checkpoint-1588/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.3699906803355079,
  "eval_steps": 318,
  "global_step": 1588,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00023299161230195712,
      "grad_norm": 6.78125,
      "learning_rate": 2e-06,
      "loss": 0.7236,
      "step": 1
    },
    {
      "epoch": 0.023299161230195712,
      "grad_norm": 0.09033203125,
      "learning_rate": 0.0002,
      "loss": 0.3799,
      "step": 100
    },
    {
      "epoch": 0.046598322460391424,
      "grad_norm": 0.11865234375,
      "learning_rate": 0.0004,
      "loss": 0.2452,
      "step": 200
    },
    {
      "epoch": 0.06989748369058714,
      "grad_norm": 0.1396484375,
      "learning_rate": 0.0006,
      "loss": 0.2131,
      "step": 300
    },
    {
      "epoch": 0.07409133271202237,
      "eval_peoplespeech-clean-transcription_loss": 3.0843491554260254,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.6247,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.376,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
      "step": 318
    },
    {
      "epoch": 0.09319664492078285,
      "grad_norm": 0.07421875,
      "learning_rate": 0.0008,
      "loss": 0.155,
      "step": 400
    },
    {
      "epoch": 0.11649580615097857,
      "grad_norm": 0.0576171875,
      "learning_rate": 0.001,
      "loss": 0.1083,
      "step": 500
    },
    {
      "epoch": 0.13979496738117428,
      "grad_norm": 0.04736328125,
      "learning_rate": 0.0012,
      "loss": 0.0926,
      "step": 600
    },
    {
      "epoch": 0.14818266542404473,
      "eval_peoplespeech-clean-transcription_loss": 1.7947263717651367,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 13.9513,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.587,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.072,
      "step": 636
    },
    {
      "epoch": 0.16309412861137,
      "grad_norm": 0.04150390625,
      "learning_rate": 0.0014,
      "loss": 0.0844,
      "step": 700
    },
    {
      "epoch": 0.1863932898415657,
      "grad_norm": 0.034912109375,
      "learning_rate": 0.0016,
      "loss": 0.0793,
      "step": 800
    },
    {
      "epoch": 0.2096924510717614,
      "grad_norm": 0.032958984375,
      "learning_rate": 0.0018000000000000002,
      "loss": 0.0758,
      "step": 900
    },
    {
      "epoch": 0.2222739981360671,
      "eval_peoplespeech-clean-transcription_loss": 1.7458150386810303,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.2202,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.501,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.07,
      "step": 954
    },
    {
      "epoch": 0.23299161230195714,
      "grad_norm": 0.03125,
      "learning_rate": 0.002,
      "loss": 0.0732,
      "step": 1000
    },
    {
      "epoch": 0.25629077353215285,
      "grad_norm": 0.02783203125,
      "learning_rate": 0.0019984487567773325,
      "loss": 0.0714,
      "step": 1100
    },
    {
      "epoch": 0.27958993476234856,
      "grad_norm": 0.0289306640625,
      "learning_rate": 0.0019938003745660765,
      "loss": 0.0691,
      "step": 1200
    },
    {
      "epoch": 0.29636533084808947,
      "eval_peoplespeech-clean-transcription_loss": 1.7118018865585327,
      "eval_peoplespeech-clean-transcription_model_preparation_time": 0.0063,
      "eval_peoplespeech-clean-transcription_runtime": 14.7608,
      "eval_peoplespeech-clean-transcription_samples_per_second": 4.336,
      "eval_peoplespeech-clean-transcription_steps_per_second": 0.068,
      "step": 1272
    },
    {
      "epoch": 0.30288909599254427,
      "grad_norm": 0.027587890625,
      "learning_rate": 0.0019860708773026797,
      "loss": 0.0676,
      "step": 1300
    },
    {
      "epoch": 0.32618825722274,
      "grad_norm": 0.0257568359375,
      "learning_rate": 0.001975286910165463,
      "loss": 0.0657,
      "step": 1400
    },
    {
      "epoch": 0.3494874184529357,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.0019614856477231713,
      "loss": 0.0648,
      "step": 1500
    }
  ],
  "logging_steps": 100,
  "max_steps": 6350,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1588,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.768821783434887e+18,
  "train_batch_size": 672,
  "trial_name": null,
  "trial_params": null
}