checkpoint-300/trainer_state.json

{
  "best_metric": 2.3161551823972206,
  "best_model_checkpoint": "./whisper-small-clean_2-400/checkpoint-300",
  "epoch": 1.1425,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0625,
      "grad_norm": 0.07539740204811096,
      "learning_rate": 5e-06,
      "loss": 0.0016,
      "step": 25
    },
    {
      "epoch": 0.125,
      "grad_norm": 1.139564037322998,
      "learning_rate": 1e-05,
      "loss": 0.0007,
      "step": 50
    },
    {
      "epoch": 0.1875,
      "grad_norm": 0.043032076209783554,
      "learning_rate": 9.285714285714288e-06,
      "loss": 0.0088,
      "step": 75
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.14359460771083832,
      "learning_rate": 8.6e-06,
      "loss": 0.0153,
      "step": 100
    },
    {
      "epoch": 0.25,
      "eval_loss": 0.03707847744226456,
      "eval_runtime": 636.4956,
      "eval_samples_per_second": 2.713,
      "eval_steps_per_second": 0.085,
      "eval_wer": 5.7324840764331215,
      "step": 100
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.2424408346414566,
      "learning_rate": 7.885714285714286e-06,
      "loss": 0.0108,
      "step": 125
    },
    {
      "epoch": 0.375,
      "grad_norm": 0.038961008191108704,
      "learning_rate": 7.1714285714285725e-06,
      "loss": 0.0031,
      "step": 150
    },
    {
      "epoch": 0.4375,
      "grad_norm": 2.9564356803894043,
      "learning_rate": 6.4571428571428575e-06,
      "loss": 0.0014,
      "step": 175
    },
    {
      "epoch": 0.5,
      "grad_norm": Infinity,
      "learning_rate": 5.771428571428572e-06,
      "loss": 0.0363,
      "step": 200
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.04158010706305504,
      "eval_runtime": 630.2516,
      "eval_samples_per_second": 2.74,
      "eval_steps_per_second": 0.086,
      "eval_wer": 4.921829762594094,
      "step": 200
    },
    {
      "epoch": 0.5625,
      "grad_norm": 3.3036911487579346,
      "learning_rate": 5.114285714285715e-06,
      "loss": 0.7511,
      "step": 225
    },
    {
      "epoch": 1.0175,
      "grad_norm": 0.39931637048721313,
      "learning_rate": 4.4e-06,
      "loss": 0.1658,
      "step": 250
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.03213050216436386,
      "learning_rate": 3.6857142857142857e-06,
      "loss": 0.0005,
      "step": 275
    },
    {
      "epoch": 1.1425,
      "grad_norm": 0.008164944127202034,
      "learning_rate": 2.9714285714285716e-06,
      "loss": 0.0008,
      "step": 300
    },
    {
      "epoch": 1.1425,
      "eval_loss": 0.017297901213169098,
      "eval_runtime": 634.7882,
      "eval_samples_per_second": 2.721,
      "eval_steps_per_second": 0.085,
      "eval_wer": 2.3161551823972206,
      "step": 300
    }
  ],
  "logging_steps": 25,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.53910817374208e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}