trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.25161290322580643,
  "eval_steps": 500,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03225806451612903,
      "grad_norm": 2.242180824279785,
      "learning_rate": 1.9959742939952393e-05,
      "loss": 0.492,
      "step": 10
    },
    {
      "epoch": 0.06451612903225806,
      "grad_norm": 0.8940081000328064,
      "learning_rate": 1.8584487936018663e-05,
      "loss": 0.1381,
      "step": 20
    },
    {
      "epoch": 0.0967741935483871,
      "grad_norm": 0.5671939849853516,
      "learning_rate": 1.5508969814521026e-05,
      "loss": 0.088,
      "step": 30
    },
    {
      "epoch": 0.12903225806451613,
      "grad_norm": 0.44803526997566223,
      "learning_rate": 1.1342332658176556e-05,
      "loss": 0.0739,
      "step": 40
    },
    {
      "epoch": 0.16129032258064516,
      "grad_norm": 0.34039735794067383,
      "learning_rate": 6.909830056250527e-06,
      "loss": 0.0676,
      "step": 50
    },
    {
      "epoch": 0.1935483870967742,
      "grad_norm": 0.3448280394077301,
      "learning_rate": 3.089373510131354e-06,
      "loss": 0.0611,
      "step": 60
    },
    {
      "epoch": 0.22580645161290322,
      "grad_norm": 0.3139958679676056,
      "learning_rate": 6.37651293602628e-07,
      "loss": 0.0583,
      "step": 70
    },
    {
      "epoch": 0.25161290322580643,
      "step": 78,
      "total_flos": 1.6179959995301888e+17,
      "train_loss": 0.13258298620199546,
      "train_runtime": 2183.8276,
      "train_samples_per_second": 2.264,
      "train_steps_per_second": 0.036
    }
  ],
  "logging_steps": 10,
  "max_steps": 78,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 1.6179959995301888e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}