trainer_state.json

{
  "best_global_step": 102,
  "best_metric": 0.6137072443962097,
  "best_model_checkpoint": "distilbert_rand_5_v1_cola/checkpoint-102",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 272,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4865912199020386,
      "learning_rate": 4.9e-05,
      "loss": 0.6177,
      "step": 34
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6142980456352234,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 0.4394,
      "eval_samples_per_second": 2373.446,
      "eval_steps_per_second": 11.378,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7455832958221436,
      "learning_rate": 4.8e-05,
      "loss": 0.5895,
      "step": 68
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6893576383590698,
      "eval_loss": 0.6215459704399109,
      "eval_matthews_correlation": 0.03026012901076368,
      "eval_runtime": 0.4741,
      "eval_samples_per_second": 2199.909,
      "eval_steps_per_second": 10.546,
      "step": 68
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.917654275894165,
      "learning_rate": 4.7e-05,
      "loss": 0.546,
      "step": 102
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6912751793861389,
      "eval_loss": 0.6137072443962097,
      "eval_matthews_correlation": 0.08689422736051329,
      "eval_runtime": 0.4335,
      "eval_samples_per_second": 2406.212,
      "eval_steps_per_second": 11.535,
      "step": 102
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.8158507347106934,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.4996,
      "step": 136
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.668264627456665,
      "eval_loss": 0.6761741638183594,
      "eval_matthews_correlation": 0.1278302795073983,
      "eval_runtime": 0.4402,
      "eval_samples_per_second": 2369.491,
      "eval_steps_per_second": 11.359,
      "step": 136
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.6593291759490967,
      "learning_rate": 4.5e-05,
      "loss": 0.442,
      "step": 170
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6548417806625366,
      "eval_loss": 0.7010028958320618,
      "eval_matthews_correlation": 0.14743668594213347,
      "eval_runtime": 0.4918,
      "eval_samples_per_second": 2120.648,
      "eval_steps_per_second": 10.166,
      "step": 170
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.5317959785461426,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.3915,
      "step": 204
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6347075700759888,
      "eval_loss": 0.7664260268211365,
      "eval_matthews_correlation": 0.05787465768343852,
      "eval_runtime": 0.4366,
      "eval_samples_per_second": 2389.053,
      "eval_steps_per_second": 11.453,
      "step": 204
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.171992778778076,
      "learning_rate": 4.3e-05,
      "loss": 0.3473,
      "step": 238
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6481304168701172,
      "eval_loss": 0.8267958760261536,
      "eval_matthews_correlation": 0.12688229904239956,
      "eval_runtime": 0.4335,
      "eval_samples_per_second": 2406.273,
      "eval_steps_per_second": 11.535,
      "step": 238
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.4829142093658447,
      "learning_rate": 4.2e-05,
      "loss": 0.3136,
      "step": 272
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6577181220054626,
      "eval_loss": 0.9088857769966125,
      "eval_matthews_correlation": 0.14812639440684075,
      "eval_runtime": 0.4583,
      "eval_samples_per_second": 2276.017,
      "eval_steps_per_second": 10.911,
      "step": 272
    },
    {
      "epoch": 8.0,
      "step": 272,
      "total_flos": 4530914903629824.0,
      "train_loss": 0.468404587577371,
      "train_runtime": 76.2563,
      "train_samples_per_second": 5606.748,
      "train_steps_per_second": 22.293
    }
  ],
  "logging_steps": 1,
  "max_steps": 1700,
  "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": 4530914903629824.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}