trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9952,
  "eval_steps": 500,
  "global_step": 117,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.256,
      "grad_norm": 1.1756333916034116,
      "learning_rate": 1.9901811253364458e-05,
      "loss": 0.8679,
      "step": 10
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.9801789637439382,
      "learning_rate": 1.912783265061319e-05,
      "loss": 0.8248,
      "step": 20
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.4629662903978508,
      "learning_rate": 1.7640373758216075e-05,
      "loss": 0.801,
      "step": 30
    },
    {
      "epoch": 1.024,
      "grad_norm": 0.8499478192536458,
      "learning_rate": 1.5555702330196024e-05,
      "loss": 0.8661,
      "step": 40
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.2645993260229654,
      "learning_rate": 1.3036767451096148e-05,
      "loss": 0.7376,
      "step": 50
    },
    {
      "epoch": 1.536,
      "grad_norm": 0.2249901319915651,
      "learning_rate": 1.028046256275869e-05,
      "loss": 0.7305,
      "step": 60
    },
    {
      "epoch": 1.792,
      "grad_norm": 0.19268576535191254,
      "learning_rate": 7.50223521832773e-06,
      "loss": 0.7321,
      "step": 70
    },
    {
      "epoch": 2.048,
      "grad_norm": 0.23711409541260844,
      "learning_rate": 4.919246547534709e-06,
      "loss": 0.7845,
      "step": 80
    },
    {
      "epoch": 2.304,
      "grad_norm": 0.19527495984227403,
      "learning_rate": 2.7333967796597317e-06,
      "loss": 0.6911,
      "step": 90
    },
    {
      "epoch": 2.56,
      "grad_norm": 0.16811729495957195,
      "learning_rate": 1.1155436402112785e-06,
      "loss": 0.6961,
      "step": 100
    },
    {
      "epoch": 2.816,
      "grad_norm": 0.16659870380805858,
      "learning_rate": 1.921471959676957e-07,
      "loss": 0.6923,
      "step": 110
    },
    {
      "epoch": 2.9952,
      "step": 117,
      "total_flos": 831446906306560.0,
      "train_loss": 0.7610167682680309,
      "train_runtime": 19507.0925,
      "train_samples_per_second": 6.152,
      "train_steps_per_second": 0.006
    }
  ],
  "logging_steps": 10,
  "max_steps": 117,
  "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": 831446906306560.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}