mms-1b-all-chichewa-1h/checkpoint-500/trainer_state.json

{
  "best_global_step": 500,
  "best_metric": 1.0,
  "best_model_checkpoint": "/home/jupyter-wb344850/chichewa-asr/models/checkpoints/mms-1b-all-chichewa-1h/checkpoint-500",
  "epoch": 125.0,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.25,
      "grad_norm": NaN,
      "learning_rate": 4.8e-05,
      "loss": 70.277255859375,
      "step": 25
    },
    {
      "epoch": 12.5,
      "grad_norm": NaN,
      "learning_rate": 9.800000000000001e-05,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 18.75,
      "grad_norm": NaN,
      "learning_rate": 0.000148,
      "loss": 0.0,
      "step": 75
    },
    {
      "epoch": 25.0,
      "grad_norm": NaN,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.0,
      "step": 100
    },
    {
      "epoch": 31.25,
      "grad_norm": NaN,
      "learning_rate": 0.000248,
      "loss": 0.0,
      "step": 125
    },
    {
      "epoch": 37.5,
      "grad_norm": NaN,
      "learning_rate": 0.000298,
      "loss": 0.0,
      "step": 150
    },
    {
      "epoch": 43.75,
      "grad_norm": NaN,
      "learning_rate": 0.000348,
      "loss": 0.0,
      "step": 175
    },
    {
      "epoch": 50.0,
      "grad_norm": NaN,
      "learning_rate": 0.000398,
      "loss": 0.0,
      "step": 200
    },
    {
      "epoch": 56.25,
      "grad_norm": NaN,
      "learning_rate": 0.000448,
      "loss": 0.0,
      "step": 225
    },
    {
      "epoch": 62.5,
      "grad_norm": NaN,
      "learning_rate": 0.000498,
      "loss": 0.0,
      "step": 250
    },
    {
      "epoch": 68.75,
      "grad_norm": NaN,
      "learning_rate": 0.0005480000000000001,
      "loss": 0.0,
      "step": 275
    },
    {
      "epoch": 75.0,
      "grad_norm": NaN,
      "learning_rate": 0.000598,
      "loss": 0.0,
      "step": 300
    },
    {
      "epoch": 81.25,
      "grad_norm": NaN,
      "learning_rate": 0.000648,
      "loss": 0.0,
      "step": 325
    },
    {
      "epoch": 87.5,
      "grad_norm": NaN,
      "learning_rate": 0.0006979999999999999,
      "loss": 0.0,
      "step": 350
    },
    {
      "epoch": 93.75,
      "grad_norm": NaN,
      "learning_rate": 0.000748,
      "loss": 0.0,
      "step": 375
    },
    {
      "epoch": 100.0,
      "grad_norm": NaN,
      "learning_rate": 0.0007980000000000001,
      "loss": 0.0,
      "step": 400
    },
    {
      "epoch": 106.25,
      "grad_norm": NaN,
      "learning_rate": 0.000848,
      "loss": 0.0,
      "step": 425
    },
    {
      "epoch": 112.5,
      "grad_norm": NaN,
      "learning_rate": 0.000898,
      "loss": 0.0,
      "step": 450
    },
    {
      "epoch": 118.75,
      "grad_norm": NaN,
      "learning_rate": 0.000948,
      "loss": 0.0,
      "step": 475
    },
    {
      "epoch": 125.0,
      "grad_norm": NaN,
      "learning_rate": 0.000998,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 125.0,
      "eval_loss": NaN,
      "eval_runtime": 1.3702,
      "eval_samples_per_second": 24.815,
      "eval_steps_per_second": 2.19,
      "eval_wer": 1.0,
      "step": 500
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1250,
  "save_steps": 500,
  "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": 7.435379130650906e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}