trainer_state.json

{
  "best_metric": 0.7371178269386292,
  "best_model_checkpoint": "hBERTv2_new_pretrain_rte/checkpoint-20",
  "epoch": 6.0,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0.00049,
      "loss": 9.1657,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.7371178269386292,
      "eval_runtime": 0.4926,
      "eval_samples_per_second": 562.316,
      "eval_steps_per_second": 6.09,
      "step": 20
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00048,
      "loss": 9.3188,
      "step": 40
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.7859036326408386,
      "eval_runtime": 0.4896,
      "eval_samples_per_second": 565.793,
      "eval_steps_per_second": 6.128,
      "step": 40
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00047,
      "loss": 9.3339,
      "step": 60
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.9236378073692322,
      "eval_runtime": 0.4893,
      "eval_samples_per_second": 566.073,
      "eval_steps_per_second": 6.131,
      "step": 60
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00046,
      "loss": 9.7488,
      "step": 80
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.914161205291748,
      "eval_runtime": 0.4899,
      "eval_samples_per_second": 565.365,
      "eval_steps_per_second": 6.123,
      "step": 80
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00045000000000000004,
      "loss": 9.1542,
      "step": 100
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.8327292799949646,
      "eval_runtime": 0.4923,
      "eval_samples_per_second": 562.691,
      "eval_steps_per_second": 6.094,
      "step": 100
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.00044,
      "loss": 9.1755,
      "step": 120
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5270758122743683,
      "eval_loss": 0.8221527934074402,
      "eval_runtime": 0.4828,
      "eval_samples_per_second": 573.69,
      "eval_steps_per_second": 6.213,
      "step": 120
    },
    {
      "epoch": 6.0,
      "step": 120,
      "total_flos": 2186307132456960.0,
      "train_loss": 9.316140111287435,
      "train_runtime": 100.0781,
      "train_samples_per_second": 1244.029,
      "train_steps_per_second": 9.992
    }
  ],
  "max_steps": 1000,
  "num_train_epochs": 50,
  "total_flos": 2186307132456960.0,
  "trial_name": null,
  "trial_params": null
}