{
  "best_metric": 1.3892344236373901,
  "best_model_checkpoint": "matsuo-lab-llm-competition-ug9pb742/checkpoint-312",
  "epoch": 3.0,
  "eval_steps": 39,
  "global_step": 390,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 0.41445887088775635,
      "learning_rate": 6.2224e-05,
      "loss": 1.6218,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.7263110876083374,
      "learning_rate": 6.058652631578948e-05,
      "loss": 1.7018,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 0.4063463509082794,
      "learning_rate": 5.894905263157895e-05,
      "loss": 1.6955,
      "step": 30
    },
    {
      "epoch": 0.3,
      "eval_loss": 1.521242618560791,
      "eval_runtime": 70.3814,
      "eval_samples_per_second": 3.282,
      "eval_steps_per_second": 0.412,
      "step": 39
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.25590476393699646,
      "learning_rate": 5.7311578947368427e-05,
      "loss": 1.4888,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 0.2585609257221222,
      "learning_rate": 5.56741052631579e-05,
      "loss": 1.4528,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.3065848648548126,
      "learning_rate": 5.4036631578947374e-05,
      "loss": 1.5525,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 0.2731223404407501,
      "learning_rate": 5.239915789473685e-05,
      "loss": 1.4864,
      "step": 70
    },
    {
      "epoch": 0.6,
      "eval_loss": 1.4685145616531372,
      "eval_runtime": 70.0638,
      "eval_samples_per_second": 3.297,
      "eval_steps_per_second": 0.414,
      "step": 78
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.3084161579608917,
      "learning_rate": 5.076168421052632e-05,
      "loss": 1.3792,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 0.3827728033065796,
      "learning_rate": 4.9124210526315794e-05,
      "loss": 1.5268,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.3320125937461853,
      "learning_rate": 4.748673684210527e-05,
      "loss": 1.5606,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 0.386620432138443,
      "learning_rate": 4.584926315789474e-05,
      "loss": 1.3499,
      "step": 110
    },
    {
      "epoch": 0.9,
      "eval_loss": 1.4356176853179932,
      "eval_runtime": 70.0699,
      "eval_samples_per_second": 3.297,
      "eval_steps_per_second": 0.414,
      "step": 117
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.48299941420555115,
      "learning_rate": 4.4211789473684215e-05,
      "loss": 1.4698,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.6564630270004272,
      "learning_rate": 4.257431578947369e-05,
      "loss": 1.6282,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 0.34602901339530945,
      "learning_rate": 4.093684210526316e-05,
      "loss": 1.3008,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 0.45014750957489014,
      "learning_rate": 3.9299368421052635e-05,
      "loss": 1.4049,
      "step": 150
    },
    {
      "epoch": 1.2,
      "eval_loss": 1.4226655960083008,
      "eval_runtime": 70.033,
      "eval_samples_per_second": 3.298,
      "eval_steps_per_second": 0.414,
      "step": 156
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 0.5505159497261047,
      "learning_rate": 3.766189473684211e-05,
      "loss": 1.5217,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 0.4141960144042969,
      "learning_rate": 3.602442105263158e-05,
      "loss": 1.2895,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 0.4828188717365265,
      "learning_rate": 3.4386947368421056e-05,
      "loss": 1.3107,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 0.5664470195770264,
      "learning_rate": 3.274947368421053e-05,
      "loss": 1.4455,
      "step": 190
    },
    {
      "epoch": 1.5,
      "eval_loss": 1.413093090057373,
      "eval_runtime": 69.9748,
      "eval_samples_per_second": 3.301,
      "eval_steps_per_second": 0.414,
      "step": 195
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.46674856543540955,
      "learning_rate": 3.1112e-05,
      "loss": 1.3529,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 0.5345956683158875,
      "learning_rate": 2.9474526315789477e-05,
      "loss": 1.2989,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 0.6872637271881104,
      "learning_rate": 2.783705263157895e-05,
      "loss": 1.4185,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 0.49507564306259155,
      "learning_rate": 2.6199578947368424e-05,
      "loss": 1.4658,
      "step": 230
    },
    {
      "epoch": 1.8,
      "eval_loss": 1.4004093408584595,
      "eval_runtime": 69.6383,
      "eval_samples_per_second": 3.317,
      "eval_steps_per_second": 0.416,
      "step": 234
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 0.5477748513221741,
      "learning_rate": 2.4562105263157897e-05,
      "loss": 1.2469,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 0.6340446472167969,
      "learning_rate": 2.292463157894737e-05,
      "loss": 1.3764,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1060491800308228,
      "learning_rate": 2.1287157894736844e-05,
      "loss": 1.4648,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 0.5065295100212097,
      "learning_rate": 1.9649684210526318e-05,
      "loss": 1.196,
      "step": 270
    },
    {
      "epoch": 2.1,
      "eval_loss": 1.3942419290542603,
      "eval_runtime": 70.0887,
      "eval_samples_per_second": 3.296,
      "eval_steps_per_second": 0.414,
      "step": 273
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 0.6876117587089539,
      "learning_rate": 1.801221052631579e-05,
      "loss": 1.3281,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 0.7879146337509155,
      "learning_rate": 1.6374736842105265e-05,
      "loss": 1.4138,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.6207026839256287,
      "learning_rate": 1.4737263157894738e-05,
      "loss": 1.203,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 0.6773239374160767,
      "learning_rate": 1.3099789473684212e-05,
      "loss": 1.2431,
      "step": 310
    },
    {
      "epoch": 2.4,
      "eval_loss": 1.3892344236373901,
      "eval_runtime": 70.3067,
      "eval_samples_per_second": 3.286,
      "eval_steps_per_second": 0.412,
      "step": 312
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 0.8685129880905151,
      "learning_rate": 1.1462315789473685e-05,
      "loss": 1.3195,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 0.5892221927642822,
      "learning_rate": 9.824842105263159e-06,
      "loss": 1.3394,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.7276835441589355,
      "learning_rate": 8.187368421052632e-06,
      "loss": 1.2584,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 0.8118067979812622,
      "learning_rate": 6.549894736842106e-06,
      "loss": 1.3422,
      "step": 350
    },
    {
      "epoch": 2.7,
      "eval_loss": 1.3936101198196411,
      "eval_runtime": 70.3959,
      "eval_samples_per_second": 3.281,
      "eval_steps_per_second": 0.412,
      "step": 351
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.7172791957855225,
      "learning_rate": 4.912421052631579e-06,
      "loss": 1.3558,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 0.7048629522323608,
      "learning_rate": 3.274947368421053e-06,
      "loss": 1.1974,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.763253927230835,
      "learning_rate": 1.6374736842105265e-06,
      "loss": 1.3164,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.4231680631637573,
      "learning_rate": 0.0,
      "loss": 1.3841,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.3893393278121948,
      "eval_runtime": 70.0743,
      "eval_samples_per_second": 3.297,
      "eval_steps_per_second": 0.414,
      "step": 390
    }
  ],
  "logging_steps": 10,
  "max_steps": 390,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 39,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9017979426870272e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}