{
  "best_metric": 1.021238923072815,
  "best_model_checkpoint": "outputs/checkpoint-174",
  "epoch": 5.849462365591398,
  "eval_steps": 500,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5734767025089605,
      "grad_norm": 1.5929073095321655,
      "learning_rate": 4e-05,
      "loss": 2.8166,
      "step": 20
    },
    {
      "epoch": 0.974910394265233,
      "eval_loss": 1.9175890684127808,
      "eval_runtime": 11.59,
      "eval_samples_per_second": 32.097,
      "eval_steps_per_second": 4.055,
      "step": 34
    },
    {
      "epoch": 1.146953405017921,
      "grad_norm": 0.6563653945922852,
      "learning_rate": 8e-05,
      "loss": 2.0599,
      "step": 40
    },
    {
      "epoch": 1.7204301075268817,
      "grad_norm": 0.7280052304267883,
      "learning_rate": 0.00012,
      "loss": 1.7423,
      "step": 60
    },
    {
      "epoch": 1.978494623655914,
      "eval_loss": 1.5290530920028687,
      "eval_runtime": 11.5703,
      "eval_samples_per_second": 32.151,
      "eval_steps_per_second": 4.062,
      "step": 69
    },
    {
      "epoch": 2.293906810035842,
      "grad_norm": 1.1366034746170044,
      "learning_rate": 0.00016,
      "loss": 1.5151,
      "step": 80
    },
    {
      "epoch": 2.867383512544803,
      "grad_norm": 1.3949623107910156,
      "learning_rate": 0.0002,
      "loss": 1.2694,
      "step": 100
    },
    {
      "epoch": 2.982078853046595,
      "eval_loss": 1.2222951650619507,
      "eval_runtime": 11.5734,
      "eval_samples_per_second": 32.143,
      "eval_steps_per_second": 4.061,
      "step": 104
    },
    {
      "epoch": 3.4408602150537635,
      "grad_norm": 1.5031639337539673,
      "learning_rate": 0.00018229838658936564,
      "loss": 1.0282,
      "step": 120
    },
    {
      "epoch": 3.985663082437276,
      "eval_loss": 1.0663368701934814,
      "eval_runtime": 11.5671,
      "eval_samples_per_second": 32.16,
      "eval_steps_per_second": 4.063,
      "step": 139
    },
    {
      "epoch": 4.014336917562724,
      "grad_norm": 1.577106237411499,
      "learning_rate": 0.00013546048870425356,
      "loss": 0.9281,
      "step": 140
    },
    {
      "epoch": 4.587813620071684,
      "grad_norm": 1.994376301765442,
      "learning_rate": 7.606843357124426e-05,
      "loss": 0.7084,
      "step": 160
    },
    {
      "epoch": 4.989247311827957,
      "eval_loss": 1.021238923072815,
      "eval_runtime": 11.604,
      "eval_samples_per_second": 32.058,
      "eval_steps_per_second": 4.05,
      "step": 174
    },
    {
      "epoch": 5.161290322580645,
      "grad_norm": 1.5775189399719238,
      "learning_rate": 2.514892518288988e-05,
      "loss": 0.6572,
      "step": 180
    },
    {
      "epoch": 5.734767025089606,
      "grad_norm": 1.6861282587051392,
      "learning_rate": 7.291125901946027e-07,
      "loss": 0.5642,
      "step": 200
    },
    {
      "epoch": 5.849462365591398,
      "eval_loss": 1.0425939559936523,
      "eval_runtime": 11.584,
      "eval_samples_per_second": 32.113,
      "eval_steps_per_second": 4.057,
      "step": 204
    }
  ],
  "logging_steps": 20,
  "max_steps": 204,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 1.073987375468544e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}