checkpoint-16/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.8484848484848486,
  "eval_steps": 500,
  "global_step": 16,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12121212121212122,
      "grad_norm": 2.50050950050354,
      "learning_rate": 2.5e-06,
      "loss": 0.8407,
      "step": 1
    },
    {
      "epoch": 0.24242424242424243,
      "grad_norm": 2.2383944988250732,
      "learning_rate": 5e-06,
      "loss": 0.7878,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.1805531978607178,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.8026,
      "step": 3
    },
    {
      "epoch": 0.48484848484848486,
      "grad_norm": 2.0964150428771973,
      "learning_rate": 1e-05,
      "loss": 0.8133,
      "step": 4
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 1.5487000942230225,
      "learning_rate": 9.968561049466214e-06,
      "loss": 0.7608,
      "step": 5
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.7998461723327637,
      "learning_rate": 9.874639560909118e-06,
      "loss": 0.7272,
      "step": 6
    },
    {
      "epoch": 0.8484848484848485,
      "grad_norm": 1.9523899555206299,
      "learning_rate": 9.719416651541839e-06,
      "loss": 0.6923,
      "step": 7
    },
    {
      "epoch": 0.9696969696969697,
      "grad_norm": 4.303847312927246,
      "learning_rate": 9.504844339512096e-06,
      "loss": 0.6853,
      "step": 8
    },
    {
      "epoch": 1.0,
      "grad_norm": 4.303847312927246,
      "learning_rate": 9.233620996141421e-06,
      "loss": 0.6532,
      "step": 9
    },
    {
      "epoch": 1.121212121212121,
      "grad_norm": 2.6484241485595703,
      "learning_rate": 8.90915741234015e-06,
      "loss": 0.6334,
      "step": 10
    },
    {
      "epoch": 1.2424242424242424,
      "grad_norm": 1.0852004289627075,
      "learning_rate": 8.535533905932739e-06,
      "loss": 0.6171,
      "step": 11
    },
    {
      "epoch": 1.3636363636363638,
      "grad_norm": 87.41819763183594,
      "learning_rate": 8.117449009293668e-06,
      "loss": 0.5768,
      "step": 12
    },
    {
      "epoch": 1.4848484848484849,
      "grad_norm": 1.8918662071228027,
      "learning_rate": 7.660160382576683e-06,
      "loss": 0.5852,
      "step": 13
    },
    {
      "epoch": 1.606060606060606,
      "grad_norm": 1.3301656246185303,
      "learning_rate": 7.169418695587791e-06,
      "loss": 0.5752,
      "step": 14
    },
    {
      "epoch": 1.7272727272727273,
      "grad_norm": 0.9969548583030701,
      "learning_rate": 6.651395309775837e-06,
      "loss": 0.5949,
      "step": 15
    },
    {
      "epoch": 1.8484848484848486,
      "grad_norm": 0.9531848430633545,
      "learning_rate": 6.112604669781572e-06,
      "loss": 0.5568,
      "step": 16
    }
  ],
  "logging_steps": 1,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 8,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 39941324668928.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}