trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.43765541911125183,
  "best_model_checkpoint": "bert_base_rand_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": 8.260040283203125,
      "learning_rate": 4.9e-05,
      "loss": 0.3841,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8211009174311926,
      "eval_loss": 0.43765541911125183,
      "eval_runtime": 0.5619,
      "eval_samples_per_second": 1551.969,
      "eval_steps_per_second": 7.119,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 12.480940818786621,
      "learning_rate": 4.8e-05,
      "loss": 0.2216,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8119266055045872,
      "eval_loss": 0.5570793151855469,
      "eval_runtime": 0.5583,
      "eval_samples_per_second": 1561.835,
      "eval_steps_per_second": 7.164,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.054980754852295,
      "learning_rate": 4.7e-05,
      "loss": 0.1627,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8153669724770642,
      "eval_loss": 0.5257858633995056,
      "eval_runtime": 0.5608,
      "eval_samples_per_second": 1555.017,
      "eval_steps_per_second": 7.133,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.704627513885498,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1243,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8038990825688074,
      "eval_loss": 0.5841220617294312,
      "eval_runtime": 0.563,
      "eval_samples_per_second": 1548.809,
      "eval_steps_per_second": 7.105,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.184584140777588,
      "learning_rate": 4.5e-05,
      "loss": 0.0986,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.805045871559633,
      "eval_loss": 0.5857242941856384,
      "eval_runtime": 0.5649,
      "eval_samples_per_second": 1543.633,
      "eval_steps_per_second": 7.081,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 5.928060531616211,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0757,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8073394495412844,
      "eval_loss": 0.8638249635696411,
      "eval_runtime": 0.5632,
      "eval_samples_per_second": 1548.171,
      "eval_steps_per_second": 7.102,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 5.316079940232192e+16,
      "train_loss": 0.17779791355133057,
      "train_runtime": 662.482,
      "train_samples_per_second": 5083.082,
      "train_steps_per_second": 19.925
    }
  ],
  "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": 5.316079940232192e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}