": -0.0024,
      "on_policy_loss": -0.0027,
      "step": 32
    },
    {
      "epoch": 2.126984126984127,
      "grad_norm": 0.023077696561813354,
      "learning_rate": 3.3500000000000005e-06,
      "loss": -0.0033,
      "on_policy_loss": -0.0033,
      "step": 34
    },
    {
      "epoch": 2.253968253968254,
      "grad_norm": 0.02600308135151863,
      "learning_rate": 3.2500000000000002e-06,
      "loss": -0.0081,
      "on_policy_loss": -0.0081,
      "step": 36
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.031187646090984344,
      "learning_rate": 3.1500000000000003e-06,
      "loss": -0.01,
      "on_policy_loss": -0.01,
      "step": 38
    },
    {
      "epoch": 2.507936507936508,
      "grad_norm": 0.0329587459564209,
      "learning_rate": 3.05e-06,
      "loss": -0.0136,
      "on_policy_loss": -0.0136,
      "step": 40
    },
    {
      "epoch": 2.634920634920635,
      "grad_norm": 0.020644811913371086,
      "learning_rate": 2.95e-06,
      "loss": -0.0127,
      "on_policy_loss": -0.0127,
      "step": 42
    },
    {
      "epoch": 2.761904761904762,
      "grad_norm": 0.02621665596961975,
      "learning_rate": 2.85e-06,
      "loss": -0.0134,
      "on_policy_loss": -0.0134,
      "step": 44
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 0.029684213921427727,
      "learning_rate": 2.7500000000000004e-06,
      "loss": -0.0111,
      "on_policy_loss": -0.0111,
      "step": 46
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.028028976172208786,
      "learning_rate": 2.6500000000000005e-06,
      "loss": -0.0115,
      "on_policy_loss": -0.0132,
      "step": 48
    },
    {
      "epoch": 3.126984126984127,
      "grad_norm": 0.01847624033689499,
      "learning_rate": 2.55e-06,
      "loss": -0.0109,
      "on_policy_loss": -0.0109,
      "step": 50
    }
  ],
  "logging_steps": 2,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}