trainer_state.json

{
  "best_metric": 0.6869498491287231,
  "best_model_checkpoint": "hBERTv1_new_pretrain_48_wnli/checkpoint-15",
  "epoch": 8.0,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 0.8579,
      "step": 5
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.8451804518699646,
      "eval_runtime": 0.1335,
      "eval_samples_per_second": 531.982,
      "eval_steps_per_second": 7.493,
      "step": 5
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.761,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.7104974985122681,
      "eval_runtime": 0.1323,
      "eval_samples_per_second": 536.677,
      "eval_steps_per_second": 7.559,
      "step": 10
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.76e-05,
      "loss": 0.7197,
      "step": 15
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6869498491287231,
      "eval_runtime": 0.1327,
      "eval_samples_per_second": 535.054,
      "eval_steps_per_second": 7.536,
      "step": 15
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.7104,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.6997689008712769,
      "eval_runtime": 0.132,
      "eval_samples_per_second": 537.994,
      "eval_steps_per_second": 7.577,
      "step": 20
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.6e-05,
      "loss": 0.6994,
      "step": 25
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6872798204421997,
      "eval_runtime": 0.1318,
      "eval_samples_per_second": 538.619,
      "eval_steps_per_second": 7.586,
      "step": 25
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.52e-05,
      "loss": 0.7049,
      "step": 30
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.43661971830985913,
      "eval_loss": 0.6955875158309937,
      "eval_runtime": 0.1319,
      "eval_samples_per_second": 538.452,
      "eval_steps_per_second": 7.584,
      "step": 30
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.44e-05,
      "loss": 0.6898,
      "step": 35
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6905810236930847,
      "eval_runtime": 0.1321,
      "eval_samples_per_second": 537.548,
      "eval_steps_per_second": 7.571,
      "step": 35
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.7002,
      "step": 40
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.688215434551239,
      "eval_runtime": 0.1322,
      "eval_samples_per_second": 537.113,
      "eval_steps_per_second": 7.565,
      "step": 40
    },
    {
      "epoch": 8.0,
      "step": 40,
      "total_flos": 752331637391360.0,
      "train_loss": 0.7304040133953095,
      "train_runtime": 55.0623,
      "train_samples_per_second": 576.62,
      "train_steps_per_second": 4.54
    }
  ],
  "max_steps": 250,
  "num_train_epochs": 50,
  "total_flos": 752331637391360.0,
  "trial_name": null,
  "trial_params": null
}