{
  "best_metric": 0.6280947923660278,
  "best_model_checkpoint": "experiments/checkpoint-300",
  "epoch": 22.58823529411765,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.75,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 3.2272,
      "step": 10
    },
    {
      "epoch": 1.51,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 3.068,
      "step": 20
    },
    {
      "epoch": 2.26,
      "learning_rate": 8.999999999999999e-05,
      "loss": 2.6558,
      "step": 30
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.00011399999999999999,
      "loss": 1.8885,
      "step": 40
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.00014399999999999998,
      "loss": 1.3458,
      "step": 50
    },
    {
      "epoch": 3.76,
      "eval_loss": 1.1582125425338745,
      "eval_runtime": 6.163,
      "eval_samples_per_second": 32.451,
      "eval_steps_per_second": 4.056,
      "step": 50
    },
    {
      "epoch": 4.52,
      "learning_rate": 0.00017399999999999997,
      "loss": 1.1036,
      "step": 60
    },
    {
      "epoch": 5.27,
      "learning_rate": 0.000204,
      "loss": 0.9892,
      "step": 70
    },
    {
      "epoch": 6.02,
      "learning_rate": 0.000234,
      "loss": 0.9294,
      "step": 80
    },
    {
      "epoch": 6.78,
      "learning_rate": 0.00026399999999999997,
      "loss": 0.8787,
      "step": 90
    },
    {
      "epoch": 7.53,
      "learning_rate": 0.000294,
      "loss": 0.7888,
      "step": 100
    },
    {
      "epoch": 7.53,
      "eval_loss": 0.7398544549942017,
      "eval_runtime": 6.2173,
      "eval_samples_per_second": 32.168,
      "eval_steps_per_second": 4.021,
      "step": 100
    },
    {
      "epoch": 8.28,
      "learning_rate": 0.00028799999999999995,
      "loss": 0.7733,
      "step": 110
    },
    {
      "epoch": 9.04,
      "learning_rate": 0.00027299999999999997,
      "loss": 0.7331,
      "step": 120
    },
    {
      "epoch": 9.79,
      "learning_rate": 0.000258,
      "loss": 0.7048,
      "step": 130
    },
    {
      "epoch": 10.54,
      "learning_rate": 0.000243,
      "loss": 0.6899,
      "step": 140
    },
    {
      "epoch": 11.29,
      "learning_rate": 0.00022799999999999999,
      "loss": 0.665,
      "step": 150
    },
    {
      "epoch": 11.29,
      "eval_loss": 0.6526339650154114,
      "eval_runtime": 6.1516,
      "eval_samples_per_second": 32.512,
      "eval_steps_per_second": 4.064,
      "step": 150
    },
    {
      "epoch": 12.05,
      "learning_rate": 0.00021299999999999997,
      "loss": 0.6446,
      "step": 160
    },
    {
      "epoch": 12.8,
      "learning_rate": 0.000198,
      "loss": 0.6263,
      "step": 170
    },
    {
      "epoch": 13.55,
      "learning_rate": 0.00018299999999999998,
      "loss": 0.631,
      "step": 180
    },
    {
      "epoch": 14.31,
      "learning_rate": 0.000168,
      "loss": 0.6041,
      "step": 190
    },
    {
      "epoch": 15.06,
      "learning_rate": 0.00015299999999999998,
      "loss": 0.6041,
      "step": 200
    },
    {
      "epoch": 15.06,
      "eval_loss": 0.6296558380126953,
      "eval_runtime": 6.2161,
      "eval_samples_per_second": 32.175,
      "eval_steps_per_second": 4.022,
      "step": 200
    },
    {
      "epoch": 15.81,
      "learning_rate": 0.000138,
      "loss": 0.5883,
      "step": 210
    },
    {
      "epoch": 16.56,
      "learning_rate": 0.00012299999999999998,
      "loss": 0.5914,
      "step": 220
    },
    {
      "epoch": 17.32,
      "learning_rate": 0.00010799999999999998,
      "loss": 0.5698,
      "step": 230
    },
    {
      "epoch": 18.07,
      "learning_rate": 9.3e-05,
      "loss": 0.5689,
      "step": 240
    },
    {
      "epoch": 18.82,
      "learning_rate": 7.8e-05,
      "loss": 0.5654,
      "step": 250
    },
    {
      "epoch": 18.82,
      "eval_loss": 0.6283386945724487,
      "eval_runtime": 6.3224,
      "eval_samples_per_second": 31.634,
      "eval_steps_per_second": 3.954,
      "step": 250
    },
    {
      "epoch": 19.58,
      "learning_rate": 6.299999999999999e-05,
      "loss": 0.5469,
      "step": 260
    },
    {
      "epoch": 20.33,
      "learning_rate": 4.7999999999999994e-05,
      "loss": 0.564,
      "step": 270
    },
    {
      "epoch": 21.08,
      "learning_rate": 3.2999999999999996e-05,
      "loss": 0.5409,
      "step": 280
    },
    {
      "epoch": 21.84,
      "learning_rate": 1.7999999999999997e-05,
      "loss": 0.5464,
      "step": 290
    },
    {
      "epoch": 22.59,
      "learning_rate": 2.9999999999999997e-06,
      "loss": 0.5396,
      "step": 300
    },
    {
      "epoch": 22.59,
      "eval_loss": 0.6280947923660278,
      "eval_runtime": 6.5628,
      "eval_samples_per_second": 30.475,
      "eval_steps_per_second": 3.809,
      "step": 300
    }
  ],
  "max_steps": 300,
  "num_train_epochs": 24,
  "total_flos": 2.072193504116736e+17,
  "trial_name": null,
  "trial_params": null
}