{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 40.0,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0,
      "grad_norm": 0.354248046875,
      "learning_rate": 0.00019876883405951377,
      "loss": 0.9502,
      "step": 10
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.322021484375,
      "learning_rate": 0.00019510565162951537,
      "loss": 0.74,
      "step": 20
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.179931640625,
      "learning_rate": 0.0001891006524188368,
      "loss": 0.6284,
      "step": 30
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.132080078125,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.5907,
      "step": 40
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.12481689453125,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.5757,
      "step": 50
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.130859375,
      "learning_rate": 0.00015877852522924732,
      "loss": 0.5632,
      "step": 60
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.143310546875,
      "learning_rate": 0.00014539904997395468,
      "loss": 0.5519,
      "step": 70
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.1419677734375,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.5405,
      "step": 80
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.1556396484375,
      "learning_rate": 0.0001156434465040231,
      "loss": 0.5309,
      "step": 90
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.16796875,
      "learning_rate": 0.0001,
      "loss": 0.5212,
      "step": 100
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.179443359375,
      "learning_rate": 8.435655349597689e-05,
      "loss": 0.513,
      "step": 110
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.19970703125,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.5052,
      "step": 120
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.1968994140625,
      "learning_rate": 5.4600950026045326e-05,
      "loss": 0.4986,
      "step": 130
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.2119140625,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.4939,
      "step": 140
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.2227783203125,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 0.49,
      "step": 150
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.200439453125,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.4867,
      "step": 160
    },
    {
      "epoch": 34.0,
      "grad_norm": 0.2117919921875,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 0.4846,
      "step": 170
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.2088623046875,
      "learning_rate": 4.8943483704846475e-06,
      "loss": 0.4832,
      "step": 180
    },
    {
      "epoch": 38.0,
      "grad_norm": 0.2152099609375,
      "learning_rate": 1.231165940486234e-06,
      "loss": 0.4829,
      "step": 190
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.211181640625,
      "learning_rate": 0.0,
      "loss": 0.4828,
      "step": 200
    },
    {
      "epoch": 40.0,
      "step": 200,
      "total_flos": 4.000287962431488e+16,
      "train_loss": 0.5556855773925782,
      "train_runtime": 573.2976,
      "train_samples_per_second": 11.163,
      "train_steps_per_second": 0.349
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 4.000287962431488e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}