last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.003352329869259135,
  "eval_steps": 3,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006704659738518271,
      "grad_norm": 4.041480541229248,
      "learning_rate": 2e-05,
      "loss": 5.117,
      "step": 1
    },
    {
      "epoch": 0.0006704659738518271,
      "eval_loss": 5.064063549041748,
      "eval_runtime": 52.7515,
      "eval_samples_per_second": 11.905,
      "eval_steps_per_second": 5.952,
      "step": 1
    },
    {
      "epoch": 0.0013409319477036541,
      "grad_norm": 3.8326361179351807,
      "learning_rate": 4e-05,
      "loss": 5.0327,
      "step": 2
    },
    {
      "epoch": 0.002011397921555481,
      "grad_norm": 3.754926919937134,
      "learning_rate": 6e-05,
      "loss": 4.8776,
      "step": 3
    },
    {
      "epoch": 0.002011397921555481,
      "eval_loss": 5.00453519821167,
      "eval_runtime": 52.682,
      "eval_samples_per_second": 11.921,
      "eval_steps_per_second": 5.96,
      "step": 3
    },
    {
      "epoch": 0.0026818638954073082,
      "grad_norm": 5.529252529144287,
      "learning_rate": 8e-05,
      "loss": 5.7991,
      "step": 4
    },
    {
      "epoch": 0.003352329869259135,
      "grad_norm": 3.168138027191162,
      "learning_rate": 0.0001,
      "loss": 4.8356,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}