{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5303867403314917,
  "eval_steps": 9,
  "global_step": 45,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011786372007366482,
      "eval_loss": 3.2576773166656494,
      "eval_runtime": 48.3038,
      "eval_samples_per_second": 5.921,
      "eval_steps_per_second": 2.96,
      "step": 1
    },
    {
      "epoch": 0.03535911602209945,
      "grad_norm": 37.830448150634766,
      "learning_rate": 2.2499999999999998e-05,
      "loss": 99.4918,
      "step": 3
    },
    {
      "epoch": 0.0707182320441989,
      "grad_norm": 30.299089431762695,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 95.1039,
      "step": 6
    },
    {
      "epoch": 0.10607734806629834,
      "grad_norm": 31.264686584472656,
      "learning_rate": 6.75e-05,
      "loss": 99.6355,
      "step": 9
    },
    {
      "epoch": 0.10607734806629834,
      "eval_loss": 2.994905948638916,
      "eval_runtime": 48.3217,
      "eval_samples_per_second": 5.919,
      "eval_steps_per_second": 2.959,
      "step": 9
    },
    {
      "epoch": 0.1414364640883978,
      "grad_norm": 34.47152328491211,
      "learning_rate": 8.999999999999999e-05,
      "loss": 91.6026,
      "step": 12
    },
    {
      "epoch": 0.17679558011049723,
      "grad_norm": 26.11569595336914,
      "learning_rate": 0.0001125,
      "loss": 85.1959,
      "step": 15
    },
    {
      "epoch": 0.2121546961325967,
      "grad_norm": 38.721710205078125,
      "learning_rate": 0.000135,
      "loss": 78.4229,
      "step": 18
    },
    {
      "epoch": 0.2121546961325967,
      "eval_loss": 2.432400703430176,
      "eval_runtime": 48.3101,
      "eval_samples_per_second": 5.92,
      "eval_steps_per_second": 2.96,
      "step": 18
    },
    {
      "epoch": 0.24751381215469614,
      "grad_norm": 37.68834686279297,
      "learning_rate": 0.00014994217771805422,
      "loss": 78.5547,
      "step": 21
    },
    {
      "epoch": 0.2828729281767956,
      "grad_norm": 36.297088623046875,
      "learning_rate": 0.00014907662554463532,
      "loss": 71.8931,
      "step": 24
    },
    {
      "epoch": 0.318232044198895,
      "grad_norm": 40.15138244628906,
      "learning_rate": 0.0001471841427340235,
      "loss": 68.7287,
      "step": 27
    },
    {
      "epoch": 0.318232044198895,
      "eval_loss": 2.1229209899902344,
      "eval_runtime": 48.3187,
      "eval_samples_per_second": 5.919,
      "eval_steps_per_second": 2.96,
      "step": 27
    },
    {
      "epoch": 0.35359116022099446,
      "grad_norm": 22.00638198852539,
      "learning_rate": 0.0001442909649383465,
      "loss": 65.7473,
      "step": 30
    },
    {
      "epoch": 0.3889502762430939,
      "grad_norm": 26.196809768676758,
      "learning_rate": 0.0001404372005304598,
      "loss": 64.8032,
      "step": 33
    },
    {
      "epoch": 0.4243093922651934,
      "grad_norm": 30.012163162231445,
      "learning_rate": 0.00013567627457812106,
      "loss": 62.5845,
      "step": 36
    },
    {
      "epoch": 0.4243093922651934,
      "eval_loss": 1.9335837364196777,
      "eval_runtime": 48.3129,
      "eval_samples_per_second": 5.92,
      "eval_steps_per_second": 2.96,
      "step": 36
    },
    {
      "epoch": 0.45966850828729283,
      "grad_norm": 24.18491554260254,
      "learning_rate": 0.0001300741882076764,
      "loss": 56.7812,
      "step": 39
    },
    {
      "epoch": 0.4950276243093923,
      "grad_norm": 29.25063133239746,
      "learning_rate": 0.00012370860362476374,
      "loss": 59.2692,
      "step": 42
    },
    {
      "epoch": 0.5303867403314917,
      "grad_norm": 26.837539672851562,
      "learning_rate": 0.00011666776747647015,
      "loss": 57.7333,
      "step": 45
    },
    {
      "epoch": 0.5303867403314917,
      "eval_loss": 1.8067072629928589,
      "eval_runtime": 48.2991,
      "eval_samples_per_second": 5.921,
      "eval_steps_per_second": 2.961,
      "step": 45
    }
  ],
  "logging_steps": 3,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 9,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.3666203419475968e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}