trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 7900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.31645569620253167,
      "grad_norm": 16.740053176879883,
      "learning_rate": 2.8101265822784812e-05,
      "loss": 1.7983,
      "step": 500
    },
    {
      "epoch": 0.6329113924050633,
      "grad_norm": 18.301847457885742,
      "learning_rate": 2.620253164556962e-05,
      "loss": 1.1858,
      "step": 1000
    },
    {
      "epoch": 0.9493670886075949,
      "grad_norm": 19.28789710998535,
      "learning_rate": 2.430379746835443e-05,
      "loss": 1.0645,
      "step": 1500
    },
    {
      "epoch": 1.2658227848101267,
      "grad_norm": 27.41464614868164,
      "learning_rate": 2.240506329113924e-05,
      "loss": 0.7401,
      "step": 2000
    },
    {
      "epoch": 1.5822784810126582,
      "grad_norm": 22.102096557617188,
      "learning_rate": 2.050632911392405e-05,
      "loss": 0.7248,
      "step": 2500
    },
    {
      "epoch": 1.8987341772151898,
      "grad_norm": 29.267433166503906,
      "learning_rate": 1.860759493670886e-05,
      "loss": 0.7217,
      "step": 3000
    },
    {
      "epoch": 2.2151898734177213,
      "grad_norm": 14.63181209564209,
      "learning_rate": 1.670886075949367e-05,
      "loss": 0.4984,
      "step": 3500
    },
    {
      "epoch": 2.5316455696202533,
      "grad_norm": 15.814390182495117,
      "learning_rate": 1.4810126582278482e-05,
      "loss": 0.4188,
      "step": 4000
    },
    {
      "epoch": 2.848101265822785,
      "grad_norm": 14.612494468688965,
      "learning_rate": 1.2911392405063291e-05,
      "loss": 0.4127,
      "step": 4500
    },
    {
      "epoch": 3.1645569620253164,
      "grad_norm": 39.51222229003906,
      "learning_rate": 1.1012658227848103e-05,
      "loss": 0.3182,
      "step": 5000
    },
    {
      "epoch": 3.481012658227848,
      "grad_norm": 13.681534767150879,
      "learning_rate": 9.113924050632912e-06,
      "loss": 0.2268,
      "step": 5500
    },
    {
      "epoch": 3.7974683544303796,
      "grad_norm": 11.261092185974121,
      "learning_rate": 7.215189873417722e-06,
      "loss": 0.2346,
      "step": 6000
    },
    {
      "epoch": 4.113924050632911,
      "grad_norm": 19.033397674560547,
      "learning_rate": 5.3164556962025316e-06,
      "loss": 0.1939,
      "step": 6500
    },
    {
      "epoch": 4.430379746835443,
      "grad_norm": 25.222740173339844,
      "learning_rate": 3.4177215189873417e-06,
      "loss": 0.1312,
      "step": 7000
    },
    {
      "epoch": 4.746835443037975,
      "grad_norm": 19.743125915527344,
      "learning_rate": 1.518987341772152e-06,
      "loss": 0.1167,
      "step": 7500
    },
    {
      "epoch": 5.0,
      "step": 7900,
      "total_flos": 1.85781994039296e+16,
      "train_loss": 0.5625110177148747,
      "train_runtime": 892.8635,
      "train_samples_per_second": 106.175,
      "train_steps_per_second": 8.848
    }
  ],
  "logging_steps": 500,
  "max_steps": 7900,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.85781994039296e+16,
  "train_batch_size": 12,
  "trial_name": null,
  "trial_params": null
}