{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.592910848549947,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34,
      "grad_norm": 0.6707318425178528,
      "learning_rate": 1.97816091954023e-05,
      "loss": 2.2888,
      "step": 20
    },
    {
      "epoch": 0.69,
      "grad_norm": NaN,
      "learning_rate": 1.9574712643678162e-05,
      "loss": 2.0848,
      "step": 40
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.7721680402755737,
      "learning_rate": 1.9344827586206897e-05,
      "loss": 2.0044,
      "step": 60
    },
    {
      "epoch": 1.37,
      "grad_norm": 1.1140433549880981,
      "learning_rate": 1.9126436781609195e-05,
      "loss": 1.8016,
      "step": 80
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.7205075621604919,
      "learning_rate": 1.8896551724137934e-05,
      "loss": 1.7217,
      "step": 100
    },
    {
      "epoch": 2.06,
      "grad_norm": 0.8933233618736267,
      "learning_rate": 1.866666666666667e-05,
      "loss": 1.5705,
      "step": 120
    },
    {
      "epoch": 2.41,
      "grad_norm": 0.7114273905754089,
      "learning_rate": 1.8436781609195404e-05,
      "loss": 1.4006,
      "step": 140
    },
    {
      "epoch": 2.75,
      "grad_norm": 0.7229479551315308,
      "learning_rate": 1.820689655172414e-05,
      "loss": 1.3137,
      "step": 160
    },
    {
      "epoch": 3.09,
      "grad_norm": 0.9370490908622742,
      "learning_rate": 1.7977011494252874e-05,
      "loss": 1.1898,
      "step": 180
    },
    {
      "epoch": 3.44,
      "grad_norm": 0.6051978468894958,
      "learning_rate": 1.774712643678161e-05,
      "loss": 1.1229,
      "step": 200
    },
    {
      "epoch": 3.78,
      "grad_norm": 0.6857028007507324,
      "learning_rate": 1.7517241379310347e-05,
      "loss": 1.051,
      "step": 220
    },
    {
      "epoch": 4.12,
      "grad_norm": 0.6715748310089111,
      "learning_rate": 1.7287356321839082e-05,
      "loss": 0.9894,
      "step": 240
    },
    {
      "epoch": 4.47,
      "grad_norm": 0.5918118953704834,
      "learning_rate": 1.7057471264367817e-05,
      "loss": 0.9687,
      "step": 260
    },
    {
      "epoch": 4.81,
      "grad_norm": 0.6621690392494202,
      "learning_rate": 1.6827586206896552e-05,
      "loss": 0.9199,
      "step": 280
    },
    {
      "epoch": 5.16,
      "grad_norm": 0.6697206497192383,
      "learning_rate": 1.659770114942529e-05,
      "loss": 0.9303,
      "step": 300
    },
    {
      "epoch": 5.5,
      "grad_norm": 0.8184316158294678,
      "learning_rate": 1.6367816091954025e-05,
      "loss": 0.8898,
      "step": 320
    },
    {
      "epoch": 5.84,
      "grad_norm": 0.6429987549781799,
      "learning_rate": 1.613793103448276e-05,
      "loss": 0.8623,
      "step": 340
    },
    {
      "epoch": 6.19,
      "grad_norm": 0.7518043518066406,
      "learning_rate": 1.5908045977011495e-05,
      "loss": 0.8239,
      "step": 360
    },
    {
      "epoch": 6.53,
      "grad_norm": 0.6667824983596802,
      "learning_rate": 1.567816091954023e-05,
      "loss": 0.8119,
      "step": 380
    },
    {
      "epoch": 6.87,
      "grad_norm": 0.8569457530975342,
      "learning_rate": 1.5448275862068965e-05,
      "loss": 0.8139,
      "step": 400
    },
    {
      "epoch": 7.22,
      "grad_norm": 0.7754850387573242,
      "learning_rate": 1.5218390804597702e-05,
      "loss": 0.7835,
      "step": 420
    },
    {
      "epoch": 7.56,
      "grad_norm": 1.159196138381958,
      "learning_rate": 1.4988505747126439e-05,
      "loss": 0.7546,
      "step": 440
    },
    {
      "epoch": 7.91,
      "grad_norm": 1.119764804840088,
      "learning_rate": 1.4758620689655174e-05,
      "loss": 0.7571,
      "step": 460
    },
    {
      "epoch": 8.25,
      "grad_norm": 1.3600786924362183,
      "learning_rate": 1.452873563218391e-05,
      "loss": 0.7451,
      "step": 480
    },
    {
      "epoch": 8.59,
      "grad_norm": 0.7608994245529175,
      "learning_rate": 1.4298850574712644e-05,
      "loss": 0.7109,
      "step": 500
    }
  ],
  "logging_steps": 20,
  "max_steps": 1740,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 1.2995638935552e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}