last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1918158567774936,
  "eval_steps": 13,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0038363171355498722,
      "eval_loss": NaN,
      "eval_runtime": 38.5176,
      "eval_samples_per_second": 17.109,
      "eval_steps_per_second": 4.284,
      "step": 1
    },
    {
      "epoch": 0.011508951406649617,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.023017902813299233,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.034526854219948847,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.04603580562659847,
      "grad_norm": NaN,
      "learning_rate": 0.00019876883405951377,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.049872122762148335,
      "eval_loss": NaN,
      "eval_runtime": 37.334,
      "eval_samples_per_second": 17.651,
      "eval_steps_per_second": 4.42,
      "step": 13
    },
    {
      "epoch": 0.05754475703324808,
      "grad_norm": NaN,
      "learning_rate": 0.0001923879532511287,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.06905370843989769,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.08056265984654731,
      "grad_norm": NaN,
      "learning_rate": 0.00016494480483301836,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.09207161125319693,
      "grad_norm": NaN,
      "learning_rate": 0.00014539904997395468,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.09974424552429667,
      "eval_loss": NaN,
      "eval_runtime": 37.425,
      "eval_samples_per_second": 17.609,
      "eval_steps_per_second": 4.409,
      "step": 26
    },
    {
      "epoch": 0.10358056265984655,
      "grad_norm": NaN,
      "learning_rate": 0.00012334453638559057,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.11508951406649616,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.1265984654731458,
      "grad_norm": NaN,
      "learning_rate": 7.66554636144095e-05,
      "loss": 0.0,
      "step": 33
    },
    {
      "epoch": 0.13810741687979539,
      "grad_norm": NaN,
      "learning_rate": 5.4600950026045326e-05,
      "loss": 0.0,
      "step": 36
    },
    {
      "epoch": 0.149616368286445,
      "grad_norm": NaN,
      "learning_rate": 3.5055195166981645e-05,
      "loss": 0.0,
      "step": 39
    },
    {
      "epoch": 0.149616368286445,
      "eval_loss": NaN,
      "eval_runtime": 37.4586,
      "eval_samples_per_second": 17.593,
      "eval_steps_per_second": 4.405,
      "step": 39
    },
    {
      "epoch": 0.16112531969309463,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 42
    },
    {
      "epoch": 0.17263427109974425,
      "grad_norm": NaN,
      "learning_rate": 7.612046748871327e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.18414322250639387,
      "grad_norm": NaN,
      "learning_rate": 1.231165940486234e-06,
      "loss": 0.0,
      "step": 48
    }
  ],
  "logging_steps": 3,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.455273774396211e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}