checkpoint-25/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.7037037037037037,
  "eval_steps": 10,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 119.2230224609375,
      "learning_rate": 6e-06,
      "loss": 10.468,
      "step": 10
    },
    {
      "epoch": 1.4814814814814814,
      "eval_loss": 6.026649475097656,
      "eval_runtime": 9.5208,
      "eval_samples_per_second": 1.26,
      "eval_steps_per_second": 0.63,
      "step": 10
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 86.91220092773438,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 3.5893,
      "step": 20
    },
    {
      "epoch": 2.962962962962963,
      "eval_loss": 1.231614589691162,
      "eval_runtime": 9.5301,
      "eval_samples_per_second": 1.259,
      "eval_steps_per_second": 0.63,
      "step": 20
    }
  ],
  "logging_steps": 10,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2058382144087104.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}