last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.009192866335723478,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00036771465342893915,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.00036771465342893915,
      "eval_loss": NaN,
      "eval_runtime": 110.949,
      "eval_samples_per_second": 20.64,
      "eval_steps_per_second": 2.587,
      "step": 1
    },
    {
      "epoch": 0.0007354293068578783,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.0011031439602868175,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0014708586137157566,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0014708586137157566,
      "eval_loss": NaN,
      "eval_runtime": 108.6019,
      "eval_samples_per_second": 21.086,
      "eval_steps_per_second": 2.643,
      "step": 4
    },
    {
      "epoch": 0.0018385732671446957,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.002206287920573635,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.002574002574002574,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0029417172274315132,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0029417172274315132,
      "eval_loss": NaN,
      "eval_runtime": 108.3915,
      "eval_samples_per_second": 21.127,
      "eval_steps_per_second": 2.648,
      "step": 8
    },
    {
      "epoch": 0.003309431880860452,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0036771465342893914,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.004044861187718331,
      "grad_norm": NaN,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.00441257584114727,
      "grad_norm": NaN,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.00441257584114727,
      "eval_loss": NaN,
      "eval_runtime": 108.7166,
      "eval_samples_per_second": 21.064,
      "eval_steps_per_second": 2.64,
      "step": 12
    },
    {
      "epoch": 0.0047802904945762085,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.005148005148005148,
      "grad_norm": NaN,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.005515719801434087,
      "grad_norm": NaN,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0058834344548630265,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.0058834344548630265,
      "eval_loss": NaN,
      "eval_runtime": 108.8123,
      "eval_samples_per_second": 21.045,
      "eval_steps_per_second": 2.638,
      "step": 16
    },
    {
      "epoch": 0.006251149108291966,
      "grad_norm": NaN,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.006618863761720904,
      "grad_norm": NaN,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0069865784151498436,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.007354293068578783,
      "grad_norm": NaN,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.007354293068578783,
      "eval_loss": NaN,
      "eval_runtime": 108.5781,
      "eval_samples_per_second": 21.091,
      "eval_steps_per_second": 2.643,
      "step": 20
    },
    {
      "epoch": 0.007722007722007722,
      "grad_norm": NaN,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.008089722375436662,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.0084574370288656,
      "grad_norm": NaN,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.00882515168229454,
      "grad_norm": NaN,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.00882515168229454,
      "eval_loss": NaN,
      "eval_runtime": 108.9589,
      "eval_samples_per_second": 21.017,
      "eval_steps_per_second": 2.634,
      "step": 24
    },
    {
      "epoch": 0.009192866335723478,
      "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": 3338642864996352.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}