checkpoint-150/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.43509789702683105,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 3.75e-05,
      "loss": 12.1562,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.125e-05,
      "loss": 8.7679,
      "step": 20
    },
    {
      "epoch": 0.09,
      "learning_rate": 7.398952095808383e-05,
      "loss": 5.3683,
      "step": 30
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.286676646706586e-05,
      "loss": 4.3219,
      "step": 40
    },
    {
      "epoch": 0.15,
      "learning_rate": 7.17440119760479e-05,
      "loss": 3.7182,
      "step": 50
    },
    {
      "epoch": 0.15,
      "eval_loss": 3.836604595184326,
      "eval_runtime": 133.4846,
      "eval_samples_per_second": 34.611,
      "eval_steps_per_second": 4.33,
      "eval_wer": 1.0,
      "step": 50
    },
    {
      "epoch": 0.17,
      "learning_rate": 7.062125748502993e-05,
      "loss": 3.478,
      "step": 60
    },
    {
      "epoch": 0.2,
      "learning_rate": 6.949850299401197e-05,
      "loss": 3.4492,
      "step": 70
    },
    {
      "epoch": 0.23,
      "learning_rate": 6.837574850299401e-05,
      "loss": 3.3928,
      "step": 80
    },
    {
      "epoch": 0.26,
      "learning_rate": 6.725299401197604e-05,
      "loss": 3.3183,
      "step": 90
    },
    {
      "epoch": 0.29,
      "learning_rate": 6.613023952095809e-05,
      "loss": 3.2075,
      "step": 100
    },
    {
      "epoch": 0.29,
      "eval_loss": 3.258362293243408,
      "eval_runtime": 126.6078,
      "eval_samples_per_second": 36.491,
      "eval_steps_per_second": 4.565,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 0.32,
      "learning_rate": 6.500748502994012e-05,
      "loss": 3.14,
      "step": 110
    },
    {
      "epoch": 0.35,
      "learning_rate": 6.388473053892215e-05,
      "loss": 3.1281,
      "step": 120
    },
    {
      "epoch": 0.38,
      "learning_rate": 6.276197604790418e-05,
      "loss": 3.0987,
      "step": 130
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.163922155688622e-05,
      "loss": 3.1003,
      "step": 140
    },
    {
      "epoch": 0.44,
      "learning_rate": 6.0516467065868256e-05,
      "loss": 3.0922,
      "step": 150
    },
    {
      "epoch": 0.44,
      "eval_loss": 3.127869129180908,
      "eval_runtime": 126.3837,
      "eval_samples_per_second": 36.555,
      "eval_steps_per_second": 4.573,
      "eval_wer": 1.0,
      "step": 150
    }
  ],
  "max_steps": 688,
  "num_train_epochs": 2,
  "total_flos": 4.4091799296804864e+17,
  "trial_name": null,
  "trial_params": null
}