trainer_state.json

{
  "best_global_step": 7976,
  "best_metric": 0.7853136155022947,
  "best_model_checkpoint": "./deberta_bio_result/checkpoint-7976",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 7976,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12537612838515547,
      "grad_norm": 7.210555553436279,
      "learning_rate": 1.874874623871615e-05,
      "loss": 0.4265,
      "step": 500
    },
    {
      "epoch": 0.25075225677031093,
      "grad_norm": 6.844402313232422,
      "learning_rate": 1.7494984954864596e-05,
      "loss": 0.3875,
      "step": 1000
    },
    {
      "epoch": 0.37612838515546637,
      "grad_norm": 3.695915460586548,
      "learning_rate": 1.624122367101304e-05,
      "loss": 0.3651,
      "step": 1500
    },
    {
      "epoch": 0.5015045135406219,
      "grad_norm": 10.602663040161133,
      "learning_rate": 1.4987462387161487e-05,
      "loss": 0.373,
      "step": 2000
    },
    {
      "epoch": 0.6268806419257773,
      "grad_norm": 2.402719259262085,
      "learning_rate": 1.3733701103309932e-05,
      "loss": 0.3744,
      "step": 2500
    },
    {
      "epoch": 0.7522567703109327,
      "grad_norm": 5.753551006317139,
      "learning_rate": 1.2479939819458375e-05,
      "loss": 0.3696,
      "step": 3000
    },
    {
      "epoch": 0.8776328986960883,
      "grad_norm": 4.795076847076416,
      "learning_rate": 1.1226178535606823e-05,
      "loss": 0.3445,
      "step": 3500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8630892678034102,
      "eval_f1": 0.7817309614231461,
      "eval_loss": 0.3573812246322632,
      "eval_runtime": 30.8925,
      "eval_samples_per_second": 516.371,
      "eval_steps_per_second": 32.273,
      "step": 3988
    },
    {
      "epoch": 1.0030090270812437,
      "grad_norm": 7.011142253875732,
      "learning_rate": 9.972417251755266e-06,
      "loss": 0.3543,
      "step": 4000
    },
    {
      "epoch": 1.1283851554663993,
      "grad_norm": 3.9173672199249268,
      "learning_rate": 8.718655967903712e-06,
      "loss": 0.3149,
      "step": 4500
    },
    {
      "epoch": 1.2537612838515546,
      "grad_norm": 3.875389814376831,
      "learning_rate": 7.464894684052158e-06,
      "loss": 0.321,
      "step": 5000
    },
    {
      "epoch": 1.3791374122367102,
      "grad_norm": 5.462708473205566,
      "learning_rate": 6.211133400200602e-06,
      "loss": 0.3176,
      "step": 5500
    },
    {
      "epoch": 1.5045135406218657,
      "grad_norm": 5.192739963531494,
      "learning_rate": 4.957372116349047e-06,
      "loss": 0.3111,
      "step": 6000
    },
    {
      "epoch": 1.629889669007021,
      "grad_norm": 23.06952476501465,
      "learning_rate": 3.703610832497493e-06,
      "loss": 0.3156,
      "step": 6500
    },
    {
      "epoch": 1.7552657973921764,
      "grad_norm": 5.229244232177734,
      "learning_rate": 2.449849548645938e-06,
      "loss": 0.3049,
      "step": 7000
    },
    {
      "epoch": 1.880641925777332,
      "grad_norm": 7.030540466308594,
      "learning_rate": 1.1960882647943831e-06,
      "loss": 0.3164,
      "step": 7500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8680416248746239,
      "eval_f1": 0.7853136155022947,
      "eval_loss": 0.36421605944633484,
      "eval_runtime": 31.0404,
      "eval_samples_per_second": 513.911,
      "eval_steps_per_second": 32.119,
      "step": 7976
    }
  ],
  "logging_steps": 500,
  "max_steps": 7976,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8393900443622400.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}