trainer_state.json

{
  "best_metric": 0.696027934551239,
  "best_model_checkpoint": "distilbert_lda_20_v1_book_wnli/checkpoint-6",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.6686683297157288,
      "learning_rate": 4.9e-05,
      "loss": 0.704,
      "step": 3
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.38028169014084506,
      "eval_loss": 0.7027398347854614,
      "eval_runtime": 0.0404,
      "eval_samples_per_second": 1759.262,
      "eval_steps_per_second": 24.778,
      "step": 3
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3325134813785553,
      "learning_rate": 4.8e-05,
      "loss": 0.6964,
      "step": 6
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5070422535211268,
      "eval_loss": 0.696027934551239,
      "eval_runtime": 0.0695,
      "eval_samples_per_second": 1020.97,
      "eval_steps_per_second": 14.38,
      "step": 6
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.32530850172042847,
      "learning_rate": 4.7e-05,
      "loss": 0.6928,
      "step": 9
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4084507042253521,
      "eval_loss": 0.700704038143158,
      "eval_runtime": 0.042,
      "eval_samples_per_second": 1690.474,
      "eval_steps_per_second": 23.809,
      "step": 9
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5654997825622559,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6942,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4788732394366197,
      "eval_loss": 0.7017493844032288,
      "eval_runtime": 0.0386,
      "eval_samples_per_second": 1841.495,
      "eval_steps_per_second": 25.937,
      "step": 12
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5982861518859863,
      "learning_rate": 4.5e-05,
      "loss": 0.6918,
      "step": 15
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3380281690140845,
      "eval_loss": 0.7092869281768799,
      "eval_runtime": 0.0481,
      "eval_samples_per_second": 1475.997,
      "eval_steps_per_second": 20.789,
      "step": 15
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6909031271934509,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6913,
      "step": 18
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.36619718309859156,
      "eval_loss": 0.7186398506164551,
      "eval_runtime": 0.0391,
      "eval_samples_per_second": 1815.949,
      "eval_steps_per_second": 25.577,
      "step": 18
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.3236004412174225,
      "learning_rate": 4.3e-05,
      "loss": 0.6914,
      "step": 21
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.39436619718309857,
      "eval_loss": 0.7135233879089355,
      "eval_runtime": 0.0469,
      "eval_samples_per_second": 1514.705,
      "eval_steps_per_second": 21.334,
      "step": 21
    },
    {
      "epoch": 7.0,
      "step": 21,
      "total_flos": 294408793512960.0,
      "train_loss": 0.6945538974943615,
      "train_runtime": 17.3638,
      "train_samples_per_second": 1828.517,
      "train_steps_per_second": 8.639
    }
  ],
  "logging_steps": 1,
  "max_steps": 150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 294408793512960.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}