{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08771929824561403,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0017543859649122807,
      "eval_loss": 0.7453810572624207,
      "eval_runtime": 30.1385,
      "eval_samples_per_second": 7.963,
      "eval_steps_per_second": 3.982,
      "step": 1
    },
    {
      "epoch": 0.008771929824561403,
      "grad_norm": 0.6642146706581116,
      "learning_rate": 5e-05,
      "loss": 0.8338,
      "step": 5
    },
    {
      "epoch": 0.017543859649122806,
      "grad_norm": 0.3736061155796051,
      "learning_rate": 0.0001,
      "loss": 0.6796,
      "step": 10
    },
    {
      "epoch": 0.017543859649122806,
      "eval_loss": 0.6746610999107361,
      "eval_runtime": 30.5654,
      "eval_samples_per_second": 7.852,
      "eval_steps_per_second": 3.926,
      "step": 10
    },
    {
      "epoch": 0.02631578947368421,
      "grad_norm": 0.9030256271362305,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.6721,
      "step": 15
    },
    {
      "epoch": 0.03508771929824561,
      "grad_norm": 0.36247438192367554,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.6739,
      "step": 20
    },
    {
      "epoch": 0.03508771929824561,
      "eval_loss": 0.6100965738296509,
      "eval_runtime": 30.6706,
      "eval_samples_per_second": 7.825,
      "eval_steps_per_second": 3.913,
      "step": 20
    },
    {
      "epoch": 0.043859649122807015,
      "grad_norm": 0.3055374324321747,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.542,
      "step": 25
    },
    {
      "epoch": 0.05263157894736842,
      "grad_norm": 0.35404542088508606,
      "learning_rate": 5e-05,
      "loss": 0.615,
      "step": 30
    },
    {
      "epoch": 0.05263157894736842,
      "eval_loss": 0.5953451991081238,
      "eval_runtime": 30.6917,
      "eval_samples_per_second": 7.82,
      "eval_steps_per_second": 3.91,
      "step": 30
    },
    {
      "epoch": 0.06140350877192982,
      "grad_norm": 0.35547080636024475,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.518,
      "step": 35
    },
    {
      "epoch": 0.07017543859649122,
      "grad_norm": 0.38075515627861023,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.6262,
      "step": 40
    },
    {
      "epoch": 0.07017543859649122,
      "eval_loss": 0.5874656438827515,
      "eval_runtime": 30.784,
      "eval_samples_per_second": 7.796,
      "eval_steps_per_second": 3.898,
      "step": 40
    },
    {
      "epoch": 0.07894736842105263,
      "grad_norm": 0.39305970072746277,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.5954,
      "step": 45
    },
    {
      "epoch": 0.08771929824561403,
      "grad_norm": 0.3901904225349426,
      "learning_rate": 0.0,
      "loss": 0.5802,
      "step": 50
    },
    {
      "epoch": 0.08771929824561403,
      "eval_loss": 0.5861813426017761,
      "eval_runtime": 30.7303,
      "eval_samples_per_second": 7.81,
      "eval_steps_per_second": 3.905,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.84956424814592e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}