{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 7068,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21222410865874364,
      "grad_norm": 66.5,
      "learning_rate": 7.072135785007072e-05,
      "loss": 2.656,
      "step": 500
    },
    {
      "epoch": 0.4244482173174873,
      "grad_norm": 37.0,
      "learning_rate": 9.94774038398841e-05,
      "loss": 2.976,
      "step": 1000
    },
    {
      "epoch": 0.6366723259762309,
      "grad_norm": 106.0,
      "learning_rate": 9.621403253948297e-05,
      "loss": 3.28,
      "step": 1500
    },
    {
      "epoch": 0.8488964346349746,
      "grad_norm": 176160768.0,
      "learning_rate": 9.014681135272878e-05,
      "loss": 2.048,
      "step": 2000
    },
    {
      "epoch": 1.061120543293718,
      "grad_norm": 421527552.0,
      "learning_rate": 8.164384443544097e-05,
      "loss": 2.048,
      "step": 2500
    },
    {
      "epoch": 1.2733446519524618,
      "grad_norm": 1333788672.0,
      "learning_rate": 7.122101498401308e-05,
      "loss": 2.048,
      "step": 3000
    },
    {
      "epoch": 1.4855687606112054,
      "grad_norm": 42205184.0,
      "learning_rate": 5.951068610477853e-05,
      "loss": 2.048,
      "step": 3500
    },
    {
      "epoch": 1.697792869269949,
      "grad_norm": 291504128.0,
      "learning_rate": 4.72233347349998e-05,
      "loss": 3.792,
      "step": 4000
    },
    {
      "epoch": 1.9100169779286928,
      "grad_norm": 20054016.0,
      "learning_rate": 3.510444632246376e-05,
      "loss": 2.048,
      "step": 4500
    },
    {
      "epoch": 2.122241086587436,
      "grad_norm": 128974848.0,
      "learning_rate": 2.3889285505009233e-05,
      "loss": 2.048,
      "step": 5000
    },
    {
      "epoch": 2.33446519524618,
      "grad_norm": 200278016.0,
      "learning_rate": 1.425828689644288e-05,
      "loss": 2.048,
      "step": 5500
    },
    {
      "epoch": 2.5466893039049237,
      "grad_norm": 511705088.0,
      "learning_rate": 6.795772462847716e-06,
      "loss": 2.048,
      "step": 6000
    },
    {
      "epoch": 2.7589134125636674,
      "grad_norm": 574619648.0,
      "learning_rate": 1.9545001458439126e-06,
      "loss": 2.048,
      "step": 6500
    },
    {
      "epoch": 2.9711375212224107,
      "grad_norm": 400556032.0,
      "learning_rate": 2.8194601989800507e-08,
      "loss": 2.048,
      "step": 7000
    },
    {
      "epoch": 3.0,
      "step": 7068,
      "total_flos": 9.210874701004407e+18,
      "train_loss": 2.3821448783248442,
      "train_runtime": 5996.2012,
      "train_samples_per_second": 75.446,
      "train_steps_per_second": 1.179
    }
  ],
  "logging_steps": 500,
  "max_steps": 7068,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 9.210874701004407e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}