{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 5400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 2.632924795150757,
      "learning_rate": 2.3169601482854498e-05,
      "loss": 0.8913,
      "step": 400
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 2.3553333282470703,
      "learning_rate": 2.1316033364226137e-05,
      "loss": 0.7055,
      "step": 800
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 1.3810371160507202,
      "learning_rate": 1.946246524559778e-05,
      "loss": 0.6621,
      "step": 1200
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 2.1988062858581543,
      "learning_rate": 1.7608897126969417e-05,
      "loss": 0.6527,
      "step": 1600
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 2.7631494998931885,
      "learning_rate": 1.5755329008341056e-05,
      "loss": 0.6426,
      "step": 2000
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 2.627460479736328,
      "learning_rate": 1.3901760889712698e-05,
      "loss": 0.6242,
      "step": 2400
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6276328563690186,
      "eval_runtime": 438.1461,
      "eval_samples_per_second": 1.369,
      "eval_steps_per_second": 0.087,
      "step": 2700
    },
    {
      "epoch": 1.037037037037037,
      "grad_norm": 2.762786865234375,
      "learning_rate": 1.2048192771084338e-05,
      "loss": 0.6002,
      "step": 2800
    },
    {
      "epoch": 1.1851851851851851,
      "grad_norm": 3.1197636127471924,
      "learning_rate": 1.0194624652455977e-05,
      "loss": 0.5341,
      "step": 3200
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 3.382873058319092,
      "learning_rate": 8.341056533827618e-06,
      "loss": 0.5167,
      "step": 3600
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 4.067070007324219,
      "learning_rate": 6.487488415199258e-06,
      "loss": 0.5067,
      "step": 4000
    },
    {
      "epoch": 1.6296296296296298,
      "grad_norm": 6.997328281402588,
      "learning_rate": 4.633920296570899e-06,
      "loss": 0.5007,
      "step": 4400
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 4.422049522399902,
      "learning_rate": 2.7803521779425395e-06,
      "loss": 0.4968,
      "step": 4800
    },
    {
      "epoch": 1.925925925925926,
      "grad_norm": 3.440484046936035,
      "learning_rate": 9.267840593141798e-07,
      "loss": 0.4926,
      "step": 5200
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5777664184570312,
      "eval_runtime": 422.1796,
      "eval_samples_per_second": 1.421,
      "eval_steps_per_second": 0.09,
      "step": 5400
    }
  ],
  "logging_steps": 400,
  "max_steps": 5400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 1.9993910803390464e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}