trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.912280701754386,
  "eval_steps": 100,
  "global_step": 84,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03508771929824561,
      "grad_norm": 13.274463118636966,
      "learning_rate": 6.666666666666667e-06,
      "loss": 1.227,
      "step": 1
    },
    {
      "epoch": 0.17543859649122806,
      "grad_norm": 13.067768872256629,
      "learning_rate": 1.996992941167792e-05,
      "loss": 1.2833,
      "step": 5
    },
    {
      "epoch": 0.3508771929824561,
      "grad_norm": 3.029737230058556,
      "learning_rate": 1.9633708786158803e-05,
      "loss": 0.9766,
      "step": 10
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 1.6344674238695178,
      "learning_rate": 1.8936326403234125e-05,
      "loss": 0.8109,
      "step": 15
    },
    {
      "epoch": 0.7017543859649122,
      "grad_norm": 1.440061411744547,
      "learning_rate": 1.7903926695187595e-05,
      "loss": 0.7554,
      "step": 20
    },
    {
      "epoch": 0.8771929824561403,
      "grad_norm": 1.37302749270618,
      "learning_rate": 1.657521368569064e-05,
      "loss": 0.7119,
      "step": 25
    },
    {
      "epoch": 1.0350877192982457,
      "grad_norm": 1.7984147567981632,
      "learning_rate": 1.5000000000000002e-05,
      "loss": 0.6836,
      "step": 30
    },
    {
      "epoch": 1.2105263157894737,
      "grad_norm": 1.005885201758115,
      "learning_rate": 1.3237339420583213e-05,
      "loss": 0.6187,
      "step": 35
    },
    {
      "epoch": 1.3859649122807016,
      "grad_norm": 0.9917450064965392,
      "learning_rate": 1.1353312997501313e-05,
      "loss": 0.5963,
      "step": 40
    },
    {
      "epoch": 1.5614035087719298,
      "grad_norm": 0.9760187114551988,
      "learning_rate": 9.418551710895243e-06,
      "loss": 0.5995,
      "step": 45
    },
    {
      "epoch": 1.736842105263158,
      "grad_norm": 0.976366471591892,
      "learning_rate": 7.505588559420188e-06,
      "loss": 0.5917,
      "step": 50
    },
    {
      "epoch": 1.912280701754386,
      "grad_norm": 0.8757263720939512,
      "learning_rate": 5.686139343187468e-06,
      "loss": 0.5868,
      "step": 55
    },
    {
      "epoch": 2.0701754385964914,
      "grad_norm": 0.7705568313648676,
      "learning_rate": 4.028414082972141e-06,
      "loss": 0.554,
      "step": 60
    },
    {
      "epoch": 2.245614035087719,
      "grad_norm": 0.7551160537629242,
      "learning_rate": 2.594559868909956e-06,
      "loss": 0.5186,
      "step": 65
    },
    {
      "epoch": 2.4210526315789473,
      "grad_norm": 0.737482342547591,
      "learning_rate": 1.4383310046973365e-06,
      "loss": 0.5083,
      "step": 70
    },
    {
      "epoch": 2.5964912280701755,
      "grad_norm": 0.7345189094446426,
      "learning_rate": 6.030737921409169e-07,
      "loss": 0.5097,
      "step": 75
    },
    {
      "epoch": 2.7719298245614032,
      "grad_norm": 0.6830233810548365,
      "learning_rate": 1.201015052319099e-07,
      "loss": 0.5063,
      "step": 80
    },
    {
      "epoch": 2.912280701754386,
      "step": 84,
      "total_flos": 6.582526671847424e+16,
      "train_loss": 0.6669661885216123,
      "train_runtime": 2548.4796,
      "train_samples_per_second": 0.533,
      "train_steps_per_second": 0.033
    }
  ],
  "logging_steps": 5,
  "max_steps": 84,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.582526671847424e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}