last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.009957679860592481,
  "eval_steps": 5,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0019915359721184964,
      "grad_norm": 1.2896974086761475,
      "learning_rate": 1e-05,
      "loss": 36.1849,
      "step": 1
    },
    {
      "epoch": 0.0019915359721184964,
      "eval_loss": 2.1876697540283203,
      "eval_runtime": 208.3799,
      "eval_samples_per_second": 4.06,
      "eval_steps_per_second": 2.03,
      "step": 1
    },
    {
      "epoch": 0.003983071944236993,
      "grad_norm": 1.1197001934051514,
      "learning_rate": 2e-05,
      "loss": 32.9364,
      "step": 2
    },
    {
      "epoch": 0.00597460791635549,
      "grad_norm": 1.066008448600769,
      "learning_rate": 3e-05,
      "loss": 36.2816,
      "step": 3
    },
    {
      "epoch": 0.007966143888473986,
      "grad_norm": 1.1513549089431763,
      "learning_rate": 4e-05,
      "loss": 34.4152,
      "step": 4
    },
    {
      "epoch": 0.009957679860592481,
      "grad_norm": 1.076801061630249,
      "learning_rate": 5e-05,
      "loss": 35.6192,
      "step": 5
    },
    {
      "epoch": 0.009957679860592481,
      "eval_loss": 2.1837847232818604,
      "eval_runtime": 58.8468,
      "eval_samples_per_second": 14.376,
      "eval_steps_per_second": 7.188,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "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": 9815295346606080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}