{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 252,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "grad_norm": 614343.125,
      "learning_rate": 0.0002988358809900258,
      "loss": 1.1347,
      "step": 10
    },
    {
      "epoch": 0.24,
      "grad_norm": 96113.765625,
      "learning_rate": 0.00029536159293436166,
      "loss": 0.6949,
      "step": 20
    },
    {
      "epoch": 0.36,
      "grad_norm": 127274.109375,
      "learning_rate": 0.00028963106229663063,
      "loss": 0.6215,
      "step": 30
    },
    {
      "epoch": 0.48,
      "grad_norm": 124623.640625,
      "learning_rate": 0.0002817332360055343,
      "loss": 0.7171,
      "step": 40
    },
    {
      "epoch": 0.6,
      "grad_norm": 94723.8046875,
      "learning_rate": 0.0002717907008573785,
      "loss": 0.6622,
      "step": 50
    },
    {
      "epoch": 0.71,
      "grad_norm": 79811.34375,
      "learning_rate": 0.0002599577807744739,
      "loss": 0.641,
      "step": 60
    },
    {
      "epoch": 0.83,
      "grad_norm": 81335.1015625,
      "learning_rate": 0.0002464181414529809,
      "loss": 0.6571,
      "step": 70
    },
    {
      "epoch": 0.95,
      "grad_norm": 92003.125,
      "learning_rate": 0.0002313819395798639,
      "loss": 0.6412,
      "step": 80
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6449399590492249,
      "eval_runtime": 123.5827,
      "eval_samples_per_second": 0.898,
      "eval_steps_per_second": 0.227,
      "step": 84
    },
    {
      "epoch": 1.07,
      "grad_norm": 80496.6328125,
      "learning_rate": 0.00021508256086763368,
      "loss": 0.5549,
      "step": 90
    },
    {
      "epoch": 1.19,
      "grad_norm": 91778.5078125,
      "learning_rate": 0.00019777299753775265,
      "loss": 0.468,
      "step": 100
    },
    {
      "epoch": 1.31,
      "grad_norm": 91146.3828125,
      "learning_rate": 0.0001797219214799096,
      "loss": 0.4822,
      "step": 110
    },
    {
      "epoch": 1.43,
      "grad_norm": 103467.90625,
      "learning_rate": 0.00016120951403796364,
      "loss": 0.539,
      "step": 120
    },
    {
      "epoch": 1.55,
      "grad_norm": 99161.296875,
      "learning_rate": 0.0001425231171508954,
      "loss": 0.5263,
      "step": 130
    },
    {
      "epoch": 1.67,
      "grad_norm": 64898.0390625,
      "learning_rate": 0.00012395277334996044,
      "loss": 0.4564,
      "step": 140
    },
    {
      "epoch": 1.79,
      "grad_norm": 64908.39453125,
      "learning_rate": 0.00010578672383836435,
      "loss": 0.4603,
      "step": 150
    },
    {
      "epoch": 1.9,
      "grad_norm": 65386.96875,
      "learning_rate": 8.830693453040829e-05,
      "loss": 0.497,
      "step": 160
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6082237958908081,
      "eval_runtime": 123.4189,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.227,
      "step": 168
    },
    {
      "epoch": 2.02,
      "grad_norm": 60998.84765625,
      "learning_rate": 7.17847194930753e-05,
      "loss": 0.4428,
      "step": 170
    },
    {
      "epoch": 2.14,
      "grad_norm": 136505.265625,
      "learning_rate": 5.6476529721189974e-05,
      "loss": 0.305,
      "step": 180
    },
    {
      "epoch": 2.26,
      "grad_norm": 72723.578125,
      "learning_rate": 4.261997261104223e-05,
      "loss": 0.3237,
      "step": 190
    },
    {
      "epoch": 2.38,
      "grad_norm": 89701.6875,
      "learning_rate": 3.0430123916561672e-05,
      "loss": 0.298,
      "step": 200
    },
    {
      "epoch": 2.5,
      "grad_norm": 72959.9453125,
      "learning_rate": 2.009618943233419e-05,
      "loss": 0.3401,
      "step": 210
    },
    {
      "epoch": 2.62,
      "grad_norm": 73923.734375,
      "learning_rate": 1.1778568219438839e-05,
      "loss": 0.3294,
      "step": 220
    },
    {
      "epoch": 2.74,
      "grad_norm": 87842.8828125,
      "learning_rate": 5.606362957498195e-06,
      "loss": 0.3063,
      "step": 230
    },
    {
      "epoch": 2.86,
      "grad_norm": 75173.09375,
      "learning_rate": 1.6753760662307215e-06,
      "loss": 0.3092,
      "step": 240
    },
    {
      "epoch": 2.98,
      "grad_norm": 59458.79296875,
      "learning_rate": 4.662269987756317e-08,
      "loss": 0.3357,
      "step": 250
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6374169588088989,
      "eval_runtime": 123.4407,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.227,
      "step": 252
    },
    {
      "epoch": 3.0,
      "step": 252,
      "total_flos": 4.311746565075763e+16,
      "train_loss": 0.5089443905486001,
      "train_runtime": 3502.9275,
      "train_samples_per_second": 0.285,
      "train_steps_per_second": 0.072
    }
  ],
  "logging_steps": 10,
  "max_steps": 252,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10,
  "total_flos": 4.311746565075763e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}