checkpoint-400/trainer_state.json

{
  "best_metric": 1.084672166739036,
  "best_model_checkpoint": "./whisper-tiny-qlora/checkpoint-200",
  "epoch": 0.24585125998770743,
  "eval_steps": 200,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015365703749231715,
      "grad_norm": 1.955202579498291,
      "learning_rate": 4.800000000000001e-07,
      "loss": 2.4438,
      "step": 25
    },
    {
      "epoch": 0.03073140749846343,
      "grad_norm": 1.465623140335083,
      "learning_rate": 9.800000000000001e-07,
      "loss": 2.0995,
      "step": 50
    },
    {
      "epoch": 0.046097111247695145,
      "grad_norm": 1.489674687385559,
      "learning_rate": 1.48e-06,
      "loss": 2.4302,
      "step": 75
    },
    {
      "epoch": 0.06146281499692686,
      "grad_norm": 1.7222563028335571,
      "learning_rate": 1.98e-06,
      "loss": 2.2989,
      "step": 100
    },
    {
      "epoch": 0.07682851874615858,
      "grad_norm": 2.750751256942749,
      "learning_rate": 2.4800000000000004e-06,
      "loss": 2.2729,
      "step": 125
    },
    {
      "epoch": 0.09219422249539029,
      "grad_norm": 2.440593957901001,
      "learning_rate": 2.9800000000000003e-06,
      "loss": 2.0241,
      "step": 150
    },
    {
      "epoch": 0.107559926244622,
      "grad_norm": 2.425014019012451,
      "learning_rate": 3.48e-06,
      "loss": 2.6992,
      "step": 175
    },
    {
      "epoch": 0.12292562999385372,
      "grad_norm": 1.8003255128860474,
      "learning_rate": 3.980000000000001e-06,
      "loss": 2.0456,
      "step": 200
    },
    {
      "epoch": 0.12292562999385372,
      "eval_loss": 2.245432138442993,
      "eval_runtime": 51.8391,
      "eval_samples_per_second": 7.851,
      "eval_steps_per_second": 0.251,
      "eval_wer": 1.084672166739036,
      "step": 200
    },
    {
      "epoch": 0.13829133374308544,
      "grad_norm": 1.9000691175460815,
      "learning_rate": 4.48e-06,
      "loss": 2.611,
      "step": 225
    },
    {
      "epoch": 0.15365703749231716,
      "grad_norm": 1.4311325550079346,
      "learning_rate": 4.960000000000001e-06,
      "loss": 2.4044,
      "step": 250
    },
    {
      "epoch": 0.16902274124154887,
      "grad_norm": 2.4365923404693604,
      "learning_rate": 5.460000000000001e-06,
      "loss": 2.1992,
      "step": 275
    },
    {
      "epoch": 0.18438844499078058,
      "grad_norm": 2.4958994388580322,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 2.5628,
      "step": 300
    },
    {
      "epoch": 0.1997541487400123,
      "grad_norm": 1.8009036779403687,
      "learning_rate": 6.460000000000001e-06,
      "loss": 1.8897,
      "step": 325
    },
    {
      "epoch": 0.215119852489244,
      "grad_norm": 1.8635932207107544,
      "learning_rate": 6.96e-06,
      "loss": 2.2831,
      "step": 350
    },
    {
      "epoch": 0.23048555623847572,
      "grad_norm": 1.8662347793579102,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 2.0718,
      "step": 375
    },
    {
      "epoch": 0.24585125998770743,
      "grad_norm": 2.4394640922546387,
      "learning_rate": 7.960000000000002e-06,
      "loss": 2.1359,
      "step": 400
    },
    {
      "epoch": 0.24585125998770743,
      "eval_loss": 2.1913249492645264,
      "eval_runtime": 49.901,
      "eval_samples_per_second": 8.156,
      "eval_steps_per_second": 0.261,
      "eval_wer": 1.1096396005210596,
      "step": 400
    }
  ],
  "logging_steps": 25,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2338970624e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}