trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9767441860465116,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15503875968992248,
      "grad_norm": 0.7852384448051453,
      "learning_rate": 1.75e-05,
      "loss": 2.506,
      "step": 10
    },
    {
      "epoch": 0.31007751937984496,
      "grad_norm": 0.5620294809341431,
      "learning_rate": 4.25e-05,
      "loss": 1.9499,
      "step": 20
    },
    {
      "epoch": 0.46511627906976744,
      "grad_norm": 0.818453848361969,
      "learning_rate": 4.9795940299380575e-05,
      "loss": 1.8121,
      "step": 30
    },
    {
      "epoch": 0.6201550387596899,
      "grad_norm": 0.5289535522460938,
      "learning_rate": 4.880447529310118e-05,
      "loss": 1.789,
      "step": 40
    },
    {
      "epoch": 0.7751937984496124,
      "grad_norm": 0.8725740313529968,
      "learning_rate": 4.723355372206297e-05,
      "loss": 1.7512,
      "step": 50
    },
    {
      "epoch": 0.9302325581395349,
      "grad_norm": 0.7308263182640076,
      "learning_rate": 4.478741221073136e-05,
      "loss": 1.6186,
      "step": 60
    },
    {
      "epoch": 1.0852713178294573,
      "grad_norm": 0.7901429533958435,
      "learning_rate": 4.1682970280555986e-05,
      "loss": 1.6679,
      "step": 70
    },
    {
      "epoch": 1.2403100775193798,
      "grad_norm": 0.4563349783420563,
      "learning_rate": 3.8023508512198256e-05,
      "loss": 1.266,
      "step": 80
    },
    {
      "epoch": 1.3953488372093024,
      "grad_norm": 0.5261668562889099,
      "learning_rate": 3.393077224502832e-05,
      "loss": 1.3869,
      "step": 90
    },
    {
      "epoch": 1.550387596899225,
      "grad_norm": 0.8815342783927917,
      "learning_rate": 2.954092127448591e-05,
      "loss": 1.2376,
      "step": 100
    },
    {
      "epoch": 1.7054263565891472,
      "grad_norm": 0.727622926235199,
      "learning_rate": 2.5e-05,
      "loss": 1.291,
      "step": 110
    },
    {
      "epoch": 1.8604651162790697,
      "grad_norm": 0.6443896293640137,
      "learning_rate": 2.0459078725514092e-05,
      "loss": 1.3153,
      "step": 120
    },
    {
      "epoch": 2.0155038759689923,
      "grad_norm": 0.7863492965698242,
      "learning_rate": 1.6069227754971683e-05,
      "loss": 1.3498,
      "step": 130
    },
    {
      "epoch": 2.1705426356589146,
      "grad_norm": 0.9250425696372986,
      "learning_rate": 1.1976491487801748e-05,
      "loss": 0.9415,
      "step": 140
    },
    {
      "epoch": 2.3255813953488373,
      "grad_norm": 0.9809682965278625,
      "learning_rate": 8.317029719444016e-06,
      "loss": 0.7475,
      "step": 150
    },
    {
      "epoch": 2.4806201550387597,
      "grad_norm": 0.8305571675300598,
      "learning_rate": 5.2125877892686496e-06,
      "loss": 0.6603,
      "step": 160
    },
    {
      "epoch": 2.6356589147286824,
      "grad_norm": 0.6964967846870422,
      "learning_rate": 2.7664462779370293e-06,
      "loss": 0.6403,
      "step": 170
    },
    {
      "epoch": 2.7906976744186047,
      "grad_norm": 1.0790668725967407,
      "learning_rate": 1.0599850022898539e-06,
      "loss": 0.8335,
      "step": 180
    },
    {
      "epoch": 2.945736434108527,
      "grad_norm": 0.8265672326087952,
      "learning_rate": 1.4997561900135238e-07,
      "loss": 0.6972,
      "step": 190
    },
    {
      "epoch": 2.9767441860465116,
      "step": 192,
      "total_flos": 1.1519499230222746e+17,
      "train_loss": 1.3342496789991856,
      "train_runtime": 2508.9029,
      "train_samples_per_second": 0.617,
      "train_steps_per_second": 0.077
    }
  ],
  "logging_steps": 10,
  "max_steps": 192,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1519499230222746e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}