trainer_state.json

{
  "best_metric": 0.4466005265712738,
  "best_model_checkpoint": "bert_tiny_lda_100_v1_sst2/checkpoint-264",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1584,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 10.585709571838379,
      "learning_rate": 4.9e-05,
      "loss": 0.4292,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.805045871559633,
      "eval_loss": 0.4466005265712738,
      "eval_runtime": 0.2781,
      "eval_samples_per_second": 3135.372,
      "eval_steps_per_second": 14.382,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.0157647132873535,
      "learning_rate": 4.8e-05,
      "loss": 0.2589,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8153669724770642,
      "eval_loss": 0.4592009484767914,
      "eval_runtime": 0.2832,
      "eval_samples_per_second": 3079.044,
      "eval_steps_per_second": 14.124,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.999380111694336,
      "learning_rate": 4.7e-05,
      "loss": 0.2007,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8130733944954128,
      "eval_loss": 0.4953426420688629,
      "eval_runtime": 0.2791,
      "eval_samples_per_second": 3124.457,
      "eval_steps_per_second": 14.332,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.423249244689941,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1652,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8302752293577982,
      "eval_loss": 0.5380478501319885,
      "eval_runtime": 0.2813,
      "eval_samples_per_second": 3099.753,
      "eval_steps_per_second": 14.219,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.089705467224121,
      "learning_rate": 4.5e-05,
      "loss": 0.137,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8096330275229358,
      "eval_loss": 0.5538350939750671,
      "eval_runtime": 0.2751,
      "eval_samples_per_second": 3169.498,
      "eval_steps_per_second": 14.539,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.120477199554443,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1185,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8096330275229358,
      "eval_loss": 0.6460174918174744,
      "eval_runtime": 0.2755,
      "eval_samples_per_second": 3164.776,
      "eval_steps_per_second": 14.517,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 1.0596790573590528e+16,
      "train_loss": 0.21827046196870128,
      "train_runtime": 225.56,
      "train_samples_per_second": 14929.289,
      "train_steps_per_second": 58.521
    }
  ],
  "logging_steps": 1,
  "max_steps": 13200,
  "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": 1.0596790573590528e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}