last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7086614173228346,
  "eval_steps": 5,
  "global_step": 45,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015748031496062992,
      "eval_loss": 2.1424481868743896,
      "eval_runtime": 20.1121,
      "eval_samples_per_second": 5.32,
      "eval_steps_per_second": 0.696,
      "step": 1
    },
    {
      "epoch": 0.047244094488188976,
      "grad_norm": 1.0549567937850952,
      "learning_rate": 3e-05,
      "loss": 8.6552,
      "step": 3
    },
    {
      "epoch": 0.07874015748031496,
      "eval_loss": 2.102262020111084,
      "eval_runtime": 20.3039,
      "eval_samples_per_second": 5.27,
      "eval_steps_per_second": 0.69,
      "step": 5
    },
    {
      "epoch": 0.09448818897637795,
      "grad_norm": 0.8591325879096985,
      "learning_rate": 6e-05,
      "loss": 8.4913,
      "step": 6
    },
    {
      "epoch": 0.14173228346456693,
      "grad_norm": 0.8985063433647156,
      "learning_rate": 9e-05,
      "loss": 8.3402,
      "step": 9
    },
    {
      "epoch": 0.15748031496062992,
      "eval_loss": 2.0339243412017822,
      "eval_runtime": 20.3137,
      "eval_samples_per_second": 5.267,
      "eval_steps_per_second": 0.689,
      "step": 10
    },
    {
      "epoch": 0.1889763779527559,
      "grad_norm": 0.6786600947380066,
      "learning_rate": 9.938441702975689e-05,
      "loss": 8.2753,
      "step": 12
    },
    {
      "epoch": 0.23622047244094488,
      "grad_norm": 0.7172123789787292,
      "learning_rate": 9.619397662556435e-05,
      "loss": 8.0933,
      "step": 15
    },
    {
      "epoch": 0.23622047244094488,
      "eval_loss": 1.996870994567871,
      "eval_runtime": 20.3307,
      "eval_samples_per_second": 5.263,
      "eval_steps_per_second": 0.689,
      "step": 15
    },
    {
      "epoch": 0.28346456692913385,
      "grad_norm": 0.6016080975532532,
      "learning_rate": 9.045084971874738e-05,
      "loss": 8.0482,
      "step": 18
    },
    {
      "epoch": 0.31496062992125984,
      "eval_loss": 1.9798306226730347,
      "eval_runtime": 20.3165,
      "eval_samples_per_second": 5.267,
      "eval_steps_per_second": 0.689,
      "step": 20
    },
    {
      "epoch": 0.33070866141732286,
      "grad_norm": 0.596634030342102,
      "learning_rate": 8.247240241650918e-05,
      "loss": 8.0154,
      "step": 21
    },
    {
      "epoch": 0.3779527559055118,
      "grad_norm": 0.6523898839950562,
      "learning_rate": 7.269952498697734e-05,
      "loss": 8.1727,
      "step": 24
    },
    {
      "epoch": 0.3937007874015748,
      "eval_loss": 1.9660494327545166,
      "eval_runtime": 20.3194,
      "eval_samples_per_second": 5.266,
      "eval_steps_per_second": 0.689,
      "step": 25
    },
    {
      "epoch": 0.4251968503937008,
      "grad_norm": 0.6933566927909851,
      "learning_rate": 6.167226819279528e-05,
      "loss": 8.0485,
      "step": 27
    },
    {
      "epoch": 0.47244094488188976,
      "grad_norm": 0.5993256568908691,
      "learning_rate": 5e-05,
      "loss": 7.8586,
      "step": 30
    },
    {
      "epoch": 0.47244094488188976,
      "eval_loss": 1.9575363397598267,
      "eval_runtime": 20.3033,
      "eval_samples_per_second": 5.27,
      "eval_steps_per_second": 0.69,
      "step": 30
    },
    {
      "epoch": 0.5196850393700787,
      "grad_norm": 0.6470586657524109,
      "learning_rate": 3.832773180720475e-05,
      "loss": 7.9857,
      "step": 33
    },
    {
      "epoch": 0.5511811023622047,
      "eval_loss": 1.9521300792694092,
      "eval_runtime": 20.3174,
      "eval_samples_per_second": 5.266,
      "eval_steps_per_second": 0.689,
      "step": 35
    },
    {
      "epoch": 0.5669291338582677,
      "grad_norm": 0.6373382210731506,
      "learning_rate": 2.7300475013022663e-05,
      "loss": 7.7433,
      "step": 36
    },
    {
      "epoch": 0.6141732283464567,
      "grad_norm": 0.6322449445724487,
      "learning_rate": 1.7527597583490822e-05,
      "loss": 7.9635,
      "step": 39
    },
    {
      "epoch": 0.6299212598425197,
      "eval_loss": 1.9487825632095337,
      "eval_runtime": 20.3222,
      "eval_samples_per_second": 5.265,
      "eval_steps_per_second": 0.689,
      "step": 40
    },
    {
      "epoch": 0.6614173228346457,
      "grad_norm": 0.6913183927536011,
      "learning_rate": 9.549150281252633e-06,
      "loss": 7.8228,
      "step": 42
    },
    {
      "epoch": 0.7086614173228346,
      "grad_norm": 0.8218957781791687,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 7.9007,
      "step": 45
    },
    {
      "epoch": 0.7086614173228346,
      "eval_loss": 1.9470714330673218,
      "eval_runtime": 20.321,
      "eval_samples_per_second": 5.265,
      "eval_steps_per_second": 0.689,
      "step": 45
    }
  ],
  "logging_steps": 3,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.320224204783616e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}