trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 138,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14492753623188406,
      "grad_norm": 12.398261352422882,
      "learning_rate": 6.4285714285714295e-06,
      "loss": 2.9235,
      "step": 10
    },
    {
      "epoch": 0.2898550724637681,
      "grad_norm": 8.017807501158464,
      "learning_rate": 9.959935885253715e-06,
      "loss": 1.4926,
      "step": 20
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 7.6729398768152,
      "learning_rate": 9.643264997861312e-06,
      "loss": 1.099,
      "step": 30
    },
    {
      "epoch": 0.5797101449275363,
      "grad_norm": 7.937653371039124,
      "learning_rate": 9.030141317270026e-06,
      "loss": 1.0419,
      "step": 40
    },
    {
      "epoch": 0.7246376811594203,
      "grad_norm": 6.935420300985396,
      "learning_rate": 8.15971019223152e-06,
      "loss": 0.9559,
      "step": 50
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 6.521364461671462,
      "learning_rate": 7.087544961425317e-06,
      "loss": 0.8724,
      "step": 60
    },
    {
      "epoch": 1.0144927536231885,
      "grad_norm": 4.940342489815436,
      "learning_rate": 5.882098831289044e-06,
      "loss": 0.8656,
      "step": 70
    },
    {
      "epoch": 1.1594202898550725,
      "grad_norm": 6.488774654663568,
      "learning_rate": 4.62033442887377e-06,
      "loss": 0.5261,
      "step": 80
    },
    {
      "epoch": 1.3043478260869565,
      "grad_norm": 5.384402568372393,
      "learning_rate": 3.3828100642538097e-06,
      "loss": 0.5177,
      "step": 90
    },
    {
      "epoch": 1.4492753623188406,
      "grad_norm": 5.522607349461902,
      "learning_rate": 2.2485364238130435e-06,
      "loss": 0.4972,
      "step": 100
    },
    {
      "epoch": 1.5942028985507246,
      "grad_norm": 5.956590886271073,
      "learning_rate": 1.2899320727454472e-06,
      "loss": 0.4665,
      "step": 110
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 5.162066297238004,
      "learning_rate": 5.681998365579594e-07,
      "loss": 0.4959,
      "step": 120
    },
    {
      "epoch": 1.8840579710144927,
      "grad_norm": 5.884961224410103,
      "learning_rate": 1.2941926002306536e-07,
      "loss": 0.4616,
      "step": 130
    },
    {
      "epoch": 2.0,
      "step": 138,
      "total_flos": 3235348660224.0,
      "train_loss": 0.9176704987235691,
      "train_runtime": 291.5306,
      "train_samples_per_second": 3.766,
      "train_steps_per_second": 0.473
    }
  ],
  "logging_steps": 10,
  "max_steps": 138,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3235348660224.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}