trainer_state.json

{
  "best_metric": 0.2739827632904053,
  "best_model_checkpoint": "spa_enpt_mBERT/checkpoint-4480",
  "epoch": 6.0,
  "global_step": 5376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.56,
      "learning_rate": 1.988839285714286e-05,
      "loss": 1.0144,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.898357595376641,
      "eval_f1": 0.7096795907377491,
      "eval_loss": 0.39573460817337036,
      "eval_precision": 0.7225191885964912,
      "eval_recall": 0.6972883597883598,
      "eval_runtime": 5.9371,
      "eval_samples_per_second": 278.587,
      "eval_steps_per_second": 17.517,
      "step": 896
    },
    {
      "epoch": 1.12,
      "learning_rate": 1.9776785714285716e-05,
      "loss": 0.4668,
      "step": 1000
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.9665178571428575e-05,
      "loss": 0.3426,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9187535995869164,
      "eval_f1": 0.7710635473701675,
      "eval_loss": 0.3058053255081177,
      "eval_precision": 0.7737595737595737,
      "eval_recall": 0.7683862433862434,
      "eval_runtime": 5.8871,
      "eval_samples_per_second": 280.952,
      "eval_steps_per_second": 17.666,
      "step": 1792
    },
    {
      "epoch": 2.23,
      "learning_rate": 1.955357142857143e-05,
      "loss": 0.2917,
      "step": 2000
    },
    {
      "epoch": 2.79,
      "learning_rate": 1.944196428571429e-05,
      "loss": 0.2536,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9238377057970727,
      "eval_f1": 0.7864055223436931,
      "eval_loss": 0.28532469272613525,
      "eval_precision": 0.7854456686679422,
      "eval_recall": 0.7873677248677249,
      "eval_runtime": 5.888,
      "eval_samples_per_second": 280.91,
      "eval_steps_per_second": 17.663,
      "step": 2688
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.9330357142857144e-05,
      "loss": 0.2101,
      "step": 3000
    },
    {
      "epoch": 3.91,
      "learning_rate": 1.9218750000000003e-05,
      "loss": 0.19,
      "step": 3500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9296169046531487,
      "eval_f1": 0.8055234382742558,
      "eval_loss": 0.2802596986293793,
      "eval_precision": 0.8046993597782325,
      "eval_recall": 0.8063492063492064,
      "eval_runtime": 5.9895,
      "eval_samples_per_second": 276.151,
      "eval_steps_per_second": 17.364,
      "step": 3584
    },
    {
      "epoch": 4.46,
      "learning_rate": 1.910714285714286e-05,
      "loss": 0.1669,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9341449367465692,
      "eval_f1": 0.8190463567381291,
      "eval_loss": 0.2739827632904053,
      "eval_precision": 0.8207052261106316,
      "eval_recall": 0.8173941798941798,
      "eval_runtime": 5.9186,
      "eval_samples_per_second": 279.459,
      "eval_steps_per_second": 17.572,
      "step": 4480
    },
    {
      "epoch": 5.02,
      "learning_rate": 1.8995535714285717e-05,
      "loss": 0.1547,
      "step": 4500
    },
    {
      "epoch": 5.58,
      "learning_rate": 1.8883928571428573e-05,
      "loss": 0.1292,
      "step": 5000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9308680714158044,
      "eval_f1": 0.8097851239669421,
      "eval_loss": 0.29914459586143494,
      "eval_precision": 0.809517514871117,
      "eval_recall": 0.81005291005291,
      "eval_runtime": 5.9033,
      "eval_samples_per_second": 280.184,
      "eval_steps_per_second": 17.617,
      "step": 5376
    }
  ],
  "max_steps": 89600,
  "num_train_epochs": 100,
  "total_flos": 3919265715030336.0,
  "trial_name": null,
  "trial_params": null
}