trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6957,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21561017680034497,
      "grad_norm": 0.09970881789922714,
      "learning_rate": 4.784389823199656e-05,
      "loss": 0.2857,
      "step": 500
    },
    {
      "epoch": 0.43122035360068994,
      "grad_norm": 0.11449956893920898,
      "learning_rate": 4.5687796463993104e-05,
      "loss": 0.1398,
      "step": 1000
    },
    {
      "epoch": 0.6468305304010349,
      "grad_norm": 0.1402336210012436,
      "learning_rate": 4.353169469598966e-05,
      "loss": 0.1241,
      "step": 1500
    },
    {
      "epoch": 0.8624407072013799,
      "grad_norm": 0.144794300198555,
      "learning_rate": 4.13755929279862e-05,
      "loss": 0.114,
      "step": 2000
    },
    {
      "epoch": 1.0780508840017249,
      "grad_norm": 0.15507571399211884,
      "learning_rate": 3.921949115998275e-05,
      "loss": 0.102,
      "step": 2500
    },
    {
      "epoch": 1.2936610608020698,
      "grad_norm": 0.1842988282442093,
      "learning_rate": 3.70633893919793e-05,
      "loss": 0.0933,
      "step": 3000
    },
    {
      "epoch": 1.5092712376024149,
      "grad_norm": 0.16924935579299927,
      "learning_rate": 3.4907287623975855e-05,
      "loss": 0.0878,
      "step": 3500
    },
    {
      "epoch": 1.7248814144027598,
      "grad_norm": 0.1321273148059845,
      "learning_rate": 3.27511858559724e-05,
      "loss": 0.0835,
      "step": 4000
    },
    {
      "epoch": 1.9404915912031049,
      "grad_norm": 0.17047573626041412,
      "learning_rate": 3.0595084087968956e-05,
      "loss": 0.0805,
      "step": 4500
    },
    {
      "epoch": 2.1561017680034498,
      "grad_norm": 0.15277768671512604,
      "learning_rate": 2.84389823199655e-05,
      "loss": 0.0751,
      "step": 5000
    },
    {
      "epoch": 2.3717119448037947,
      "grad_norm": 0.15870638191699982,
      "learning_rate": 2.628288055196205e-05,
      "loss": 0.0728,
      "step": 5500
    },
    {
      "epoch": 2.5873221216041395,
      "grad_norm": 0.23912857472896576,
      "learning_rate": 2.4126778783958602e-05,
      "loss": 0.0714,
      "step": 6000
    },
    {
      "epoch": 2.8029322984044844,
      "grad_norm": 0.14251725375652313,
      "learning_rate": 2.1970677015955153e-05,
      "loss": 0.07,
      "step": 6500
    }
  ],
  "logging_steps": 500,
  "max_steps": 11595,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.29100420056064e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}