trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 11760,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "learning_rate": 9.921387504651096e-05,
      "loss": 2.0594,
      "step": 1000
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.687402448381859e-05,
      "loss": 1.7419,
      "step": 2000
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.304978806833158e-05,
      "loss": 1.6843,
      "step": 3000
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.786952888283983e-05,
      "loss": 1.664,
      "step": 4000
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.149369270017385e-05,
      "loss": 1.6364,
      "step": 5000
    },
    {
      "epoch": 1.02,
      "learning_rate": 7.413137010913054e-05,
      "loss": 1.6023,
      "step": 6000
    },
    {
      "epoch": 1.19,
      "learning_rate": 6.599184470769908e-05,
      "loss": 1.5497,
      "step": 7000
    },
    {
      "epoch": 1.36,
      "learning_rate": 5.7355240963431157e-05,
      "loss": 1.5687,
      "step": 8000
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.8486425275577716e-05,
      "loss": 1.5239,
      "step": 9000
    },
    {
      "epoch": 1.7,
      "learning_rate": 3.96653945794421e-05,
      "loss": 1.5011,
      "step": 10000
    },
    {
      "epoch": 1.87,
      "learning_rate": 3.116238830710666e-05,
      "loss": 1.504,
      "step": 11000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 17640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10000,
  "total_flos": 2.13229959118848e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}