checkpoint-868/trainer_state.json

{
  "best_global_step": 868,
  "best_metric": 0.02894793264567852,
  "best_model_checkpoint": "coedit-gec-output/checkpoint-868",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 868,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gleu": 0.5903191424034809,
      "eval_loss": 0.05094386264681816,
      "eval_runtime": 274.5123,
      "eval_samples_per_second": 21.664,
      "eval_steps_per_second": 1.355,
      "step": 434
    },
    {
      "epoch": 1.1520737327188941,
      "grad_norm": 2.6467134952545166,
      "learning_rate": 1.707373271889401e-05,
      "loss": 4.5342,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_gleu": 0.6136322248136682,
      "eval_loss": 0.02894793264567852,
      "eval_runtime": 275.4106,
      "eval_samples_per_second": 21.593,
      "eval_steps_per_second": 1.351,
      "step": 868
    }
  ],
  "logging_steps": 500,
  "max_steps": 8680,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 4230351547269120.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}