tmp-checkpoint-4380/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 1000000000,
  "global_step": 4380,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "learning_rate": 3.680365296803653e-06,
      "loss": 39.2485,
      "step": 500
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.354207436399217e-06,
      "loss": 6.6145,
      "step": 1000
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.0280495759947813e-06,
      "loss": 5.3605,
      "step": 1500
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.7018917155903456e-06,
      "loss": 4.9587,
      "step": 2000
    },
    {
      "epoch": 1.14,
      "learning_rate": 2.37573385518591e-06,
      "loss": 4.7091,
      "step": 2500
    },
    {
      "epoch": 1.37,
      "learning_rate": 2.0495759947814745e-06,
      "loss": 4.4555,
      "step": 3000
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.7234181343770385e-06,
      "loss": 4.2992,
      "step": 3500
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.3972602739726028e-06,
      "loss": 4.2272,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 6132,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 4577396584501248.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}