{
  "best_metric": 1.1660184860229492,
  "best_model_checkpoint": "experiments/checkpoint-200",
  "epoch": 21.867881548974943,
  "eval_steps": 50,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.73,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 2.5173,
      "step": 10
    },
    {
      "epoch": 1.46,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 2.4599,
      "step": 20
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.999999999999999e-05,
      "loss": 2.2163,
      "step": 30
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.00011999999999999999,
      "loss": 1.7728,
      "step": 40
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.00015,
      "loss": 1.2772,
      "step": 50
    },
    {
      "epoch": 3.64,
      "eval_loss": 1.2498592138290405,
      "eval_runtime": 8.4312,
      "eval_samples_per_second": 23.721,
      "eval_steps_per_second": 2.965,
      "step": 50
    },
    {
      "epoch": 4.37,
      "learning_rate": 0.00017999999999999998,
      "loss": 1.1737,
      "step": 60
    },
    {
      "epoch": 5.1,
      "learning_rate": 0.00020999999999999998,
      "loss": 1.1581,
      "step": 70
    },
    {
      "epoch": 5.83,
      "learning_rate": 0.00023999999999999998,
      "loss": 1.1455,
      "step": 80
    },
    {
      "epoch": 6.56,
      "learning_rate": 0.00027,
      "loss": 1.121,
      "step": 90
    },
    {
      "epoch": 7.29,
      "learning_rate": 0.0003,
      "loss": 1.1123,
      "step": 100
    },
    {
      "epoch": 7.29,
      "eval_loss": 1.187015175819397,
      "eval_runtime": 10.2578,
      "eval_samples_per_second": 19.497,
      "eval_steps_per_second": 2.437,
      "step": 100
    },
    {
      "epoch": 8.02,
      "learning_rate": 0.000285,
      "loss": 1.1124,
      "step": 110
    },
    {
      "epoch": 8.75,
      "learning_rate": 0.00027,
      "loss": 1.089,
      "step": 120
    },
    {
      "epoch": 9.48,
      "learning_rate": 0.00025499999999999996,
      "loss": 1.0991,
      "step": 130
    },
    {
      "epoch": 10.21,
      "learning_rate": 0.00023999999999999998,
      "loss": 1.0759,
      "step": 140
    },
    {
      "epoch": 10.93,
      "learning_rate": 0.000225,
      "loss": 1.0762,
      "step": 150
    },
    {
      "epoch": 10.93,
      "eval_loss": 1.1669864654541016,
      "eval_runtime": 10.2044,
      "eval_samples_per_second": 19.599,
      "eval_steps_per_second": 2.45,
      "step": 150
    },
    {
      "epoch": 11.66,
      "learning_rate": 0.00020999999999999998,
      "loss": 1.0738,
      "step": 160
    },
    {
      "epoch": 12.39,
      "learning_rate": 0.000195,
      "loss": 1.0424,
      "step": 170
    },
    {
      "epoch": 13.12,
      "learning_rate": 0.00017999999999999998,
      "loss": 1.0701,
      "step": 180
    },
    {
      "epoch": 13.85,
      "learning_rate": 0.000165,
      "loss": 1.0486,
      "step": 190
    },
    {
      "epoch": 14.58,
      "learning_rate": 0.00015,
      "loss": 1.0409,
      "step": 200
    },
    {
      "epoch": 14.58,
      "eval_loss": 1.1660184860229492,
      "eval_runtime": 10.2373,
      "eval_samples_per_second": 19.536,
      "eval_steps_per_second": 2.442,
      "step": 200
    },
    {
      "epoch": 15.31,
      "learning_rate": 0.000135,
      "loss": 1.0404,
      "step": 210
    },
    {
      "epoch": 16.04,
      "learning_rate": 0.00011999999999999999,
      "loss": 1.027,
      "step": 220
    },
    {
      "epoch": 16.77,
      "learning_rate": 0.00010499999999999999,
      "loss": 1.0251,
      "step": 230
    },
    {
      "epoch": 17.49,
      "learning_rate": 8.999999999999999e-05,
      "loss": 1.0183,
      "step": 240
    },
    {
      "epoch": 18.22,
      "learning_rate": 7.5e-05,
      "loss": 1.0063,
      "step": 250
    },
    {
      "epoch": 18.22,
      "eval_loss": 1.1671104431152344,
      "eval_runtime": 10.1032,
      "eval_samples_per_second": 19.796,
      "eval_steps_per_second": 2.474,
      "step": 250
    },
    {
      "epoch": 18.95,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 1.0167,
      "step": 260
    },
    {
      "epoch": 19.68,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 0.9973,
      "step": 270
    },
    {
      "epoch": 20.41,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 1.0219,
      "step": 280
    },
    {
      "epoch": 21.14,
      "learning_rate": 1.4999999999999999e-05,
      "loss": 0.993,
      "step": 290
    },
    {
      "epoch": 21.87,
      "learning_rate": 0.0,
      "loss": 0.9988,
      "step": 300
    },
    {
      "epoch": 21.87,
      "eval_loss": 1.1713457107543945,
      "eval_runtime": 10.0918,
      "eval_samples_per_second": 19.818,
      "eval_steps_per_second": 2.477,
      "step": 300
    }
  ],
  "logging_steps": 10,
  "max_steps": 300,
  "num_train_epochs": 24,
  "save_steps": 50,
  "total_flos": 3.898336244387021e+17,
  "trial_name": null,
  "trial_params": null
}