{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5747126436781609,
  "eval_steps": 9,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.022988505747126436,
      "grad_norm": 8.76852035522461,
      "learning_rate": 1e-05,
      "loss": 20.8384,
      "step": 1
    },
    {
      "epoch": 0.022988505747126436,
      "eval_loss": 21.21566390991211,
      "eval_runtime": 2.179,
      "eval_samples_per_second": 67.004,
      "eval_steps_per_second": 2.295,
      "step": 1
    },
    {
      "epoch": 0.04597701149425287,
      "grad_norm": 7.297959327697754,
      "learning_rate": 2e-05,
      "loss": 20.9799,
      "step": 2
    },
    {
      "epoch": 0.06896551724137931,
      "grad_norm": 8.222094535827637,
      "learning_rate": 3e-05,
      "loss": 20.9579,
      "step": 3
    },
    {
      "epoch": 0.09195402298850575,
      "grad_norm": 8.73771858215332,
      "learning_rate": 4e-05,
      "loss": 21.4096,
      "step": 4
    },
    {
      "epoch": 0.11494252873563218,
      "grad_norm": 9.340126037597656,
      "learning_rate": 5e-05,
      "loss": 20.8177,
      "step": 5
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 10.907496452331543,
      "learning_rate": 6e-05,
      "loss": 20.0583,
      "step": 6
    },
    {
      "epoch": 0.16091954022988506,
      "grad_norm": 13.2294340133667,
      "learning_rate": 7e-05,
      "loss": 19.5326,
      "step": 7
    },
    {
      "epoch": 0.1839080459770115,
      "grad_norm": 16.79814910888672,
      "learning_rate": 8e-05,
      "loss": 17.6944,
      "step": 8
    },
    {
      "epoch": 0.20689655172413793,
      "grad_norm": 20.354450225830078,
      "learning_rate": 9e-05,
      "loss": 15.8224,
      "step": 9
    },
    {
      "epoch": 0.20689655172413793,
      "eval_loss": 13.134273529052734,
      "eval_runtime": 2.1879,
      "eval_samples_per_second": 66.732,
      "eval_steps_per_second": 2.285,
      "step": 9
    },
    {
      "epoch": 0.22988505747126436,
      "grad_norm": 28.363195419311523,
      "learning_rate": 0.0001,
      "loss": 13.1636,
      "step": 10
    },
    {
      "epoch": 0.25287356321839083,
      "grad_norm": 20.680286407470703,
      "learning_rate": 9.99695413509548e-05,
      "loss": 10.6378,
      "step": 11
    },
    {
      "epoch": 0.27586206896551724,
      "grad_norm": 18.535959243774414,
      "learning_rate": 9.987820251299122e-05,
      "loss": 9.0366,
      "step": 12
    },
    {
      "epoch": 0.2988505747126437,
      "grad_norm": 13.819395065307617,
      "learning_rate": 9.972609476841367e-05,
      "loss": 6.7707,
      "step": 13
    },
    {
      "epoch": 0.3218390804597701,
      "grad_norm": 10.581539154052734,
      "learning_rate": 9.951340343707852e-05,
      "loss": 6.3245,
      "step": 14
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 9.617717742919922,
      "learning_rate": 9.924038765061042e-05,
      "loss": 5.1628,
      "step": 15
    },
    {
      "epoch": 0.367816091954023,
      "grad_norm": 6.453233242034912,
      "learning_rate": 9.890738003669029e-05,
      "loss": 4.81,
      "step": 16
    },
    {
      "epoch": 0.39080459770114945,
      "grad_norm": 4.609827041625977,
      "learning_rate": 9.851478631379982e-05,
      "loss": 4.5561,
      "step": 17
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 4.55856466293335,
      "learning_rate": 9.806308479691595e-05,
      "loss": 4.1678,
      "step": 18
    },
    {
      "epoch": 0.41379310344827586,
      "eval_loss": 4.0816650390625,
      "eval_runtime": 2.1607,
      "eval_samples_per_second": 67.571,
      "eval_steps_per_second": 2.314,
      "step": 18
    },
    {
      "epoch": 0.4367816091954023,
      "grad_norm": 5.335641384124756,
      "learning_rate": 9.755282581475769e-05,
      "loss": 4.1428,
      "step": 19
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 5.87034273147583,
      "learning_rate": 9.698463103929542e-05,
      "loss": 4.1433,
      "step": 20
    },
    {
      "epoch": 0.4827586206896552,
      "grad_norm": 10.226850509643555,
      "learning_rate": 9.635919272833938e-05,
      "loss": 3.8563,
      "step": 21
    },
    {
      "epoch": 0.5057471264367817,
      "grad_norm": 5.1733317375183105,
      "learning_rate": 9.567727288213005e-05,
      "loss": 3.7241,
      "step": 22
    },
    {
      "epoch": 0.5287356321839081,
      "grad_norm": 7.562571048736572,
      "learning_rate": 9.493970231495835e-05,
      "loss": 3.6946,
      "step": 23
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 8.14027214050293,
      "learning_rate": 9.414737964294636e-05,
      "loss": 3.7754,
      "step": 24
    },
    {
      "epoch": 0.5747126436781609,
      "grad_norm": 10.570993423461914,
      "learning_rate": 9.330127018922194e-05,
      "loss": 3.5848,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.41887283560448e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}