checkpoint-1458/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1458,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "learning_rate": 0.00048,
      "loss": 7.0366,
      "step": 100
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0004933240611961058,
      "loss": 0.4467,
      "step": 200
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00048643949930458967,
      "loss": 0.3946,
      "step": 300
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0004794853963838665,
      "loss": 0.399,
      "step": 400
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0004726008344923505,
      "loss": 0.3696,
      "step": 500
    },
    {
      "epoch": 0.69,
      "eval_loss": 0.3043254017829895,
      "eval_runtime": 65.1553,
      "eval_samples_per_second": 11.664,
      "eval_steps_per_second": 1.458,
      "eval_wer": 0.265993265993266,
      "step": 500
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0004656467315716273,
      "loss": 0.3677,
      "step": 600
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00045869262865090403,
      "loss": 0.3534,
      "step": 700
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00045173852573018083,
      "loss": 0.34,
      "step": 800
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0004447844228094576,
      "loss": 0.3428,
      "step": 900
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0004378998609179416,
      "loss": 0.3533,
      "step": 1000
    },
    {
      "epoch": 1.37,
      "eval_loss": 0.27337372303009033,
      "eval_runtime": 65.7997,
      "eval_samples_per_second": 11.55,
      "eval_steps_per_second": 1.444,
      "eval_wer": 0.23931623931623933,
      "step": 1000
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00043094575799721834,
      "loss": 0.336,
      "step": 1100
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00042399165507649514,
      "loss": 0.3453,
      "step": 1200
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00041703755215577194,
      "loss": 0.342,
      "step": 1300
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0004100834492350487,
      "loss": 0.3232,
      "step": 1400
    }
  ],
  "logging_steps": 100,
  "max_steps": 7290,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.0100542581793982e+19,
  "trial_name": null,
  "trial_params": null
}