trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.4467853009700775,
  "best_model_checkpoint": "tiny_bert_rand_10_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": 7.41895866394043,
      "learning_rate": 4.9e-05,
      "loss": 0.4231,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 0.4467853009700775,
      "eval_runtime": 0.2679,
      "eval_samples_per_second": 3254.856,
      "eval_steps_per_second": 14.931,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 13.120146751403809,
      "learning_rate": 4.8e-05,
      "loss": 0.239,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.786697247706422,
      "eval_loss": 0.5577822327613831,
      "eval_runtime": 0.2716,
      "eval_samples_per_second": 3210.829,
      "eval_steps_per_second": 14.729,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.774768352508545,
      "learning_rate": 4.7e-05,
      "loss": 0.1866,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7912844036697247,
      "eval_loss": 0.5245605111122131,
      "eval_runtime": 0.2693,
      "eval_samples_per_second": 3237.836,
      "eval_steps_per_second": 14.852,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.856670379638672,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.155,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7935779816513762,
      "eval_loss": 0.5888988375663757,
      "eval_runtime": 0.2688,
      "eval_samples_per_second": 3243.486,
      "eval_steps_per_second": 14.878,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 12.111465454101562,
      "learning_rate": 4.5e-05,
      "loss": 0.1281,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7993119266055045,
      "eval_loss": 0.6581941246986389,
      "eval_runtime": 0.2683,
      "eval_samples_per_second": 3249.607,
      "eval_steps_per_second": 14.906,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 9.845342636108398,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.1091,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.801605504587156,
      "eval_loss": 0.8329399228096008,
      "eval_runtime": 0.2643,
      "eval_samples_per_second": 3299.839,
      "eval_steps_per_second": 15.137,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 1.0596790573590528e+16,
      "train_loss": 0.20680861521248867,
      "train_runtime": 220.6024,
      "train_samples_per_second": 15264.792,
      "train_steps_per_second": 59.836
    }
  ],
  "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
}