last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00021064161435733244,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.1064161435733244e-05,
      "grad_norm": 7.505989074707031,
      "learning_rate": 2e-05,
      "loss": 35.5579,
      "step": 1
    },
    {
      "epoch": 2.1064161435733244e-05,
      "eval_loss": 8.20571517944336,
      "eval_runtime": 1210.7745,
      "eval_samples_per_second": 16.51,
      "eval_steps_per_second": 8.255,
      "step": 1
    },
    {
      "epoch": 4.212832287146649e-05,
      "grad_norm": 6.046815872192383,
      "learning_rate": 4e-05,
      "loss": 43.1733,
      "step": 2
    },
    {
      "epoch": 6.319248430719973e-05,
      "grad_norm": 6.785678386688232,
      "learning_rate": 6e-05,
      "loss": 31.3483,
      "step": 3
    },
    {
      "epoch": 6.319248430719973e-05,
      "eval_loss": 8.196510314941406,
      "eval_runtime": 1210.0563,
      "eval_samples_per_second": 16.52,
      "eval_steps_per_second": 8.26,
      "step": 3
    },
    {
      "epoch": 8.425664574293298e-05,
      "grad_norm": 8.55991268157959,
      "learning_rate": 8e-05,
      "loss": 35.9895,
      "step": 4
    },
    {
      "epoch": 0.00010532080717866622,
      "grad_norm": 7.0496110916137695,
      "learning_rate": 0.0001,
      "loss": 30.85,
      "step": 5
    },
    {
      "epoch": 0.00012638496861439946,
      "grad_norm": 7.777755260467529,
      "learning_rate": 0.00012,
      "loss": 33.1663,
      "step": 6
    },
    {
      "epoch": 0.00012638496861439946,
      "eval_loss": 8.057297706604004,
      "eval_runtime": 1210.3996,
      "eval_samples_per_second": 16.515,
      "eval_steps_per_second": 8.258,
      "step": 6
    },
    {
      "epoch": 0.0001474491300501327,
      "grad_norm": 8.272865295410156,
      "learning_rate": 0.00014,
      "loss": 27.6331,
      "step": 7
    },
    {
      "epoch": 0.00016851329148586595,
      "grad_norm": 10.752893447875977,
      "learning_rate": 0.00016,
      "loss": 27.2656,
      "step": 8
    },
    {
      "epoch": 0.0001895774529215992,
      "grad_norm": 14.83069133758545,
      "learning_rate": 0.00018,
      "loss": 26.1396,
      "step": 9
    },
    {
      "epoch": 0.0001895774529215992,
      "eval_loss": 6.5651936531066895,
      "eval_runtime": 1210.5612,
      "eval_samples_per_second": 16.513,
      "eval_steps_per_second": 8.257,
      "step": 9
    },
    {
      "epoch": 0.00021064161435733244,
      "grad_norm": 25.13823699951172,
      "learning_rate": 0.0002,
      "loss": 27.1915,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3737104102195200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}