trainer_state.json

{
  "best_metric": 0.9008125554721984,
  "best_model_checkpoint": "rubert_classification/checkpoint-3500",
  "epoch": 1.674641148325359,
  "eval_steps": 500,
  "global_step": 3500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24,
      "eval_accuracy": 0.7823053307284399,
      "eval_f1": 0.7752692727748154,
      "eval_loss": 0.6053181886672974,
      "eval_precision": 0.7849961270931392,
      "eval_recall": 0.7823053307284399,
      "eval_runtime": 38.9969,
      "eval_samples_per_second": 183.758,
      "eval_steps_per_second": 11.488,
      "step": 500
    },
    {
      "epoch": 0.29,
      "grad_norm": 22.682289123535156,
      "learning_rate": 4.975490196078432e-05,
      "loss": 0.9433,
      "step": 600
    },
    {
      "epoch": 0.48,
      "eval_accuracy": 0.8368685459112476,
      "eval_f1": 0.8330214371969107,
      "eval_loss": 0.46620890498161316,
      "eval_precision": 0.840415327990016,
      "eval_recall": 0.8368685459112476,
      "eval_runtime": 36.1817,
      "eval_samples_per_second": 198.056,
      "eval_steps_per_second": 12.382,
      "step": 1000
    },
    {
      "epoch": 0.57,
      "grad_norm": 12.139711380004883,
      "learning_rate": 4.82843137254902e-05,
      "loss": 0.5182,
      "step": 1200
    },
    {
      "epoch": 0.72,
      "eval_accuracy": 0.8428691041027072,
      "eval_f1": 0.8319455570081821,
      "eval_loss": 0.4372938573360443,
      "eval_precision": 0.8395761804457677,
      "eval_recall": 0.8428691041027072,
      "eval_runtime": 35.6356,
      "eval_samples_per_second": 201.091,
      "eval_steps_per_second": 12.572,
      "step": 1500
    },
    {
      "epoch": 0.86,
      "grad_norm": 37.54426574707031,
      "learning_rate": 4.681372549019608e-05,
      "loss": 0.4508,
      "step": 1800
    },
    {
      "epoch": 0.96,
      "eval_accuracy": 0.8644990231649455,
      "eval_f1": 0.8633019943338417,
      "eval_loss": 0.39909306168556213,
      "eval_precision": 0.8707480255460968,
      "eval_recall": 0.8644990231649455,
      "eval_runtime": 35.6445,
      "eval_samples_per_second": 201.041,
      "eval_steps_per_second": 12.569,
      "step": 2000
    },
    {
      "epoch": 1.15,
      "grad_norm": 3.496920108795166,
      "learning_rate": 4.5343137254901966e-05,
      "loss": 0.3875,
      "step": 2400
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.8898967345799609,
      "eval_f1": 0.8875919037418019,
      "eval_loss": 0.33941856026649475,
      "eval_precision": 0.8906320921346105,
      "eval_recall": 0.8898967345799609,
      "eval_runtime": 35.5422,
      "eval_samples_per_second": 201.619,
      "eval_steps_per_second": 12.605,
      "step": 2500
    },
    {
      "epoch": 1.44,
      "grad_norm": 16.364843368530273,
      "learning_rate": 4.387254901960784e-05,
      "loss": 0.3401,
      "step": 3000
    },
    {
      "epoch": 1.44,
      "eval_accuracy": 0.8869662294166899,
      "eval_f1": 0.8851872123624062,
      "eval_loss": 0.3447243273258209,
      "eval_precision": 0.8869707495957468,
      "eval_recall": 0.8869662294166899,
      "eval_runtime": 35.7225,
      "eval_samples_per_second": 200.602,
      "eval_steps_per_second": 12.541,
      "step": 3000
    },
    {
      "epoch": 1.67,
      "eval_accuracy": 0.9016187552330449,
      "eval_f1": 0.9008125554721984,
      "eval_loss": 0.325568288564682,
      "eval_precision": 0.9021842242763984,
      "eval_recall": 0.9016187552330449,
      "eval_runtime": 35.7388,
      "eval_samples_per_second": 200.51,
      "eval_steps_per_second": 12.535,
      "step": 3500
    }
  ],
  "logging_steps": 600,
  "max_steps": 20900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1640277252096000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}