trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.84,
  "eval_steps": 500,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "grad_norm": 18.83793067932129,
      "learning_rate": 0.00018333333333333334,
      "loss": 9.2102,
      "step": 2
    },
    {
      "epoch": 0.64,
      "grad_norm": 22.66363525390625,
      "learning_rate": 0.0001666666666666667,
      "loss": 8.2322,
      "step": 4
    },
    {
      "epoch": 0.96,
      "grad_norm": 22.428932189941406,
      "learning_rate": 0.00015000000000000001,
      "loss": 7.2138,
      "step": 6
    },
    {
      "epoch": 1.28,
      "grad_norm": 17.410696029663086,
      "learning_rate": 0.00014166666666666668,
      "loss": 6.6662,
      "step": 8
    },
    {
      "epoch": 1.6,
      "grad_norm": 12.254047393798828,
      "learning_rate": 0.000125,
      "loss": 6.6019,
      "step": 10
    },
    {
      "epoch": 1.92,
      "grad_norm": 10.925116539001465,
      "learning_rate": 0.00010833333333333333,
      "loss": 6.2901,
      "step": 12
    },
    {
      "epoch": 2.24,
      "grad_norm": 11.835088729858398,
      "learning_rate": 9.166666666666667e-05,
      "loss": 6.1701,
      "step": 14
    },
    {
      "epoch": 2.56,
      "grad_norm": 6.8969950675964355,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.0758,
      "step": 16
    },
    {
      "epoch": 2.88,
      "grad_norm": 10.971280097961426,
      "learning_rate": 5.833333333333334e-05,
      "loss": 5.953,
      "step": 18
    },
    {
      "epoch": 3.2,
      "grad_norm": 6.5526275634765625,
      "learning_rate": 4.166666666666667e-05,
      "loss": 6.1228,
      "step": 20
    },
    {
      "epoch": 3.52,
      "grad_norm": 4.4527459144592285,
      "learning_rate": 2.5e-05,
      "loss": 6.0307,
      "step": 22
    },
    {
      "epoch": 3.84,
      "grad_norm": 8.93628978729248,
      "learning_rate": 8.333333333333334e-06,
      "loss": 5.9651,
      "step": 24
    },
    {
      "epoch": 3.84,
      "step": 24,
      "total_flos": 110537516692488.0,
      "train_loss": 6.710986336072286,
      "train_runtime": 634.4855,
      "train_samples_per_second": 0.63,
      "train_steps_per_second": 0.038
    }
  ],
  "logging_steps": 2,
  "max_steps": 24,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 110537516692488.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}