{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 40000,
  "global_step": 5400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15,
      "grad_norm": 2.7106876373291016,
      "learning_rate": 2.3169601482854498e-05,
      "loss": 0.6705,
      "step": 400
    },
    {
      "epoch": 0.3,
      "grad_norm": 3.0413529872894287,
      "learning_rate": 2.1316033364226137e-05,
      "loss": 0.5494,
      "step": 800
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.6685305833816528,
      "learning_rate": 1.946246524559778e-05,
      "loss": 0.5135,
      "step": 1200
    },
    {
      "epoch": 0.59,
      "grad_norm": 2.7073967456817627,
      "learning_rate": 1.7608897126969417e-05,
      "loss": 0.5007,
      "step": 1600
    },
    {
      "epoch": 0.74,
      "grad_norm": 2.5799527168273926,
      "learning_rate": 1.5755329008341056e-05,
      "loss": 0.491,
      "step": 2000
    },
    {
      "epoch": 0.89,
      "grad_norm": 3.942272901535034,
      "learning_rate": 1.3901760889712698e-05,
      "loss": 0.4789,
      "step": 2400
    },
    {
      "epoch": 1.04,
      "grad_norm": 3.0384867191314697,
      "learning_rate": 1.2048192771084338e-05,
      "loss": 0.4501,
      "step": 2800
    },
    {
      "epoch": 1.19,
      "grad_norm": 4.298065662384033,
      "learning_rate": 1.0194624652455977e-05,
      "loss": 0.3766,
      "step": 3200
    },
    {
      "epoch": 1.33,
      "grad_norm": 4.440635681152344,
      "learning_rate": 8.341056533827618e-06,
      "loss": 0.3603,
      "step": 3600
    },
    {
      "epoch": 1.48,
      "grad_norm": 3.9259891510009766,
      "learning_rate": 6.487488415199258e-06,
      "loss": 0.3597,
      "step": 4000
    },
    {
      "epoch": 1.63,
      "grad_norm": 5.881516933441162,
      "learning_rate": 4.633920296570899e-06,
      "loss": 0.3443,
      "step": 4400
    },
    {
      "epoch": 1.78,
      "grad_norm": 5.677618980407715,
      "learning_rate": 2.7803521779425395e-06,
      "loss": 0.3441,
      "step": 4800
    },
    {
      "epoch": 1.93,
      "grad_norm": 3.5774009227752686,
      "learning_rate": 9.267840593141798e-07,
      "loss": 0.34,
      "step": 5200
    }
  ],
  "logging_steps": 400,
  "max_steps": 5400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 1.8944577171382272e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}