last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.002931175987806308,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00011724703951225231,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.00011724703951225231,
      "eval_loss": NaN,
      "eval_runtime": 433.9007,
      "eval_samples_per_second": 16.552,
      "eval_steps_per_second": 2.07,
      "step": 1
    },
    {
      "epoch": 0.00023449407902450463,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.00035174111853675694,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00046898815804900925,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.00046898815804900925,
      "eval_loss": NaN,
      "eval_runtime": 433.0334,
      "eval_samples_per_second": 16.585,
      "eval_steps_per_second": 2.074,
      "step": 4
    },
    {
      "epoch": 0.0005862351975612616,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0007034822370735139,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0008207292765857662,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0009379763160980185,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0009379763160980185,
      "eval_loss": NaN,
      "eval_runtime": 433.1489,
      "eval_samples_per_second": 16.581,
      "eval_steps_per_second": 2.073,
      "step": 8
    },
    {
      "epoch": 0.001055223355610271,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0011724703951225231,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0012897174346347756,
      "grad_norm": NaN,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.0014069644741470278,
      "grad_norm": NaN,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0014069644741470278,
      "eval_loss": NaN,
      "eval_runtime": 433.0996,
      "eval_samples_per_second": 16.583,
      "eval_steps_per_second": 2.073,
      "step": 12
    },
    {
      "epoch": 0.0015242115136592802,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.0016414585531715324,
      "grad_norm": NaN,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.0017587055926837848,
      "grad_norm": NaN,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.001875952632196037,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.001875952632196037,
      "eval_loss": NaN,
      "eval_runtime": 433.0076,
      "eval_samples_per_second": 16.586,
      "eval_steps_per_second": 2.074,
      "step": 16
    },
    {
      "epoch": 0.0019931996717082894,
      "grad_norm": NaN,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.002110446711220542,
      "grad_norm": NaN,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.002227693750732794,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.0023449407902450463,
      "grad_norm": NaN,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0023449407902450463,
      "eval_loss": NaN,
      "eval_runtime": 433.0097,
      "eval_samples_per_second": 16.586,
      "eval_steps_per_second": 2.074,
      "step": 20
    },
    {
      "epoch": 0.0024621878297572987,
      "grad_norm": NaN,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.002579434869269551,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.002696681908781803,
      "grad_norm": NaN,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.0028139289482940555,
      "grad_norm": NaN,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.0028139289482940555,
      "eval_loss": NaN,
      "eval_runtime": 433.0417,
      "eval_samples_per_second": 16.585,
      "eval_steps_per_second": 2.074,
      "step": 24
    },
    {
      "epoch": 0.002931175987806308,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 25,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6987107323084800.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}