{
  "best_metric": 2.166426420211792,
  "best_model_checkpoint": "./tomrp/checkpoint-15",
  "epoch": 5.0,
  "eval_steps": 3,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.33,
      "learning_rate": 5e-06,
      "loss": 2.779,
      "step": 1
    },
    {
      "epoch": 0.33,
      "eval_loss": 2.6370952129364014,
      "eval_runtime": 2.2587,
      "eval_samples_per_second": 12.397,
      "eval_steps_per_second": 0.885,
      "step": 1
    },
    {
      "epoch": 0.67,
      "learning_rate": 1e-05,
      "loss": 2.8308,
      "step": 2
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.5e-05,
      "loss": 2.8055,
      "step": 3
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.6105611324310303,
      "eval_runtime": 2.2632,
      "eval_samples_per_second": 12.372,
      "eval_steps_per_second": 0.884,
      "step": 3
    },
    {
      "epoch": 1.33,
      "learning_rate": 2e-05,
      "loss": 2.7115,
      "step": 4
    },
    {
      "epoch": 1.67,
      "learning_rate": 2.5e-05,
      "loss": 2.8004,
      "step": 5
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 2.4321,
      "step": 6
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.45676589012146,
      "eval_runtime": 2.2658,
      "eval_samples_per_second": 12.358,
      "eval_steps_per_second": 0.883,
      "step": 6
    },
    {
      "epoch": 2.33,
      "learning_rate": 3.5e-05,
      "loss": 2.5158,
      "step": 7
    },
    {
      "epoch": 2.67,
      "learning_rate": 4e-05,
      "loss": 2.2594,
      "step": 8
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.5e-05,
      "loss": 2.0601,
      "step": 9
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.2703487873077393,
      "eval_runtime": 2.2646,
      "eval_samples_per_second": 12.364,
      "eval_steps_per_second": 0.883,
      "step": 9
    },
    {
      "epoch": 3.33,
      "learning_rate": 5e-05,
      "loss": 1.9706,
      "step": 10
    },
    {
      "epoch": 3.67,
      "learning_rate": 4.9692208514878444e-05,
      "loss": 1.7659,
      "step": 11
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.877641290737884e-05,
      "loss": 2.0104,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.1692585945129395,
      "eval_runtime": 2.2637,
      "eval_samples_per_second": 12.369,
      "eval_steps_per_second": 0.884,
      "step": 12
    },
    {
      "epoch": 4.33,
      "learning_rate": 4.72751631047092e-05,
      "loss": 1.5315,
      "step": 13
    },
    {
      "epoch": 4.67,
      "learning_rate": 4.522542485937369e-05,
      "loss": 1.4955,
      "step": 14
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.267766952966369e-05,
      "loss": 1.5786,
      "step": 15
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.166426420211792,
      "eval_runtime": 2.2642,
      "eval_samples_per_second": 12.367,
      "eval_steps_per_second": 0.883,
      "step": 15
    }
  ],
  "logging_steps": 1,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3,
  "total_flos": 3399415706419200.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}