trainer_state.json

{
  "best_metric": 0.9136,
  "best_model_checkpoint": "checkpoint/deberta-base/imdb_42/checkpoint-6256",
  "epoch": 10.0,
  "global_step": 7820,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.91168,
      "eval_loss": 0.22001588344573975,
      "eval_runtime": 22.3141,
      "eval_samples_per_second": 1120.369,
      "eval_steps_per_second": 4.392,
      "step": 782
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.360613810741688e-05,
      "loss": 0.2741,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.90984,
      "eval_loss": 0.23234927654266357,
      "eval_runtime": 22.2845,
      "eval_samples_per_second": 1121.854,
      "eval_steps_per_second": 4.398,
      "step": 1564
    },
    {
      "epoch": 2.56,
      "learning_rate": 3.721227621483376e-05,
      "loss": 0.1524,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.91116,
      "eval_loss": 0.31343552470207214,
      "eval_runtime": 22.2601,
      "eval_samples_per_second": 1123.083,
      "eval_steps_per_second": 4.402,
      "step": 2346
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.081841432225064e-05,
      "loss": 0.0863,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.91352,
      "eval_loss": 0.36432042717933655,
      "eval_runtime": 22.1164,
      "eval_samples_per_second": 1130.384,
      "eval_steps_per_second": 4.431,
      "step": 3128
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.90884,
      "eval_loss": 0.3902149498462677,
      "eval_runtime": 22.2414,
      "eval_samples_per_second": 1124.03,
      "eval_steps_per_second": 4.406,
      "step": 3910
    },
    {
      "epoch": 5.12,
      "learning_rate": 2.442455242966752e-05,
      "loss": 0.046,
      "step": 4000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9102,
      "eval_loss": 0.45257332921028137,
      "eval_runtime": 22.1568,
      "eval_samples_per_second": 1128.324,
      "eval_steps_per_second": 4.423,
      "step": 4692
    },
    {
      "epoch": 6.39,
      "learning_rate": 1.80306905370844e-05,
      "loss": 0.0244,
      "step": 5000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.91224,
      "eval_loss": 0.5029028058052063,
      "eval_runtime": 22.1112,
      "eval_samples_per_second": 1130.65,
      "eval_steps_per_second": 4.432,
      "step": 5474
    },
    {
      "epoch": 7.67,
      "learning_rate": 1.163682864450128e-05,
      "loss": 0.019,
      "step": 6000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9136,
      "eval_loss": 0.5574696063995361,
      "eval_runtime": 22.1411,
      "eval_samples_per_second": 1129.122,
      "eval_steps_per_second": 4.426,
      "step": 6256
    },
    {
      "epoch": 8.95,
      "learning_rate": 5.242966751918159e-06,
      "loss": 0.0114,
      "step": 7000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.91356,
      "eval_loss": 0.5506225228309631,
      "eval_runtime": 22.2458,
      "eval_samples_per_second": 1123.808,
      "eval_steps_per_second": 4.405,
      "step": 7038
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.91312,
      "eval_loss": 0.559421956539154,
      "eval_runtime": 22.097,
      "eval_samples_per_second": 1131.374,
      "eval_steps_per_second": 4.435,
      "step": 7820
    },
    {
      "epoch": 10.0,
      "step": 7820,
      "total_flos": 1.6444735872e+16,
      "train_loss": 0.07955141689466394,
      "train_runtime": 1147.935,
      "train_samples_per_second": 217.782,
      "train_steps_per_second": 6.812
    }
  ],
  "max_steps": 7820,
  "num_train_epochs": 10,
  "total_flos": 1.6444735872e+16,
  "trial_name": null,
  "trial_params": null
}