last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0028410705153701916,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00011364282061480765,
      "grad_norm": NaN,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.00011364282061480765,
      "eval_loss": NaN,
      "eval_runtime": 431.3888,
      "eval_samples_per_second": 8.589,
      "eval_steps_per_second": 2.149,
      "step": 1
    },
    {
      "epoch": 0.0002272856412296153,
      "grad_norm": NaN,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.000340928461844423,
      "grad_norm": NaN,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0004545712824592306,
      "grad_norm": NaN,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0004545712824592306,
      "eval_loss": NaN,
      "eval_runtime": 431.0085,
      "eval_samples_per_second": 8.596,
      "eval_steps_per_second": 2.151,
      "step": 4
    },
    {
      "epoch": 0.0005682141030740383,
      "grad_norm": NaN,
      "learning_rate": 5e-06,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.000681856923688846,
      "grad_norm": NaN,
      "learning_rate": 6e-06,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0007954997443036536,
      "grad_norm": NaN,
      "learning_rate": 7e-06,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0009091425649184612,
      "grad_norm": NaN,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0009091425649184612,
      "eval_loss": NaN,
      "eval_runtime": 431.1027,
      "eval_samples_per_second": 8.594,
      "eval_steps_per_second": 2.15,
      "step": 8
    },
    {
      "epoch": 0.001022785385533269,
      "grad_norm": NaN,
      "learning_rate": 9e-06,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0011364282061480766,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0012500710267628842,
      "grad_norm": NaN,
      "learning_rate": 9.890738003669029e-06,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.001363713847377692,
      "grad_norm": NaN,
      "learning_rate": 9.567727288213005e-06,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.001363713847377692,
      "eval_loss": NaN,
      "eval_runtime": 431.6008,
      "eval_samples_per_second": 8.584,
      "eval_steps_per_second": 2.148,
      "step": 12
    },
    {
      "epoch": 0.0014773566679924996,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.0015909994886073072,
      "grad_norm": NaN,
      "learning_rate": 8.345653031794292e-06,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.0017046423092221148,
      "grad_norm": NaN,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0018182851298369225,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.0018182851298369225,
      "eval_loss": NaN,
      "eval_runtime": 430.7717,
      "eval_samples_per_second": 8.601,
      "eval_steps_per_second": 2.152,
      "step": 16
    },
    {
      "epoch": 0.0019319279504517303,
      "grad_norm": NaN,
      "learning_rate": 5.522642316338268e-06,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.002045570771066538,
      "grad_norm": NaN,
      "learning_rate": 4.477357683661734e-06,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0021592135916813455,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.002272856412296153,
      "grad_norm": NaN,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.002272856412296153,
      "eval_loss": NaN,
      "eval_runtime": 430.7811,
      "eval_samples_per_second": 8.601,
      "eval_steps_per_second": 2.152,
      "step": 20
    },
    {
      "epoch": 0.0023864992329109607,
      "grad_norm": NaN,
      "learning_rate": 1.6543469682057105e-06,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0025001420535257684,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.002613784874140576,
      "grad_norm": NaN,
      "learning_rate": 4.322727117869951e-07,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.002727427694755384,
      "grad_norm": NaN,
      "learning_rate": 1.0926199633097156e-07,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.002727427694755384,
      "eval_loss": NaN,
      "eval_runtime": 431.2707,
      "eval_samples_per_second": 8.591,
      "eval_steps_per_second": 2.149,
      "step": 24
    },
    {
      "epoch": 0.0028410705153701916,
      "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": 9273591044505600.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}