{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.6923076923076925,
  "eval_steps": 100,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.6472,
      "step": 10
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.3259,
      "step": 20
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.224,
      "step": 30
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002,
      "loss": 0.1295,
      "step": 40
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002,
      "loss": 0.046,
      "step": 50
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.0327,
      "step": 60
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002,
      "loss": 0.0207,
      "step": 70
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0002,
      "loss": 0.0176,
      "step": 80
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.015,
      "step": 90
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002,
      "loss": 0.0139,
      "step": 100
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.01566016674041748,
      "eval_runtime": 4.9917,
      "eval_samples_per_second": 13.022,
      "eval_steps_per_second": 1.803,
      "step": 100
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0002,
      "loss": 0.0117,
      "step": 110
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.0112,
      "step": 120
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0002,
      "loss": 0.0105,
      "step": 130
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0002,
      "loss": 0.0097,
      "step": 140
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.0002,
      "loss": 0.0104,
      "step": 150
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0002,
      "loss": 0.0118,
      "step": 160
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.0002,
      "loss": 0.01,
      "step": 170
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0002,
      "loss": 0.0091,
      "step": 180
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.0002,
      "loss": 0.0072,
      "step": 190
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0002,
      "loss": 0.0079,
      "step": 200
    },
    {
      "epoch": 3.08,
      "eval_loss": 0.012308033183217049,
      "eval_runtime": 4.9748,
      "eval_samples_per_second": 13.066,
      "eval_steps_per_second": 1.809,
      "step": 200
    },
    {
      "epoch": 3.23,
      "learning_rate": 0.0002,
      "loss": 0.0077,
      "step": 210
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0002,
      "loss": 0.0081,
      "step": 220
    },
    {
      "epoch": 3.54,
      "learning_rate": 0.0002,
      "loss": 0.0072,
      "step": 230
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0002,
      "loss": 0.0073,
      "step": 240
    },
    {
      "epoch": 3.85,
      "learning_rate": 0.0002,
      "loss": 0.0068,
      "step": 250
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0002,
      "loss": 0.007,
      "step": 260
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.0002,
      "loss": 0.0057,
      "step": 270
    },
    {
      "epoch": 4.31,
      "learning_rate": 0.0002,
      "loss": 0.0072,
      "step": 280
    },
    {
      "epoch": 4.46,
      "learning_rate": 0.0002,
      "loss": 0.006,
      "step": 290
    },
    {
      "epoch": 4.62,
      "learning_rate": 0.0002,
      "loss": 0.0062,
      "step": 300
    },
    {
      "epoch": 4.62,
      "eval_loss": 0.011374865658581257,
      "eval_runtime": 4.9747,
      "eval_samples_per_second": 13.066,
      "eval_steps_per_second": 1.809,
      "step": 300
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.0002,
      "loss": 0.0063,
      "step": 310
    },
    {
      "epoch": 4.92,
      "learning_rate": 0.0002,
      "loss": 0.0067,
      "step": 320
    },
    {
      "epoch": 5.08,
      "learning_rate": 0.0002,
      "loss": 0.0052,
      "step": 330
    },
    {
      "epoch": 5.23,
      "learning_rate": 0.0002,
      "loss": 0.0069,
      "step": 340
    },
    {
      "epoch": 5.38,
      "learning_rate": 0.0002,
      "loss": 0.0054,
      "step": 350
    },
    {
      "epoch": 5.54,
      "learning_rate": 0.0002,
      "loss": 0.0061,
      "step": 360
    },
    {
      "epoch": 5.69,
      "learning_rate": 0.0002,
      "loss": 0.006,
      "step": 370
    },
    {
      "epoch": 5.85,
      "learning_rate": 0.0002,
      "loss": 0.0061,
      "step": 380
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.0002,
      "loss": 0.0064,
      "step": 390
    },
    {
      "epoch": 6.15,
      "learning_rate": 0.0002,
      "loss": 0.0047,
      "step": 400
    },
    {
      "epoch": 6.15,
      "eval_loss": 0.009761717170476913,
      "eval_runtime": 4.974,
      "eval_samples_per_second": 13.068,
      "eval_steps_per_second": 1.809,
      "step": 400
    },
    {
      "epoch": 6.31,
      "learning_rate": 0.0002,
      "loss": 0.0051,
      "step": 410
    },
    {
      "epoch": 6.46,
      "learning_rate": 0.0002,
      "loss": 0.0053,
      "step": 420
    },
    {
      "epoch": 6.62,
      "learning_rate": 0.0002,
      "loss": 0.0051,
      "step": 430
    },
    {
      "epoch": 6.77,
      "learning_rate": 0.0002,
      "loss": 0.0052,
      "step": 440
    },
    {
      "epoch": 6.92,
      "learning_rate": 0.0002,
      "loss": 0.005,
      "step": 450
    },
    {
      "epoch": 7.08,
      "learning_rate": 0.0002,
      "loss": 0.0052,
      "step": 460
    },
    {
      "epoch": 7.23,
      "learning_rate": 0.0002,
      "loss": 0.0042,
      "step": 470
    },
    {
      "epoch": 7.38,
      "learning_rate": 0.0002,
      "loss": 0.0051,
      "step": 480
    },
    {
      "epoch": 7.54,
      "learning_rate": 0.0002,
      "loss": 0.005,
      "step": 490
    },
    {
      "epoch": 7.69,
      "learning_rate": 0.0002,
      "loss": 0.0049,
      "step": 500
    },
    {
      "epoch": 7.69,
      "eval_loss": 0.01100065279752016,
      "eval_runtime": 4.9709,
      "eval_samples_per_second": 13.076,
      "eval_steps_per_second": 1.811,
      "step": 500
    }
  ],
  "logging_steps": 10,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 8.7710798708736e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}