trainer_state.json

{
  "best_metric": 0.7339349985122681,
  "best_model_checkpoint": "bert_base_lda_5_v1_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": 1.3261046409606934,
      "learning_rate": 4.9e-05,
      "loss": 0.7595,
      "step": 3
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.38028169014084506,
      "eval_loss": 0.7546765208244324,
      "eval_runtime": 0.0536,
      "eval_samples_per_second": 1325.604,
      "eval_steps_per_second": 18.67,
      "step": 3
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0250415802001953,
      "learning_rate": 4.8e-05,
      "loss": 0.7035,
      "step": 6
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.38028169014084506,
      "eval_loss": 0.7339349985122681,
      "eval_runtime": 0.0533,
      "eval_samples_per_second": 1332.156,
      "eval_steps_per_second": 18.763,
      "step": 6
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.9258949756622314,
      "learning_rate": 4.7e-05,
      "loss": 0.6886,
      "step": 9
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2112676056338028,
      "eval_loss": 0.760481059551239,
      "eval_runtime": 0.0557,
      "eval_samples_per_second": 1274.684,
      "eval_steps_per_second": 17.953,
      "step": 9
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.567131519317627,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.6933,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.22535211267605634,
      "eval_loss": 0.7813875675201416,
      "eval_runtime": 0.0531,
      "eval_samples_per_second": 1336.887,
      "eval_steps_per_second": 18.829,
      "step": 12
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.0627994537353516,
      "learning_rate": 4.5e-05,
      "loss": 0.6894,
      "step": 15
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.22535211267605634,
      "eval_loss": 0.814260721206665,
      "eval_runtime": 0.0562,
      "eval_samples_per_second": 1264.203,
      "eval_steps_per_second": 17.806,
      "step": 15
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.213302493095398,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.6803,
      "step": 18
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.1267605633802817,
      "eval_loss": 0.8257593512535095,
      "eval_runtime": 0.0534,
      "eval_samples_per_second": 1330.175,
      "eval_steps_per_second": 18.735,
      "step": 18
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.9398391842842102,
      "learning_rate": 4.3e-05,
      "loss": 0.6799,
      "step": 21
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.16901408450704225,
      "eval_loss": 0.854671061038971,
      "eval_runtime": 0.0542,
      "eval_samples_per_second": 1310.841,
      "eval_steps_per_second": 18.463,
      "step": 21
    },
    {
      "epoch": 7.0,
      "step": 21,
      "total_flos": 584764320537600.0,
      "train_loss": 0.6992167291187105,
      "train_runtime": 26.7084,
      "train_samples_per_second": 1188.763,
      "train_steps_per_second": 5.616
    }
  ],
  "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": 584764320537600.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}