{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03676470588235294,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007352941176470588,
      "eval_loss": 2.2312276363372803,
      "eval_runtime": 55.0264,
      "eval_samples_per_second": 10.413,
      "eval_steps_per_second": 5.216,
      "step": 1
    },
    {
      "epoch": 0.003676470588235294,
      "grad_norm": 1.701627254486084,
      "learning_rate": 5e-05,
      "loss": 8.9151,
      "step": 5
    },
    {
      "epoch": 0.007352941176470588,
      "grad_norm": 2.1781604290008545,
      "learning_rate": 0.0001,
      "loss": 8.8291,
      "step": 10
    },
    {
      "epoch": 0.007352941176470588,
      "eval_loss": 2.177993059158325,
      "eval_runtime": 55.6871,
      "eval_samples_per_second": 10.29,
      "eval_steps_per_second": 5.154,
      "step": 10
    },
    {
      "epoch": 0.011029411764705883,
      "grad_norm": 2.438439130783081,
      "learning_rate": 9.619397662556435e-05,
      "loss": 8.845,
      "step": 15
    },
    {
      "epoch": 0.014705882352941176,
      "grad_norm": 2.578474998474121,
      "learning_rate": 8.535533905932738e-05,
      "loss": 8.6161,
      "step": 20
    },
    {
      "epoch": 0.014705882352941176,
      "eval_loss": 2.130336046218872,
      "eval_runtime": 55.7713,
      "eval_samples_per_second": 10.274,
      "eval_steps_per_second": 5.146,
      "step": 20
    },
    {
      "epoch": 0.01838235294117647,
      "grad_norm": 2.473172903060913,
      "learning_rate": 6.91341716182545e-05,
      "loss": 8.6475,
      "step": 25
    },
    {
      "epoch": 0.022058823529411766,
      "grad_norm": 2.351734161376953,
      "learning_rate": 5e-05,
      "loss": 8.1037,
      "step": 30
    },
    {
      "epoch": 0.022058823529411766,
      "eval_loss": 2.119811773300171,
      "eval_runtime": 55.9096,
      "eval_samples_per_second": 10.249,
      "eval_steps_per_second": 5.133,
      "step": 30
    },
    {
      "epoch": 0.025735294117647058,
      "grad_norm": 3.0046396255493164,
      "learning_rate": 3.086582838174551e-05,
      "loss": 8.2632,
      "step": 35
    },
    {
      "epoch": 0.029411764705882353,
      "grad_norm": 2.4054322242736816,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 8.2349,
      "step": 40
    },
    {
      "epoch": 0.029411764705882353,
      "eval_loss": 2.1164493560791016,
      "eval_runtime": 56.0657,
      "eval_samples_per_second": 10.22,
      "eval_steps_per_second": 5.119,
      "step": 40
    },
    {
      "epoch": 0.03308823529411765,
      "grad_norm": 2.259829044342041,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 8.6449,
      "step": 45
    },
    {
      "epoch": 0.03676470588235294,
      "grad_norm": 2.5194344520568848,
      "learning_rate": 0.0,
      "loss": 8.4466,
      "step": 50
    },
    {
      "epoch": 0.03676470588235294,
      "eval_loss": 2.1157116889953613,
      "eval_runtime": 55.9108,
      "eval_samples_per_second": 10.248,
      "eval_steps_per_second": 5.133,
      "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": 1.4260472119296e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}