trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 17.0,
  "global_step": 8772,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.96046511627907e-05,
      "loss": 1.5989,
      "step": 516
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.9204651162790698e-05,
      "loss": 0.7816,
      "step": 1032
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.88046511627907e-05,
      "loss": 0.6128,
      "step": 1548
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.84046511627907e-05,
      "loss": 0.5083,
      "step": 2064
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.80046511627907e-05,
      "loss": 0.4314,
      "step": 2580
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.76046511627907e-05,
      "loss": 0.3745,
      "step": 3096
    },
    {
      "epoch": 7.0,
      "learning_rate": 1.72046511627907e-05,
      "loss": 0.3258,
      "step": 3612
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.68046511627907e-05,
      "loss": 0.2907,
      "step": 4128
    },
    {
      "epoch": 9.0,
      "learning_rate": 1.6404651162790698e-05,
      "loss": 0.2571,
      "step": 4644
    },
    {
      "epoch": 10.0,
      "learning_rate": 1.60046511627907e-05,
      "loss": 0.2337,
      "step": 5160
    },
    {
      "epoch": 11.0,
      "learning_rate": 1.56046511627907e-05,
      "loss": 0.211,
      "step": 5676
    },
    {
      "epoch": 12.0,
      "learning_rate": 1.5204651162790698e-05,
      "loss": 0.191,
      "step": 6192
    },
    {
      "epoch": 13.0,
      "learning_rate": 1.48046511627907e-05,
      "loss": 0.1765,
      "step": 6708
    },
    {
      "epoch": 14.0,
      "learning_rate": 1.4404651162790698e-05,
      "loss": 0.1636,
      "step": 7224
    },
    {
      "epoch": 15.0,
      "learning_rate": 1.4004651162790699e-05,
      "loss": 0.1494,
      "step": 7740
    },
    {
      "epoch": 16.0,
      "learning_rate": 1.36046511627907e-05,
      "loss": 0.1378,
      "step": 8256
    },
    {
      "epoch": 17.0,
      "learning_rate": 1.3204651162790698e-05,
      "loss": 0.1302,
      "step": 8772
    }
  ],
  "max_steps": 25800,
  "num_train_epochs": 50,
  "total_flos": 4177415800160256.0,
  "trial_name": null,
  "trial_params": null
}