checkpoint-88200/trainer_state.json

{
  "best_metric": 0.5593105349247343,
  "best_model_checkpoint": "model/checkpoint-52920",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 88200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 50.548851013183594,
      "learning_rate": 4.5246352478135424e-06,
      "loss": 1.5427,
      "step": 17640
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5058956916099773,
      "eval_f1_macro": 0.5205801166115408,
      "eval_f1_weighted": 0.49617134581102434,
      "eval_loss": 1.5204957723617554,
      "eval_precision_macro": 0.5811975776248032,
      "eval_precision_weighted": 0.5725457955618433,
      "eval_recall_macro": 0.5336316110063488,
      "eval_recall_weighted": 0.5058956916099773,
      "eval_runtime": 90.4124,
      "eval_samples_per_second": 48.776,
      "eval_steps_per_second": 24.388,
      "step": 17640
    },
    {
      "epoch": 2.0,
      "grad_norm": 34.12895202636719,
      "learning_rate": 3.2750841586103932e-06,
      "loss": 1.3601,
      "step": 35280
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5390022675736962,
      "eval_f1_macro": 0.5526684923116755,
      "eval_f1_weighted": 0.5363903680411256,
      "eval_loss": 1.4914500713348389,
      "eval_precision_macro": 0.5690443791451115,
      "eval_precision_weighted": 0.5437222363835312,
      "eval_recall_macro": 0.5477102309891048,
      "eval_recall_weighted": 0.5390022675736962,
      "eval_runtime": 92.1119,
      "eval_samples_per_second": 47.877,
      "eval_steps_per_second": 23.938,
      "step": 35280
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.428272247314453,
      "learning_rate": 1.7291524533131918e-06,
      "loss": 1.2854,
      "step": 52920
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.546031746031746,
      "eval_f1_macro": 0.5593105349247343,
      "eval_f1_weighted": 0.5404116337093552,
      "eval_loss": 1.6935405731201172,
      "eval_precision_macro": 0.563042729738896,
      "eval_precision_weighted": 0.5453144110800754,
      "eval_recall_macro": 0.5628661101651561,
      "eval_recall_weighted": 0.546031746031746,
      "eval_runtime": 92.7796,
      "eval_samples_per_second": 47.532,
      "eval_steps_per_second": 23.766,
      "step": 52920
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.4480783939361572,
      "learning_rate": 4.779813469311145e-07,
      "loss": 1.2134,
      "step": 70560
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5424036281179139,
      "eval_f1_macro": 0.556857581435203,
      "eval_f1_weighted": 0.540084379855431,
      "eval_loss": 1.850308895111084,
      "eval_precision_macro": 0.5534590158556113,
      "eval_precision_weighted": 0.5403231343551113,
      "eval_recall_macro": 0.5627436062490211,
      "eval_recall_weighted": 0.5424036281179139,
      "eval_runtime": 90.1464,
      "eval_samples_per_second": 48.92,
      "eval_steps_per_second": 24.46,
      "step": 70560
    },
    {
      "epoch": 5.0,
      "grad_norm": 97.99571228027344,
      "learning_rate": 0.0,
      "loss": 1.1551,
      "step": 88200
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5380952380952381,
      "eval_f1_macro": 0.5523656911346606,
      "eval_f1_weighted": 0.535792504428007,
      "eval_loss": 1.9902939796447754,
      "eval_precision_macro": 0.5460891197767054,
      "eval_precision_weighted": 0.5351070225287136,
      "eval_recall_macro": 0.5601702556233763,
      "eval_recall_weighted": 0.5380952380952381,
      "eval_runtime": 89.9788,
      "eval_samples_per_second": 49.012,
      "eval_steps_per_second": 24.506,
      "step": 88200
    }
  ],
  "logging_steps": 500,
  "max_steps": 88200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 2
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.21982903146496e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}