trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 8145,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.061387354205033766,
      "grad_norm": 14.549583435058594,
      "learning_rate": 6.137981831573779e-06,
      "loss": 5.8577,
      "step": 500
    },
    {
      "epoch": 0.12277470841006753,
      "grad_norm": 17.52524185180664,
      "learning_rate": 1.2275963663147557e-05,
      "loss": 3.557,
      "step": 1000
    },
    {
      "epoch": 0.1841620626151013,
      "grad_norm": 8.65881061553955,
      "learning_rate": 1.8413945494721336e-05,
      "loss": 1.6768,
      "step": 1500
    },
    {
      "epoch": 0.24554941682013506,
      "grad_norm": 13.167006492614746,
      "learning_rate": 2.4551927326295114e-05,
      "loss": 1.1943,
      "step": 2000
    },
    {
      "epoch": 0.3069367710251688,
      "grad_norm": 11.799457550048828,
      "learning_rate": 3.068990915786889e-05,
      "loss": 1.0562,
      "step": 2500
    },
    {
      "epoch": 0.3683241252302026,
      "grad_norm": 9.352530479431152,
      "learning_rate": 3.682789098944267e-05,
      "loss": 0.9976,
      "step": 3000
    },
    {
      "epoch": 0.42971147943523635,
      "grad_norm": 10.113947868347168,
      "learning_rate": 4.296587282101645e-05,
      "loss": 0.9335,
      "step": 3500
    },
    {
      "epoch": 0.4910988336402701,
      "grad_norm": 16.53941535949707,
      "learning_rate": 4.910385465259023e-05,
      "loss": 0.9027,
      "step": 4000
    },
    {
      "epoch": 0.5524861878453039,
      "grad_norm": 11.04137134552002,
      "learning_rate": 4.972407821445649e-05,
      "loss": 0.9036,
      "step": 4500
    },
    {
      "epoch": 0.6138735420503376,
      "grad_norm": 7.330798149108887,
      "learning_rate": 4.9400984788761525e-05,
      "loss": 0.8831,
      "step": 5000
    },
    {
      "epoch": 0.6752608962553714,
      "grad_norm": 7.521310329437256,
      "learning_rate": 4.9077891363066546e-05,
      "loss": 0.8716,
      "step": 5500
    },
    {
      "epoch": 0.7366482504604052,
      "grad_norm": 6.216422080993652,
      "learning_rate": 4.8754797937371574e-05,
      "loss": 0.8635,
      "step": 6000
    },
    {
      "epoch": 0.7980356046654389,
      "grad_norm": 7.46917200088501,
      "learning_rate": 4.84317045116766e-05,
      "loss": 0.8184,
      "step": 6500
    },
    {
      "epoch": 0.8594229588704727,
      "grad_norm": 12.236076354980469,
      "learning_rate": 4.810861108598162e-05,
      "loss": 0.7897,
      "step": 7000
    },
    {
      "epoch": 0.9208103130755064,
      "grad_norm": 4.151940822601318,
      "learning_rate": 4.778551766028665e-05,
      "loss": 0.7685,
      "step": 7500
    },
    {
      "epoch": 0.9821976672805403,
      "grad_norm": 6.2876362800598145,
      "learning_rate": 4.746242423459168e-05,
      "loss": 0.7761,
      "step": 8000
    },
    {
      "epoch": 1.0,
      "eval_HasAns_exact": 76.61943319838056,
      "eval_HasAns_f1": 83.0111719601073,
      "eval_HasAns_total": 5928,
      "eval_NoAns_exact": 85.95458368376788,
      "eval_NoAns_f1": 85.95458368376788,
      "eval_NoAns_total": 5945,
      "eval_best_exact": 81.29369156910637,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 84.48498503996589,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 81.29369156910637,
      "eval_f1": 84.48498503996598,
      "eval_runtime": 40.9091,
      "eval_samples_per_second": 290.229,
      "eval_steps_per_second": 18.162,
      "eval_total": 11873,
      "step": 8145
    }
  ],
  "logging_steps": 500,
  "max_steps": 81450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 4.417078415257805e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}