{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003092671913900014,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.185343827800028e-05,
      "eval_loss": 2.7244412899017334,
      "eval_runtime": 968.1024,
      "eval_samples_per_second": 7.032,
      "eval_steps_per_second": 3.516,
      "step": 1
    },
    {
      "epoch": 0.0003092671913900014,
      "grad_norm": 3.0380706787109375,
      "learning_rate": 5e-05,
      "loss": 2.6228,
      "step": 5
    },
    {
      "epoch": 0.0006185343827800028,
      "grad_norm": 3.1688232421875,
      "learning_rate": 0.0001,
      "loss": 2.5449,
      "step": 10
    },
    {
      "epoch": 0.0006185343827800028,
      "eval_loss": 2.5041794776916504,
      "eval_runtime": 968.8379,
      "eval_samples_per_second": 7.027,
      "eval_steps_per_second": 3.513,
      "step": 10
    },
    {
      "epoch": 0.0009278015741700042,
      "grad_norm": 3.517625331878662,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.5198,
      "step": 15
    },
    {
      "epoch": 0.0012370687655600056,
      "grad_norm": 4.388193607330322,
      "learning_rate": 8.535533905932738e-05,
      "loss": 2.4901,
      "step": 20
    },
    {
      "epoch": 0.0012370687655600056,
      "eval_loss": 2.343156337738037,
      "eval_runtime": 968.5965,
      "eval_samples_per_second": 7.029,
      "eval_steps_per_second": 3.514,
      "step": 20
    },
    {
      "epoch": 0.001546335956950007,
      "grad_norm": 4.974583625793457,
      "learning_rate": 6.91341716182545e-05,
      "loss": 2.2323,
      "step": 25
    },
    {
      "epoch": 0.0018556031483400084,
      "grad_norm": 4.070034027099609,
      "learning_rate": 5e-05,
      "loss": 2.468,
      "step": 30
    },
    {
      "epoch": 0.0018556031483400084,
      "eval_loss": 2.291248083114624,
      "eval_runtime": 970.6296,
      "eval_samples_per_second": 7.014,
      "eval_steps_per_second": 3.507,
      "step": 30
    },
    {
      "epoch": 0.00216487033973001,
      "grad_norm": 3.880756378173828,
      "learning_rate": 3.086582838174551e-05,
      "loss": 2.3391,
      "step": 35
    },
    {
      "epoch": 0.002474137531120011,
      "grad_norm": 3.2438924312591553,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 2.2421,
      "step": 40
    },
    {
      "epoch": 0.002474137531120011,
      "eval_loss": 2.263791799545288,
      "eval_runtime": 969.6415,
      "eval_samples_per_second": 7.021,
      "eval_steps_per_second": 3.511,
      "step": 40
    },
    {
      "epoch": 0.0027834047225100125,
      "grad_norm": 3.466553211212158,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 2.28,
      "step": 45
    },
    {
      "epoch": 0.003092671913900014,
      "grad_norm": 3.2413833141326904,
      "learning_rate": 0.0,
      "loss": 2.2653,
      "step": 50
    },
    {
      "epoch": 0.003092671913900014,
      "eval_loss": 2.259215831756592,
      "eval_runtime": 970.5443,
      "eval_samples_per_second": 7.015,
      "eval_steps_per_second": 3.507,
      "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": 1.772537288589312e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}