{
  "best_metric": 0.12418080866336823,
  "best_model_checkpoint": "/content/drive/MyDrive/ml2/ML_project_2_course/weights1/checkpoint-700",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 788,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 1.3563565015792847,
      "learning_rate": 0.00019746192893401016,
      "loss": 0.6637,
      "step": 10
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.3717408180236816,
      "learning_rate": 0.00019492385786802033,
      "loss": 0.5792,
      "step": 20
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.0099543333053589,
      "learning_rate": 0.00019238578680203048,
      "loss": 0.5253,
      "step": 30
    },
    {
      "epoch": 0.2,
      "grad_norm": 5.920238494873047,
      "learning_rate": 0.0001898477157360406,
      "loss": 0.3762,
      "step": 40
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.513320207595825,
      "learning_rate": 0.00018730964467005077,
      "loss": 0.3987,
      "step": 50
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.2757742404937744,
      "learning_rate": 0.00018477157360406092,
      "loss": 0.3336,
      "step": 60
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.0954958200454712,
      "learning_rate": 0.00018223350253807107,
      "loss": 0.4492,
      "step": 70
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.1074856519699097,
      "learning_rate": 0.00017969543147208124,
      "loss": 0.3453,
      "step": 80
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.97078800201416,
      "learning_rate": 0.0001771573604060914,
      "loss": 0.3103,
      "step": 90
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.3375837802886963,
      "learning_rate": 0.0001746192893401015,
      "loss": 0.2034,
      "step": 100
    },
    {
      "epoch": 0.51,
      "eval_f1": 0.8307086614173228,
      "eval_loss": 0.5720954537391663,
      "eval_runtime": 5.4754,
      "eval_samples_per_second": 77.255,
      "eval_steps_per_second": 9.68,
      "step": 100
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.1420457363128662,
      "learning_rate": 0.00017208121827411168,
      "loss": 0.3871,
      "step": 110
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.5014958381652832,
      "learning_rate": 0.00016954314720812183,
      "loss": 0.242,
      "step": 120
    },
    {
      "epoch": 0.66,
      "grad_norm": 2.7499921321868896,
      "learning_rate": 0.00016700507614213198,
      "loss": 0.205,
      "step": 130
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.3920105695724487,
      "learning_rate": 0.00016446700507614215,
      "loss": 0.2539,
      "step": 140
    },
    {
      "epoch": 0.76,
      "grad_norm": 2.2138984203338623,
      "learning_rate": 0.0001619289340101523,
      "loss": 0.3,
      "step": 150
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.4511910080909729,
      "learning_rate": 0.00015939086294416242,
      "loss": 0.1906,
      "step": 160
    },
    {
      "epoch": 0.86,
      "grad_norm": 3.9145667552948,
      "learning_rate": 0.0001568527918781726,
      "loss": 0.2873,
      "step": 170
    },
    {
      "epoch": 0.91,
      "grad_norm": 7.277167797088623,
      "learning_rate": 0.00015431472081218274,
      "loss": 0.1775,
      "step": 180
    },
    {
      "epoch": 0.96,
      "grad_norm": 3.8661720752716064,
      "learning_rate": 0.0001517766497461929,
      "loss": 0.1682,
      "step": 190
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.29035553336143494,
      "learning_rate": 0.00014923857868020306,
      "loss": 0.201,
      "step": 200
    },
    {
      "epoch": 1.02,
      "eval_f1": 0.9207048458149779,
      "eval_loss": 0.2542865574359894,
      "eval_runtime": 5.3252,
      "eval_samples_per_second": 79.434,
      "eval_steps_per_second": 9.953,
      "step": 200
    },
    {
      "epoch": 1.07,
      "grad_norm": 7.454772472381592,
      "learning_rate": 0.0001467005076142132,
      "loss": 0.219,
      "step": 210
    },
    {
      "epoch": 1.12,
      "grad_norm": 2.4295766353607178,
      "learning_rate": 0.00014416243654822336,
      "loss": 0.1328,
      "step": 220
    },
    {
      "epoch": 1.17,
      "grad_norm": 2.6288206577301025,
      "learning_rate": 0.0001416243654822335,
      "loss": 0.1695,
      "step": 230
    },
    {
      "epoch": 1.22,
      "grad_norm": 1.8671033382415771,
      "learning_rate": 0.00013908629441624365,
      "loss": 0.1648,
      "step": 240
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.6234340667724609,
      "learning_rate": 0.0001365482233502538,
      "loss": 0.1318,
      "step": 250
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.7282599210739136,
      "learning_rate": 0.00013401015228426397,
      "loss": 0.1646,
      "step": 260
    },
    {
      "epoch": 1.37,
      "grad_norm": 8.562129020690918,
      "learning_rate": 0.00013147208121827412,
      "loss": 0.0765,
      "step": 270
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.10765184462070465,
      "learning_rate": 0.00012893401015228427,
      "loss": 0.1424,
      "step": 280
    },
    {
      "epoch": 1.47,
      "grad_norm": 1.785663366317749,
      "learning_rate": 0.00012639593908629442,
      "loss": 0.0838,
      "step": 290
    },
    {
      "epoch": 1.52,
      "grad_norm": 2.627047538757324,
      "learning_rate": 0.00012385786802030456,
      "loss": 0.2025,
      "step": 300
    },
    {
      "epoch": 1.52,
      "eval_f1": 0.9406392694063928,
      "eval_loss": 0.21563898026943207,
      "eval_runtime": 5.5211,
      "eval_samples_per_second": 76.615,
      "eval_steps_per_second": 9.599,
      "step": 300
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.36638501286506653,
      "learning_rate": 0.00012131979695431472,
      "loss": 0.1436,
      "step": 310
    },
    {
      "epoch": 1.62,
      "grad_norm": 2.350069761276245,
      "learning_rate": 0.00011878172588832489,
      "loss": 0.0701,
      "step": 320
    },
    {
      "epoch": 1.68,
      "grad_norm": 1.4287878274917603,
      "learning_rate": 0.00011624365482233503,
      "loss": 0.0843,
      "step": 330
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.32531386613845825,
      "learning_rate": 0.0001137055837563452,
      "loss": 0.1415,
      "step": 340
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.5714915990829468,
      "learning_rate": 0.00011116751269035533,
      "loss": 0.0751,
      "step": 350
    },
    {
      "epoch": 1.83,
      "grad_norm": 4.953071117401123,
      "learning_rate": 0.00010862944162436547,
      "loss": 0.126,
      "step": 360
    },
    {
      "epoch": 1.88,
      "grad_norm": 4.382538318634033,
      "learning_rate": 0.00010609137055837564,
      "loss": 0.1279,
      "step": 370
    },
    {
      "epoch": 1.93,
      "grad_norm": 2.4561574459075928,
      "learning_rate": 0.0001035532994923858,
      "loss": 0.0932,
      "step": 380
    },
    {
      "epoch": 1.98,
      "grad_norm": 1.9325863122940063,
      "learning_rate": 0.00010101522842639594,
      "loss": 0.1361,
      "step": 390
    },
    {
      "epoch": 2.03,
      "grad_norm": 0.1038457602262497,
      "learning_rate": 9.847715736040609e-05,
      "loss": 0.0564,
      "step": 400
    },
    {
      "epoch": 2.03,
      "eval_f1": 0.9452954048140045,
      "eval_loss": 0.1903565376996994,
      "eval_runtime": 6.5595,
      "eval_samples_per_second": 64.486,
      "eval_steps_per_second": 8.08,
      "step": 400
    },
    {
      "epoch": 2.08,
      "grad_norm": 0.12188208103179932,
      "learning_rate": 9.593908629441625e-05,
      "loss": 0.0947,
      "step": 410
    },
    {
      "epoch": 2.13,
      "grad_norm": 0.8643466234207153,
      "learning_rate": 9.34010152284264e-05,
      "loss": 0.058,
      "step": 420
    },
    {
      "epoch": 2.18,
      "grad_norm": 0.06271246075630188,
      "learning_rate": 9.086294416243655e-05,
      "loss": 0.065,
      "step": 430
    },
    {
      "epoch": 2.23,
      "grad_norm": 0.593055784702301,
      "learning_rate": 8.83248730964467e-05,
      "loss": 0.0585,
      "step": 440
    },
    {
      "epoch": 2.28,
      "grad_norm": 0.20611844956874847,
      "learning_rate": 8.578680203045685e-05,
      "loss": 0.0695,
      "step": 450
    },
    {
      "epoch": 2.34,
      "grad_norm": 0.09795264154672623,
      "learning_rate": 8.324873096446701e-05,
      "loss": 0.0351,
      "step": 460
    },
    {
      "epoch": 2.39,
      "grad_norm": 0.04693546146154404,
      "learning_rate": 8.071065989847716e-05,
      "loss": 0.0367,
      "step": 470
    },
    {
      "epoch": 2.44,
      "grad_norm": 0.09200535714626312,
      "learning_rate": 7.817258883248731e-05,
      "loss": 0.0266,
      "step": 480
    },
    {
      "epoch": 2.49,
      "grad_norm": 0.134303480386734,
      "learning_rate": 7.563451776649747e-05,
      "loss": 0.0306,
      "step": 490
    },
    {
      "epoch": 2.54,
      "grad_norm": 0.041602715849876404,
      "learning_rate": 7.309644670050762e-05,
      "loss": 0.0262,
      "step": 500
    },
    {
      "epoch": 2.54,
      "eval_f1": 0.9495412844036697,
      "eval_loss": 0.21375472843647003,
      "eval_runtime": 5.9305,
      "eval_samples_per_second": 71.326,
      "eval_steps_per_second": 8.937,
      "step": 500
    },
    {
      "epoch": 2.59,
      "grad_norm": 5.353679656982422,
      "learning_rate": 7.055837563451776e-05,
      "loss": 0.0615,
      "step": 510
    },
    {
      "epoch": 2.64,
      "grad_norm": 7.85841703414917,
      "learning_rate": 6.802030456852793e-05,
      "loss": 0.0927,
      "step": 520
    },
    {
      "epoch": 2.69,
      "grad_norm": 0.5664938688278198,
      "learning_rate": 6.548223350253807e-05,
      "loss": 0.1466,
      "step": 530
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.04644864797592163,
      "learning_rate": 6.294416243654822e-05,
      "loss": 0.0148,
      "step": 540
    },
    {
      "epoch": 2.79,
      "grad_norm": 0.043112464249134064,
      "learning_rate": 6.040609137055838e-05,
      "loss": 0.041,
      "step": 550
    },
    {
      "epoch": 2.84,
      "grad_norm": 0.9218811392784119,
      "learning_rate": 5.786802030456853e-05,
      "loss": 0.0998,
      "step": 560
    },
    {
      "epoch": 2.89,
      "grad_norm": 0.09559272974729538,
      "learning_rate": 5.532994923857868e-05,
      "loss": 0.0447,
      "step": 570
    },
    {
      "epoch": 2.94,
      "grad_norm": 0.04861769080162048,
      "learning_rate": 5.2791878172588836e-05,
      "loss": 0.0296,
      "step": 580
    },
    {
      "epoch": 2.99,
      "grad_norm": 0.13610342144966125,
      "learning_rate": 5.0253807106598984e-05,
      "loss": 0.037,
      "step": 590
    },
    {
      "epoch": 3.05,
      "grad_norm": 0.048563599586486816,
      "learning_rate": 4.771573604060914e-05,
      "loss": 0.0222,
      "step": 600
    },
    {
      "epoch": 3.05,
      "eval_f1": 0.9508196721311475,
      "eval_loss": 0.15546846389770508,
      "eval_runtime": 5.4832,
      "eval_samples_per_second": 77.145,
      "eval_steps_per_second": 9.666,
      "step": 600
    },
    {
      "epoch": 3.1,
      "grad_norm": 0.05200362578034401,
      "learning_rate": 4.517766497461929e-05,
      "loss": 0.0118,
      "step": 610
    },
    {
      "epoch": 3.15,
      "grad_norm": 1.2061017751693726,
      "learning_rate": 4.2639593908629446e-05,
      "loss": 0.0469,
      "step": 620
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.05518200248479843,
      "learning_rate": 4.010152284263959e-05,
      "loss": 0.0228,
      "step": 630
    },
    {
      "epoch": 3.25,
      "grad_norm": 0.19791868329048157,
      "learning_rate": 3.756345177664975e-05,
      "loss": 0.0184,
      "step": 640
    },
    {
      "epoch": 3.3,
      "grad_norm": 0.10160644352436066,
      "learning_rate": 3.50253807106599e-05,
      "loss": 0.0267,
      "step": 650
    },
    {
      "epoch": 3.35,
      "grad_norm": 0.04751751944422722,
      "learning_rate": 3.248730964467005e-05,
      "loss": 0.0096,
      "step": 660
    },
    {
      "epoch": 3.4,
      "grad_norm": 0.04644055664539337,
      "learning_rate": 2.9949238578680206e-05,
      "loss": 0.0367,
      "step": 670
    },
    {
      "epoch": 3.45,
      "grad_norm": 0.060281168669462204,
      "learning_rate": 2.7411167512690357e-05,
      "loss": 0.0068,
      "step": 680
    },
    {
      "epoch": 3.5,
      "grad_norm": 0.034235160797834396,
      "learning_rate": 2.4873096446700507e-05,
      "loss": 0.0398,
      "step": 690
    },
    {
      "epoch": 3.55,
      "grad_norm": 0.3630073368549347,
      "learning_rate": 2.233502538071066e-05,
      "loss": 0.0069,
      "step": 700
    },
    {
      "epoch": 3.55,
      "eval_f1": 0.963302752293578,
      "eval_loss": 0.12418080866336823,
      "eval_runtime": 6.5914,
      "eval_samples_per_second": 64.175,
      "eval_steps_per_second": 8.041,
      "step": 700
    },
    {
      "epoch": 3.6,
      "grad_norm": 0.040880296379327774,
      "learning_rate": 1.9796954314720815e-05,
      "loss": 0.0045,
      "step": 710
    },
    {
      "epoch": 3.65,
      "grad_norm": 0.5524014830589294,
      "learning_rate": 1.7258883248730966e-05,
      "loss": 0.04,
      "step": 720
    },
    {
      "epoch": 3.71,
      "grad_norm": 0.03459872677922249,
      "learning_rate": 1.4720812182741117e-05,
      "loss": 0.0131,
      "step": 730
    },
    {
      "epoch": 3.76,
      "grad_norm": 2.4353725910186768,
      "learning_rate": 1.218274111675127e-05,
      "loss": 0.0114,
      "step": 740
    },
    {
      "epoch": 3.81,
      "grad_norm": 0.049515530467033386,
      "learning_rate": 9.644670050761421e-06,
      "loss": 0.0163,
      "step": 750
    },
    {
      "epoch": 3.86,
      "grad_norm": 0.033101875334978104,
      "learning_rate": 7.106598984771575e-06,
      "loss": 0.0108,
      "step": 760
    },
    {
      "epoch": 3.91,
      "grad_norm": 0.08629802614450455,
      "learning_rate": 4.568527918781726e-06,
      "loss": 0.0435,
      "step": 770
    },
    {
      "epoch": 3.96,
      "grad_norm": 0.09244952350854874,
      "learning_rate": 2.030456852791878e-06,
      "loss": 0.0195,
      "step": 780
    },
    {
      "epoch": 4.0,
      "step": 788,
      "total_flos": 9.74539261393109e+17,
      "train_loss": 0.13476083650806833,
      "train_runtime": 481.1459,
      "train_samples_per_second": 26.138,
      "train_steps_per_second": 1.638
    }
  ],
  "logging_steps": 10,
  "max_steps": 788,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 100,
  "total_flos": 9.74539261393109e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}