{
  "best_metric": 0.016913937404751778,
  "best_model_checkpoint": "saves/ftni/llm3br256/checkpoint-15",
  "epoch": 4.615384615384615,
  "eval_steps": 5,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.3152237832546234,
      "learning_rate": 5e-05,
      "loss": 0.086,
      "step": 1
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.3210161626338959,
      "learning_rate": 0.0001,
      "loss": 0.0854,
      "step": 2
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.1479671597480774,
      "learning_rate": 9.85470908713026e-05,
      "loss": 0.0444,
      "step": 3
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.19353915750980377,
      "learning_rate": 9.42728012826605e-05,
      "loss": 0.0553,
      "step": 4
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.058054178953170776,
      "learning_rate": 8.742553740855506e-05,
      "loss": 0.0269,
      "step": 5
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 0.02396741323173046,
      "eval_runtime": 2.2689,
      "eval_samples_per_second": 22.038,
      "eval_steps_per_second": 5.73,
      "step": 5
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.04341084882616997,
      "learning_rate": 7.840323733655778e-05,
      "loss": 0.0225,
      "step": 6
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.08417680859565735,
      "learning_rate": 6.773024435212678e-05,
      "loss": 0.036,
      "step": 7
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.032517991960048676,
      "learning_rate": 5.602683401276615e-05,
      "loss": 0.0161,
      "step": 8
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.029575150460004807,
      "learning_rate": 4.397316598723385e-05,
      "loss": 0.0172,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.05908314138650894,
      "learning_rate": 3.226975564787322e-05,
      "loss": 0.0264,
      "step": 10
    },
    {
      "epoch": 3.076923076923077,
      "eval_loss": 0.017921479418873787,
      "eval_runtime": 2.2303,
      "eval_samples_per_second": 22.418,
      "eval_steps_per_second": 5.829,
      "step": 10
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 0.0319657139480114,
      "learning_rate": 2.1596762663442218e-05,
      "loss": 0.0166,
      "step": 11
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 0.023381099104881287,
      "learning_rate": 1.257446259144494e-05,
      "loss": 0.0099,
      "step": 12
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.068178690969944,
      "learning_rate": 5.727198717339511e-06,
      "loss": 0.0257,
      "step": 13
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 0.028482988476753235,
      "learning_rate": 1.4529091286973995e-06,
      "loss": 0.0139,
      "step": 14
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 0.024495579302310944,
      "learning_rate": 0.0,
      "loss": 0.0123,
      "step": 15
    },
    {
      "epoch": 4.615384615384615,
      "eval_loss": 0.016913937404751778,
      "eval_runtime": 2.2317,
      "eval_samples_per_second": 22.404,
      "eval_steps_per_second": 5.825,
      "step": 15
    },
    {
      "epoch": 4.615384615384615,
      "step": 15,
      "total_flos": 9464253241540608.0,
      "train_loss": 0.03296535598735015,
      "train_runtime": 107.9418,
      "train_samples_per_second": 4.771,
      "train_steps_per_second": 0.139
    }
  ],
  "logging_steps": 1,
  "max_steps": 15,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9464253241540608.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}