run-10/checkpoint-3180/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 3180,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9968553459119497,
      "grad_norm": 5.211723804473877,
      "learning_rate": 1.8006289308176103e-05,
      "loss": 6.0788,
      "step": 317
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7067741935483871,
      "eval_loss": 4.211826324462891,
      "eval_runtime": 1.7417,
      "eval_samples_per_second": 1779.886,
      "eval_steps_per_second": 37.32,
      "step": 318
    },
    {
      "epoch": 1.9937106918238994,
      "grad_norm": 5.270339488983154,
      "learning_rate": 1.6012578616352204e-05,
      "loss": 3.1783,
      "step": 634
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8548387096774194,
      "eval_loss": 1.911292314529419,
      "eval_runtime": 1.8073,
      "eval_samples_per_second": 1715.276,
      "eval_steps_per_second": 35.965,
      "step": 636
    },
    {
      "epoch": 2.990566037735849,
      "grad_norm": 5.103645324707031,
      "learning_rate": 1.4018867924528304e-05,
      "loss": 1.5075,
      "step": 951
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9087096774193548,
      "eval_loss": 1.0654947757720947,
      "eval_runtime": 1.8019,
      "eval_samples_per_second": 1720.413,
      "eval_steps_per_second": 36.073,
      "step": 954
    },
    {
      "epoch": 3.9874213836477987,
      "grad_norm": 3.8782594203948975,
      "learning_rate": 1.2025157232704403e-05,
      "loss": 0.8843,
      "step": 1268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9332258064516129,
      "eval_loss": 0.8456443548202515,
      "eval_runtime": 1.7851,
      "eval_samples_per_second": 1736.633,
      "eval_steps_per_second": 36.413,
      "step": 1272
    },
    {
      "epoch": 4.984276729559748,
      "grad_norm": 3.119723081588745,
      "learning_rate": 1.0031446540880504e-05,
      "loss": 0.6849,
      "step": 1585
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9380645161290323,
      "eval_loss": 0.7831763625144958,
      "eval_runtime": 1.79,
      "eval_samples_per_second": 1731.853,
      "eval_steps_per_second": 36.313,
      "step": 1590
    },
    {
      "epoch": 5.981132075471698,
      "grad_norm": 3.575148582458496,
      "learning_rate": 8.037735849056606e-06,
      "loss": 0.606,
      "step": 1902
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9425806451612904,
      "eval_loss": 0.756099283695221,
      "eval_runtime": 1.7871,
      "eval_samples_per_second": 1734.629,
      "eval_steps_per_second": 36.371,
      "step": 1908
    },
    {
      "epoch": 6.977987421383648,
      "grad_norm": 1.812427043914795,
      "learning_rate": 6.044025157232704e-06,
      "loss": 0.5681,
      "step": 2219
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9451612903225807,
      "eval_loss": 0.7347410321235657,
      "eval_runtime": 1.8217,
      "eval_samples_per_second": 1701.678,
      "eval_steps_per_second": 35.68,
      "step": 2226
    },
    {
      "epoch": 7.9748427672955975,
      "grad_norm": 1.113770842552185,
      "learning_rate": 4.0503144654088055e-06,
      "loss": 0.5453,
      "step": 2536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9438709677419355,
      "eval_loss": 0.722730278968811,
      "eval_runtime": 1.78,
      "eval_samples_per_second": 1741.528,
      "eval_steps_per_second": 36.516,
      "step": 2544
    },
    {
      "epoch": 8.971698113207546,
      "grad_norm": 1.670234203338623,
      "learning_rate": 2.056603773584906e-06,
      "loss": 0.5339,
      "step": 2853
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9451612903225807,
      "eval_loss": 0.7155605554580688,
      "eval_runtime": 2.8971,
      "eval_samples_per_second": 1070.04,
      "eval_steps_per_second": 22.436,
      "step": 2862
    },
    {
      "epoch": 9.968553459119496,
      "grad_norm": 1.3198106288909912,
      "learning_rate": 6.289308176100629e-08,
      "loss": 0.527,
      "step": 3170
    }
  ],
  "logging_steps": 317,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000000000.0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 825254092458012.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7585301269848748,
    "num_train_epochs": 10,
    "temperature": 2
  }
}