{
  "best_metric": 0.2845354974269867,
  "best_model_checkpoint": "outputs/checkpoint-101",
  "epoch": 7.724137931034483,
  "eval_steps": 500,
  "global_step": 112,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.69,
      "learning_rate": 0.00018219165847995,
      "loss": 1.693,
      "step": 10
    },
    {
      "epoch": 0.97,
      "eval_loss": 0.8468072414398193,
      "eval_runtime": 47.2393,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.042,
      "step": 14
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001643297311779941,
      "loss": 0.8001,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5586440563201904,
      "eval_runtime": 46.728,
      "eval_samples_per_second": 0.321,
      "eval_steps_per_second": 0.043,
      "step": 29
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.00014646780387603822,
      "loss": 0.6269,
      "step": 30
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.00012860587657408234,
      "loss": 0.3671,
      "step": 40
    },
    {
      "epoch": 2.97,
      "eval_loss": 0.33814382553100586,
      "eval_runtime": 48.0352,
      "eval_samples_per_second": 0.312,
      "eval_steps_per_second": 0.042,
      "step": 43
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.00011074394927212647,
      "loss": 0.2759,
      "step": 50
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.31169143319129944,
      "eval_runtime": 47.7275,
      "eval_samples_per_second": 0.314,
      "eval_steps_per_second": 0.042,
      "step": 58
    },
    {
      "epoch": 4.14,
      "learning_rate": 9.288202197017058e-05,
      "loss": 0.2149,
      "step": 60
    },
    {
      "epoch": 4.83,
      "learning_rate": 7.50200946682147e-05,
      "loss": 0.164,
      "step": 70
    },
    {
      "epoch": 4.97,
      "eval_loss": 0.29888349771499634,
      "eval_runtime": 47.9342,
      "eval_samples_per_second": 0.313,
      "eval_steps_per_second": 0.042,
      "step": 72
    },
    {
      "epoch": 5.52,
      "learning_rate": 5.7158167366258816e-05,
      "loss": 0.1221,
      "step": 80
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.2934713065624237,
      "eval_runtime": 48.2143,
      "eval_samples_per_second": 0.311,
      "eval_steps_per_second": 0.041,
      "step": 87
    },
    {
      "epoch": 6.21,
      "learning_rate": 3.929624006430294e-05,
      "loss": 0.1183,
      "step": 90
    },
    {
      "epoch": 6.9,
      "learning_rate": 2.143431276234706e-05,
      "loss": 0.0866,
      "step": 100
    },
    {
      "epoch": 6.97,
      "eval_loss": 0.2845354974269867,
      "eval_runtime": 46.7151,
      "eval_samples_per_second": 0.321,
      "eval_steps_per_second": 0.043,
      "step": 101
    },
    {
      "epoch": 7.59,
      "learning_rate": 3.572385460391176e-06,
      "loss": 0.0837,
      "step": 110
    },
    {
      "epoch": 7.72,
      "eval_loss": 0.2850027084350586,
      "eval_runtime": 47.2985,
      "eval_samples_per_second": 0.317,
      "eval_steps_per_second": 0.042,
      "step": 112
    }
  ],
  "logging_steps": 10,
  "max_steps": 112,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 2.546768317169664e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}