checkpoint-141/trainer_state.json

{
  "best_global_step": 141,
  "best_metric": 2.384185791015625e-07,
  "best_model_checkpoint": "./star_trek_guard_finetuned/checkpoint-141",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 141,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0071111111111111115,
      "grad_norm": 863.2054443359375,
      "learning_rate": 0.0,
      "loss": 31.4624,
      "step": 1
    },
    {
      "epoch": 0.07111111111111111,
      "grad_norm": 975.65576171875,
      "learning_rate": 4.186046511627907e-05,
      "loss": 35.8397,
      "step": 10
    },
    {
      "epoch": 0.14222222222222222,
      "grad_norm": 198.86065673828125,
      "learning_rate": 8.837209302325582e-05,
      "loss": 8.8149,
      "step": 20
    },
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 8.080985069274902,
      "learning_rate": 0.00013488372093023256,
      "loss": 0.5709,
      "step": 30
    },
    {
      "epoch": 0.28444444444444444,
      "grad_norm": 0.00014901161193847656,
      "learning_rate": 0.0001813953488372093,
      "loss": 0.1449,
      "step": 40
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 231.45864868164062,
      "learning_rate": 0.00019987699691483048,
      "loss": 2.3554,
      "step": 50
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 0.17169412970542908,
      "learning_rate": 0.00019912640693269752,
      "loss": 0.7842,
      "step": 60
    },
    {
      "epoch": 0.49777777777777776,
      "grad_norm": 0.0002152260858565569,
      "learning_rate": 0.00019769868307835994,
      "loss": 0.3842,
      "step": 70
    },
    {
      "epoch": 0.5688888888888889,
      "grad_norm": 0.7667607665061951,
      "learning_rate": 0.00019560357815343577,
      "loss": 0.1437,
      "step": 80
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.00011813640594482422,
      "learning_rate": 0.00019285540384897073,
      "loss": 0.1796,
      "step": 90
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 0.014821142889559269,
      "learning_rate": 0.00018947293298207635,
      "loss": 0.0,
      "step": 100
    },
    {
      "epoch": 0.7822222222222223,
      "grad_norm": 0.0,
      "learning_rate": 0.0001854792712585539,
      "loss": 0.0002,
      "step": 110
    },
    {
      "epoch": 0.8533333333333334,
      "grad_norm": 0.0,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 120
    },
    {
      "epoch": 0.9244444444444444,
      "grad_norm": 0.0003181487263645977,
      "learning_rate": 0.0001757714869760335,
      "loss": 0.4035,
      "step": 130
    },
    {
      "epoch": 0.9955555555555555,
      "grad_norm": 0.0,
      "learning_rate": 0.00017012367842724887,
      "loss": 0.3667,
      "step": 140
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.384185791015625e-07,
      "eval_runtime": 24.7269,
      "eval_samples_per_second": 20.221,
      "eval_steps_per_second": 5.055,
      "step": 141
    }
  ],
  "logging_steps": 10,
  "max_steps": 423,
  "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": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.068641927168e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}