checkpoint-6620/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.998112495281238,
  "eval_steps": 500,
  "global_step": 6620,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.38,
      "learning_rate": 4.9800000000000004e-05,
      "loss": 0.0323,
      "step": 500
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.91875452905494e-05,
      "loss": 0.0536,
      "step": 1000
    },
    {
      "epoch": 1.13,
      "learning_rate": 4.6790417797358055e-05,
      "loss": 0.0491,
      "step": 1500
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.296565601079351e-05,
      "loss": 0.0441,
      "step": 2000
    },
    {
      "epoch": 1.89,
      "learning_rate": 3.7963845355204304e-05,
      "loss": 0.0431,
      "step": 2500
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.2112687506037174e-05,
      "loss": 0.0328,
      "step": 3000
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.5795530487138815e-05,
      "loss": 0.0287,
      "step": 3500
    },
    {
      "epoch": 3.02,
      "learning_rate": 1.9426253007980113e-05,
      "loss": 0.0262,
      "step": 4000
    },
    {
      "epoch": 3.4,
      "learning_rate": 1.3444902911492174e-05,
      "loss": 0.0141,
      "step": 4500
    },
    {
      "epoch": 3.78,
      "learning_rate": 8.195580089631732e-06,
      "loss": 0.0138,
      "step": 5000
    },
    {
      "epoch": 4.15,
      "learning_rate": 4.047225884682987e-06,
      "loss": 0.0106,
      "step": 5500
    },
    {
      "epoch": 4.53,
      "learning_rate": 1.2716263991024713e-06,
      "loss": 0.012,
      "step": 6000
    },
    {
      "epoch": 4.91,
      "learning_rate": 5.144911675032671e-08,
      "loss": 0.0113,
      "step": 6500
    }
  ],
  "logging_steps": 500,
  "max_steps": 6620,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 5.612325759297946e+16,
  "trial_name": null,
  "trial_params": null
}