{
  "best_metric": 1.0670849084854126,
  "best_model_checkpoint": "outputs/checkpoint-170",
  "epoch": 4.874551971326165,
  "eval_steps": 500,
  "global_step": 170,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5734767025089605,
      "grad_norm": 1.373448371887207,
      "learning_rate": 2e-05,
      "loss": 2.8781,
      "step": 20
    },
    {
      "epoch": 0.974910394265233,
      "eval_loss": 2.158191204071045,
      "eval_runtime": 90.2612,
      "eval_samples_per_second": 12.364,
      "eval_steps_per_second": 1.551,
      "step": 34
    },
    {
      "epoch": 1.146953405017921,
      "grad_norm": 0.6855719685554504,
      "learning_rate": 4e-05,
      "loss": 2.3786,
      "step": 40
    },
    {
      "epoch": 1.7204301075268817,
      "grad_norm": 0.6112047433853149,
      "learning_rate": 6e-05,
      "loss": 1.8949,
      "step": 60
    },
    {
      "epoch": 1.978494623655914,
      "eval_loss": 1.6969424486160278,
      "eval_runtime": 35.3087,
      "eval_samples_per_second": 31.607,
      "eval_steps_per_second": 3.965,
      "step": 69
    },
    {
      "epoch": 2.293906810035842,
      "grad_norm": 0.7167624235153198,
      "learning_rate": 8e-05,
      "loss": 1.707,
      "step": 80
    },
    {
      "epoch": 2.867383512544803,
      "grad_norm": 1.0710257291793823,
      "learning_rate": 0.0001,
      "loss": 1.5109,
      "step": 100
    },
    {
      "epoch": 2.982078853046595,
      "eval_loss": 1.366994857788086,
      "eval_runtime": 35.3012,
      "eval_samples_per_second": 31.614,
      "eval_steps_per_second": 3.966,
      "step": 104
    },
    {
      "epoch": 3.4408602150537635,
      "grad_norm": 1.3240500688552856,
      "learning_rate": 8.117449009293668e-05,
      "loss": 1.3227,
      "step": 120
    },
    {
      "epoch": 3.985663082437276,
      "eval_loss": 1.1181360483169556,
      "eval_runtime": 35.2954,
      "eval_samples_per_second": 31.619,
      "eval_steps_per_second": 3.967,
      "step": 139
    },
    {
      "epoch": 4.014336917562724,
      "grad_norm": 1.4254589080810547,
      "learning_rate": 3.887395330218429e-05,
      "loss": 1.214,
      "step": 140
    },
    {
      "epoch": 4.587813620071684,
      "grad_norm": 1.6918621063232422,
      "learning_rate": 4.951556604879048e-06,
      "loss": 1.0894,
      "step": 160
    },
    {
      "epoch": 4.874551971326165,
      "eval_loss": 1.0670849084854126,
      "eval_runtime": 35.2971,
      "eval_samples_per_second": 31.617,
      "eval_steps_per_second": 3.966,
      "step": 170
    }
  ],
  "logging_steps": 20,
  "max_steps": 170,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 8953785954189312.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}