last-checkpoint/trainer_state.json

{
  "best_metric": 0.04674109071493149,
  "best_model_checkpoint": "doc-topic-model_eval-00_train-03/checkpoint-5000",
  "epoch": 2.465483234714004,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2465483234714004,
      "grad_norm": 0.3815906345844269,
      "learning_rate": 1.9950690335305722e-05,
      "loss": 0.1667,
      "step": 500
    },
    {
      "epoch": 0.4930966469428008,
      "grad_norm": 0.36234650015830994,
      "learning_rate": 1.9901380670611442e-05,
      "loss": 0.0923,
      "step": 1000
    },
    {
      "epoch": 0.4930966469428008,
      "eval_accuracy": 0.9814660487265615,
      "eval_f1": 0.0,
      "eval_loss": 0.08645907044410706,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 11.8049,
      "eval_samples_per_second": 687.004,
      "eval_steps_per_second": 2.711,
      "step": 1000
    },
    {
      "epoch": 0.7396449704142012,
      "grad_norm": 0.39963558316230774,
      "learning_rate": 1.9852071005917162e-05,
      "loss": 0.084,
      "step": 1500
    },
    {
      "epoch": 0.9861932938856016,
      "grad_norm": 0.37207868695259094,
      "learning_rate": 1.980276134122288e-05,
      "loss": 0.0734,
      "step": 2000
    },
    {
      "epoch": 0.9861932938856016,
      "eval_accuracy": 0.9814660487265615,
      "eval_f1": 0.0,
      "eval_loss": 0.0668005496263504,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 11.8932,
      "eval_samples_per_second": 681.903,
      "eval_steps_per_second": 2.691,
      "step": 2000
    },
    {
      "epoch": 1.232741617357002,
      "grad_norm": 0.39202699065208435,
      "learning_rate": 1.9753451676528602e-05,
      "loss": 0.0653,
      "step": 2500
    },
    {
      "epoch": 1.4792899408284024,
      "grad_norm": 0.3334597647190094,
      "learning_rate": 1.9704142011834322e-05,
      "loss": 0.0606,
      "step": 3000
    },
    {
      "epoch": 1.4792899408284024,
      "eval_accuracy": 0.9823617784202843,
      "eval_f1": 0.12648276830209867,
      "eval_loss": 0.05524122714996338,
      "eval_precision": 0.7700854700854701,
      "eval_recall": 0.06889959470826643,
      "eval_runtime": 14.1698,
      "eval_samples_per_second": 572.346,
      "eval_steps_per_second": 2.258,
      "step": 3000
    },
    {
      "epoch": 1.725838264299803,
      "grad_norm": 0.3517291843891144,
      "learning_rate": 1.965483234714004e-05,
      "loss": 0.055,
      "step": 3500
    },
    {
      "epoch": 1.972386587771203,
      "grad_norm": 0.43956857919692993,
      "learning_rate": 1.9605522682445763e-05,
      "loss": 0.0532,
      "step": 4000
    },
    {
      "epoch": 1.972386587771203,
      "eval_accuracy": 0.9840511926527489,
      "eval_f1": 0.2943500344892456,
      "eval_loss": 0.0491451695561409,
      "eval_precision": 0.8177700348432055,
      "eval_recall": 0.1794754148505009,
      "eval_runtime": 14.1385,
      "eval_samples_per_second": 573.612,
      "eval_steps_per_second": 2.263,
      "step": 4000
    },
    {
      "epoch": 2.2189349112426036,
      "grad_norm": 0.33584991097450256,
      "learning_rate": 1.9556213017751483e-05,
      "loss": 0.0483,
      "step": 4500
    },
    {
      "epoch": 2.465483234714004,
      "grad_norm": 0.35681256651878357,
      "learning_rate": 1.95069033530572e-05,
      "loss": 0.0466,
      "step": 5000
    },
    {
      "epoch": 2.465483234714004,
      "eval_accuracy": 0.9850886517283898,
      "eval_f1": 0.43230993363189985,
      "eval_loss": 0.04674109071493149,
      "eval_precision": 0.7342375366568915,
      "eval_recall": 0.3063393744742678,
      "eval_runtime": 14.36,
      "eval_samples_per_second": 564.763,
      "eval_steps_per_second": 2.228,
      "step": 5000
    }
  ],
  "logging_steps": 500,
  "max_steps": 202800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 1000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 112700461753656.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}