{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04782608695652174,
  "eval_steps": 27,
  "global_step": 22,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002173913043478261,
      "grad_norm": 0.31070002913475037,
      "learning_rate": 0.0,
      "loss": 1.0674,
      "step": 1
    },
    {
      "epoch": 0.002173913043478261,
      "eval_loss": 4.08291482925415,
      "eval_runtime": 14.2183,
      "eval_samples_per_second": 27.289,
      "eval_steps_per_second": 1.758,
      "step": 1
    },
    {
      "epoch": 0.004347826086956522,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 2.6769,
      "step": 2
    },
    {
      "epoch": 0.006521739130434782,
      "grad_norm": 0.6840159893035889,
      "learning_rate": 2e-05,
      "loss": 3.095,
      "step": 3
    },
    {
      "epoch": 0.008695652173913044,
      "grad_norm": 0.9547252655029297,
      "learning_rate": 4e-05,
      "loss": 3.3761,
      "step": 4
    },
    {
      "epoch": 0.010869565217391304,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 6.3563,
      "step": 5
    },
    {
      "epoch": 0.013043478260869565,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 6.813,
      "step": 6
    },
    {
      "epoch": 0.015217391304347827,
      "grad_norm": 1.6616406440734863,
      "learning_rate": 6e-05,
      "loss": 2.632,
      "step": 7
    },
    {
      "epoch": 0.017391304347826087,
      "grad_norm": 0.5449294447898865,
      "learning_rate": 8e-05,
      "loss": 1.8506,
      "step": 8
    },
    {
      "epoch": 0.01956521739130435,
      "grad_norm": 1.0437792539596558,
      "learning_rate": 0.0001,
      "loss": 2.8176,
      "step": 9
    },
    {
      "epoch": 0.021739130434782608,
      "grad_norm": 1.3472729921340942,
      "learning_rate": 0.00012,
      "loss": 3.2039,
      "step": 10
    },
    {
      "epoch": 0.02391304347826087,
      "grad_norm": 3.4967548847198486,
      "learning_rate": 0.00014,
      "loss": 4.6888,
      "step": 11
    },
    {
      "epoch": 0.02608695652173913,
      "grad_norm": 8.567302703857422,
      "learning_rate": 0.00016,
      "loss": 6.068,
      "step": 12
    },
    {
      "epoch": 0.02826086956521739,
      "grad_norm": 6.242945194244385,
      "learning_rate": 0.00018,
      "loss": 3.8248,
      "step": 13
    },
    {
      "epoch": 0.030434782608695653,
      "grad_norm": 1.132996678352356,
      "learning_rate": 0.0002,
      "loss": 1.7484,
      "step": 14
    },
    {
      "epoch": 0.03260869565217391,
      "grad_norm": 0.8223450779914856,
      "learning_rate": 0.0001999486216200688,
      "loss": 3.1641,
      "step": 15
    },
    {
      "epoch": 0.034782608695652174,
      "grad_norm": 2.4242587089538574,
      "learning_rate": 0.00019979453927503364,
      "loss": 3.1671,
      "step": 16
    },
    {
      "epoch": 0.03695652173913044,
      "grad_norm": 2.9778759479522705,
      "learning_rate": 0.00019953791129491983,
      "loss": 4.0796,
      "step": 17
    },
    {
      "epoch": 0.0391304347826087,
      "grad_norm": 5.1453986167907715,
      "learning_rate": 0.0001991790013823246,
      "loss": 4.1276,
      "step": 18
    },
    {
      "epoch": 0.041304347826086954,
      "grad_norm": 5.806064605712891,
      "learning_rate": 0.00019871817834144504,
      "loss": 4.0408,
      "step": 19
    },
    {
      "epoch": 0.043478260869565216,
      "grad_norm": 2.1800103187561035,
      "learning_rate": 0.00019815591569910654,
      "loss": 1.1865,
      "step": 20
    },
    {
      "epoch": 0.04565217391304348,
      "grad_norm": 5.802156925201416,
      "learning_rate": 0.00019749279121818235,
      "loss": 2.7308,
      "step": 21
    },
    {
      "epoch": 0.04782608695652174,
      "grad_norm": 7.91619348526001,
      "learning_rate": 0.00019672948630390294,
      "loss": 3.4633,
      "step": 22
    }
  ],
  "logging_steps": 1,
  "max_steps": 108,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 22,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8207648948224000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}