{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.018587602440923952,
  "eval_steps": 100000000,
  "global_step": 20000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 7.7296,
      "step": 500
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.9999535305620204e-05,
      "loss": 6.5786,
      "step": 1000
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.99990706112404e-05,
      "loss": 6.2317,
      "step": 1500
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.99986059168606e-05,
      "loss": 6.0053,
      "step": 2000
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.99981412224808e-05,
      "loss": 5.8615,
      "step": 2500
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.9997676528101e-05,
      "loss": 5.6794,
      "step": 3000
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.99972118337212e-05,
      "loss": 5.5481,
      "step": 3500
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.9996747139341396e-05,
      "loss": 5.4375,
      "step": 4000
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.99962824449616e-05,
      "loss": 5.3391,
      "step": 4500
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.99958177505818e-05,
      "loss": 5.2598,
      "step": 5000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9995353056202e-05,
      "loss": 5.17,
      "step": 5500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9994888361822204e-05,
      "loss": 5.0886,
      "step": 6000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.99944236674424e-05,
      "loss": 5.0128,
      "step": 6500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.99939589730626e-05,
      "loss": 4.9504,
      "step": 7000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.99934942786828e-05,
      "loss": 4.8755,
      "step": 7500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9993029584303e-05,
      "loss": 4.8402,
      "step": 8000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.99925648899232e-05,
      "loss": 4.7674,
      "step": 8500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9992100195543396e-05,
      "loss": 4.7254,
      "step": 9000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.99916355011636e-05,
      "loss": 4.6749,
      "step": 9500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.99911708067838e-05,
      "loss": 4.6135,
      "step": 10000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9990706112403995e-05,
      "loss": 4.5239,
      "step": 10500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.99902414180242e-05,
      "loss": 4.4729,
      "step": 11000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.998977672364439e-05,
      "loss": 4.4434,
      "step": 11500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9989312029264594e-05,
      "loss": 4.3587,
      "step": 12000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9988847334884796e-05,
      "loss": 4.3126,
      "step": 12500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.998838264050499e-05,
      "loss": 4.2658,
      "step": 13000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.998791794612519e-05,
      "loss": 4.2412,
      "step": 13500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9987453251745395e-05,
      "loss": 4.2035,
      "step": 14000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.998698855736559e-05,
      "loss": 4.1695,
      "step": 14500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.998652386298579e-05,
      "loss": 4.1406,
      "step": 15000
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9986059168605994e-05,
      "loss": 4.1275,
      "step": 15500
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.9985594474226196e-05,
      "loss": 4.1032,
      "step": 16000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.99851297798464e-05,
      "loss": 4.0588,
      "step": 16500
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.9984665085466594e-05,
      "loss": 4.0536,
      "step": 17000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.9984200391086796e-05,
      "loss": 3.9979,
      "step": 17500
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.998373569670699e-05,
      "loss": 3.9913,
      "step": 18000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.998327100232719e-05,
      "loss": 3.988,
      "step": 18500
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.9982806307947395e-05,
      "loss": 3.9575,
      "step": 19000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.998234161356759e-05,
      "loss": 3.942,
      "step": 19500
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.998187691918779e-05,
      "loss": 3.9258,
      "step": 20000
    }
  ],
  "logging_steps": 500,
  "max_steps": 53799300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 1000,
  "total_flos": 3.04664543232e+17,
  "train_batch_size": 7,
  "trial_name": null,
  "trial_params": null
}