checkpoint-65/trainer_state.json

{
  "best_metric": 0.42300194931773877,
  "best_model_checkpoint": "brand-safety-model\\checkpoint-65",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 65,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.6331872940063477,
      "learning_rate": 2.7692307692307694e-05,
      "loss": 3.319,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.42300194931773877,
      "eval_loss": 2.7462074756622314,
      "eval_runtime": 1.5522,
      "eval_samples_per_second": 330.495,
      "eval_steps_per_second": 10.952,
      "step": 65
    }
  ],
  "logging_steps": 50,
  "max_steps": 650,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 81574836915456.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}