{
  "best_metric": 0.6857386848847139,
  "best_model_checkpoint": "vet-sm/checkpoint-1035",
  "epoch": 4.9879518072289155,
  "eval_steps": 500,
  "global_step": 1035,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 1.454635500907898,
      "learning_rate": 4.807692307692308e-06,
      "loss": 2.0699,
      "step": 10
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.6190325021743774,
      "learning_rate": 9.615384615384616e-06,
      "loss": 2.0402,
      "step": 20
    },
    {
      "epoch": 0.14,
      "grad_norm": 1.361482858657837,
      "learning_rate": 1.4423076923076923e-05,
      "loss": 1.9857,
      "step": 30
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.3599398136138916,
      "learning_rate": 1.923076923076923e-05,
      "loss": 1.9319,
      "step": 40
    },
    {
      "epoch": 0.24,
      "grad_norm": 1.6369996070861816,
      "learning_rate": 2.4038461538461542e-05,
      "loss": 1.8814,
      "step": 50
    },
    {
      "epoch": 0.29,
      "grad_norm": 1.192236065864563,
      "learning_rate": 2.8846153846153845e-05,
      "loss": 1.8459,
      "step": 60
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.7909432649612427,
      "learning_rate": 3.365384615384616e-05,
      "loss": 1.8128,
      "step": 70
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.8163617849349976,
      "learning_rate": 3.846153846153846e-05,
      "loss": 1.7424,
      "step": 80
    },
    {
      "epoch": 0.43,
      "grad_norm": 2.151453733444214,
      "learning_rate": 4.326923076923077e-05,
      "loss": 1.6964,
      "step": 90
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.8711392879486084,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 1.6367,
      "step": 100
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.7750871181488037,
      "learning_rate": 4.967776584317938e-05,
      "loss": 1.6045,
      "step": 110
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.7674392461776733,
      "learning_rate": 4.9140708915145005e-05,
      "loss": 1.626,
      "step": 120
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.8520715236663818,
      "learning_rate": 4.860365198711064e-05,
      "loss": 1.5904,
      "step": 130
    },
    {
      "epoch": 0.67,
      "grad_norm": 2.2221274375915527,
      "learning_rate": 4.806659505907626e-05,
      "loss": 1.5845,
      "step": 140
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.5263822078704834,
      "learning_rate": 4.7529538131041896e-05,
      "loss": 1.5236,
      "step": 150
    },
    {
      "epoch": 0.77,
      "grad_norm": 3.3485989570617676,
      "learning_rate": 4.699248120300752e-05,
      "loss": 1.4698,
      "step": 160
    },
    {
      "epoch": 0.82,
      "grad_norm": 2.506840944290161,
      "learning_rate": 4.645542427497315e-05,
      "loss": 1.4167,
      "step": 170
    },
    {
      "epoch": 0.87,
      "grad_norm": 3.2644572257995605,
      "learning_rate": 4.591836734693878e-05,
      "loss": 1.4426,
      "step": 180
    },
    {
      "epoch": 0.92,
      "grad_norm": 2.8360097408294678,
      "learning_rate": 4.5381310418904406e-05,
      "loss": 1.3924,
      "step": 190
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.6382710933685303,
      "learning_rate": 4.484425349087004e-05,
      "loss": 1.3437,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5456874466268147,
      "eval_loss": 1.3443350791931152,
      "eval_runtime": 402.2174,
      "eval_samples_per_second": 2.911,
      "eval_steps_per_second": 0.184,
      "step": 207
    },
    {
      "epoch": 1.01,
      "grad_norm": 3.3896210193634033,
      "learning_rate": 4.4307196562835664e-05,
      "loss": 1.413,
      "step": 210
    },
    {
      "epoch": 1.06,
      "grad_norm": 6.438076972961426,
      "learning_rate": 4.3770139634801297e-05,
      "loss": 1.3462,
      "step": 220
    },
    {
      "epoch": 1.11,
      "grad_norm": 3.4236090183258057,
      "learning_rate": 4.323308270676692e-05,
      "loss": 1.2821,
      "step": 230
    },
    {
      "epoch": 1.16,
      "grad_norm": 1.8709990978240967,
      "learning_rate": 4.269602577873255e-05,
      "loss": 1.2731,
      "step": 240
    },
    {
      "epoch": 1.2,
      "grad_norm": 2.8092925548553467,
      "learning_rate": 4.215896885069818e-05,
      "loss": 1.2207,
      "step": 250
    },
    {
      "epoch": 1.25,
      "grad_norm": 3.147840976715088,
      "learning_rate": 4.1621911922663806e-05,
      "loss": 1.2131,
      "step": 260
    },
    {
      "epoch": 1.3,
      "grad_norm": 3.0452332496643066,
      "learning_rate": 4.108485499462943e-05,
      "loss": 1.1956,
      "step": 270
    },
    {
      "epoch": 1.35,
      "grad_norm": 3.2367162704467773,
      "learning_rate": 4.054779806659506e-05,
      "loss": 1.1639,
      "step": 280
    },
    {
      "epoch": 1.4,
      "grad_norm": 3.8542873859405518,
      "learning_rate": 4.0010741138560684e-05,
      "loss": 1.1577,
      "step": 290
    },
    {
      "epoch": 1.45,
      "grad_norm": 2.901261568069458,
      "learning_rate": 3.9473684210526316e-05,
      "loss": 1.1535,
      "step": 300
    },
    {
      "epoch": 1.49,
      "grad_norm": 4.736692905426025,
      "learning_rate": 3.893662728249194e-05,
      "loss": 1.1863,
      "step": 310
    },
    {
      "epoch": 1.54,
      "grad_norm": 2.820009708404541,
      "learning_rate": 3.8399570354457575e-05,
      "loss": 1.2104,
      "step": 320
    },
    {
      "epoch": 1.59,
      "grad_norm": 4.279903888702393,
      "learning_rate": 3.78625134264232e-05,
      "loss": 1.162,
      "step": 330
    },
    {
      "epoch": 1.64,
      "grad_norm": 4.341091156005859,
      "learning_rate": 3.732545649838883e-05,
      "loss": 1.1758,
      "step": 340
    },
    {
      "epoch": 1.69,
      "grad_norm": 3.7722089290618896,
      "learning_rate": 3.678839957035446e-05,
      "loss": 1.153,
      "step": 350
    },
    {
      "epoch": 1.73,
      "grad_norm": 2.462113857269287,
      "learning_rate": 3.6251342642320084e-05,
      "loss": 1.1653,
      "step": 360
    },
    {
      "epoch": 1.78,
      "grad_norm": 3.2860846519470215,
      "learning_rate": 3.571428571428572e-05,
      "loss": 1.0895,
      "step": 370
    },
    {
      "epoch": 1.83,
      "grad_norm": 4.360082626342773,
      "learning_rate": 3.517722878625134e-05,
      "loss": 1.0896,
      "step": 380
    },
    {
      "epoch": 1.88,
      "grad_norm": 3.952023506164551,
      "learning_rate": 3.4640171858216975e-05,
      "loss": 1.1883,
      "step": 390
    },
    {
      "epoch": 1.93,
      "grad_norm": 4.6793742179870605,
      "learning_rate": 3.41031149301826e-05,
      "loss": 1.0039,
      "step": 400
    },
    {
      "epoch": 1.98,
      "grad_norm": 5.783022403717041,
      "learning_rate": 3.3566058002148234e-05,
      "loss": 1.0892,
      "step": 410
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.627668659265585,
      "eval_loss": 1.0833462476730347,
      "eval_runtime": 386.7346,
      "eval_samples_per_second": 3.028,
      "eval_steps_per_second": 0.191,
      "step": 415
    },
    {
      "epoch": 2.02,
      "grad_norm": 2.7940785884857178,
      "learning_rate": 3.302900107411386e-05,
      "loss": 1.0866,
      "step": 420
    },
    {
      "epoch": 2.07,
      "grad_norm": 2.254930257797241,
      "learning_rate": 3.2491944146079485e-05,
      "loss": 0.9175,
      "step": 430
    },
    {
      "epoch": 2.12,
      "grad_norm": 4.631977558135986,
      "learning_rate": 3.195488721804512e-05,
      "loss": 0.8898,
      "step": 440
    },
    {
      "epoch": 2.17,
      "grad_norm": 4.229618549346924,
      "learning_rate": 3.1417830290010743e-05,
      "loss": 0.9022,
      "step": 450
    },
    {
      "epoch": 2.22,
      "grad_norm": 2.8936877250671387,
      "learning_rate": 3.0880773361976376e-05,
      "loss": 0.902,
      "step": 460
    },
    {
      "epoch": 2.27,
      "grad_norm": 3.5928826332092285,
      "learning_rate": 3.0343716433942e-05,
      "loss": 0.8681,
      "step": 470
    },
    {
      "epoch": 2.31,
      "grad_norm": 3.8170666694641113,
      "learning_rate": 2.980665950590763e-05,
      "loss": 0.7539,
      "step": 480
    },
    {
      "epoch": 2.36,
      "grad_norm": 4.207719326019287,
      "learning_rate": 2.9269602577873257e-05,
      "loss": 0.9404,
      "step": 490
    },
    {
      "epoch": 2.41,
      "grad_norm": 5.372262001037598,
      "learning_rate": 2.8732545649838882e-05,
      "loss": 0.8323,
      "step": 500
    },
    {
      "epoch": 2.46,
      "grad_norm": 4.851659297943115,
      "learning_rate": 2.8195488721804515e-05,
      "loss": 0.8612,
      "step": 510
    },
    {
      "epoch": 2.51,
      "grad_norm": 4.7718682289123535,
      "learning_rate": 2.765843179377014e-05,
      "loss": 0.8427,
      "step": 520
    },
    {
      "epoch": 2.55,
      "grad_norm": 5.901302337646484,
      "learning_rate": 2.712137486573577e-05,
      "loss": 0.8489,
      "step": 530
    },
    {
      "epoch": 2.6,
      "grad_norm": 5.132293224334717,
      "learning_rate": 2.6584317937701396e-05,
      "loss": 0.8522,
      "step": 540
    },
    {
      "epoch": 2.65,
      "grad_norm": 4.27319860458374,
      "learning_rate": 2.6047261009667025e-05,
      "loss": 0.8314,
      "step": 550
    },
    {
      "epoch": 2.7,
      "grad_norm": 5.610922336578369,
      "learning_rate": 2.5510204081632654e-05,
      "loss": 0.8804,
      "step": 560
    },
    {
      "epoch": 2.75,
      "grad_norm": 2.8495535850524902,
      "learning_rate": 2.4973147153598283e-05,
      "loss": 0.7591,
      "step": 570
    },
    {
      "epoch": 2.8,
      "grad_norm": 4.183429718017578,
      "learning_rate": 2.443609022556391e-05,
      "loss": 0.8702,
      "step": 580
    },
    {
      "epoch": 2.84,
      "grad_norm": 5.251755237579346,
      "learning_rate": 2.3899033297529538e-05,
      "loss": 0.8275,
      "step": 590
    },
    {
      "epoch": 2.89,
      "grad_norm": 4.696440696716309,
      "learning_rate": 2.3361976369495167e-05,
      "loss": 0.8131,
      "step": 600
    },
    {
      "epoch": 2.94,
      "grad_norm": 3.233407735824585,
      "learning_rate": 2.2824919441460796e-05,
      "loss": 0.7995,
      "step": 610
    },
    {
      "epoch": 2.99,
      "grad_norm": 6.206390857696533,
      "learning_rate": 2.2287862513426426e-05,
      "loss": 0.883,
      "step": 620
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6567036720751495,
      "eval_loss": 0.99440997838974,
      "eval_runtime": 391.9131,
      "eval_samples_per_second": 2.988,
      "eval_steps_per_second": 0.189,
      "step": 622
    },
    {
      "epoch": 3.04,
      "grad_norm": 3.8997249603271484,
      "learning_rate": 2.1750805585392055e-05,
      "loss": 0.7696,
      "step": 630
    },
    {
      "epoch": 3.08,
      "grad_norm": 3.5104353427886963,
      "learning_rate": 2.121374865735768e-05,
      "loss": 0.6644,
      "step": 640
    },
    {
      "epoch": 3.13,
      "grad_norm": 4.623937129974365,
      "learning_rate": 2.067669172932331e-05,
      "loss": 0.7019,
      "step": 650
    },
    {
      "epoch": 3.18,
      "grad_norm": 3.191859722137451,
      "learning_rate": 2.0139634801288935e-05,
      "loss": 0.6541,
      "step": 660
    },
    {
      "epoch": 3.23,
      "grad_norm": 3.7715423107147217,
      "learning_rate": 1.9602577873254565e-05,
      "loss": 0.5893,
      "step": 670
    },
    {
      "epoch": 3.28,
      "grad_norm": 3.1906309127807617,
      "learning_rate": 1.9065520945220194e-05,
      "loss": 0.5868,
      "step": 680
    },
    {
      "epoch": 3.33,
      "grad_norm": 6.5846099853515625,
      "learning_rate": 1.8528464017185823e-05,
      "loss": 0.6708,
      "step": 690
    },
    {
      "epoch": 3.37,
      "grad_norm": 3.970404863357544,
      "learning_rate": 1.7991407089151452e-05,
      "loss": 0.6117,
      "step": 700
    },
    {
      "epoch": 3.42,
      "grad_norm": 4.3821845054626465,
      "learning_rate": 1.7454350161117078e-05,
      "loss": 0.6104,
      "step": 710
    },
    {
      "epoch": 3.47,
      "grad_norm": 5.36595344543457,
      "learning_rate": 1.6917293233082707e-05,
      "loss": 0.5853,
      "step": 720
    },
    {
      "epoch": 3.52,
      "grad_norm": 4.7466301918029785,
      "learning_rate": 1.6380236305048336e-05,
      "loss": 0.5884,
      "step": 730
    },
    {
      "epoch": 3.57,
      "grad_norm": 3.9577629566192627,
      "learning_rate": 1.5843179377013965e-05,
      "loss": 0.5961,
      "step": 740
    },
    {
      "epoch": 3.61,
      "grad_norm": 5.096789360046387,
      "learning_rate": 1.5306122448979594e-05,
      "loss": 0.6155,
      "step": 750
    },
    {
      "epoch": 3.66,
      "grad_norm": 6.176750183105469,
      "learning_rate": 1.4769065520945222e-05,
      "loss": 0.5,
      "step": 760
    },
    {
      "epoch": 3.71,
      "grad_norm": 3.7734062671661377,
      "learning_rate": 1.4232008592910851e-05,
      "loss": 0.5895,
      "step": 770
    },
    {
      "epoch": 3.76,
      "grad_norm": 3.7732903957366943,
      "learning_rate": 1.3694951664876477e-05,
      "loss": 0.5221,
      "step": 780
    },
    {
      "epoch": 3.81,
      "grad_norm": 5.260674953460693,
      "learning_rate": 1.3157894736842106e-05,
      "loss": 0.6303,
      "step": 790
    },
    {
      "epoch": 3.86,
      "grad_norm": 4.469279766082764,
      "learning_rate": 1.2620837808807733e-05,
      "loss": 0.6151,
      "step": 800
    },
    {
      "epoch": 3.9,
      "grad_norm": 4.325742244720459,
      "learning_rate": 1.2083780880773363e-05,
      "loss": 0.5958,
      "step": 810
    },
    {
      "epoch": 3.95,
      "grad_norm": 3.5742740631103516,
      "learning_rate": 1.1546723952738992e-05,
      "loss": 0.5177,
      "step": 820
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.369570732116699,
      "learning_rate": 1.100966702470462e-05,
      "loss": 0.5199,
      "step": 830
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6754910333048676,
      "eval_loss": 0.9295239448547363,
      "eval_runtime": 398.874,
      "eval_samples_per_second": 2.936,
      "eval_steps_per_second": 0.186,
      "step": 830
    },
    {
      "epoch": 4.05,
      "grad_norm": 4.601716041564941,
      "learning_rate": 1.0472610096670248e-05,
      "loss": 0.4864,
      "step": 840
    },
    {
      "epoch": 4.1,
      "grad_norm": 2.6591596603393555,
      "learning_rate": 9.935553168635876e-06,
      "loss": 0.4057,
      "step": 850
    },
    {
      "epoch": 4.14,
      "grad_norm": 4.309298038482666,
      "learning_rate": 9.398496240601503e-06,
      "loss": 0.4228,
      "step": 860
    },
    {
      "epoch": 4.19,
      "grad_norm": 4.057296276092529,
      "learning_rate": 8.861439312567132e-06,
      "loss": 0.4299,
      "step": 870
    },
    {
      "epoch": 4.24,
      "grad_norm": 3.2583279609680176,
      "learning_rate": 8.324382384532762e-06,
      "loss": 0.4938,
      "step": 880
    },
    {
      "epoch": 4.29,
      "grad_norm": 4.6558451652526855,
      "learning_rate": 7.787325456498389e-06,
      "loss": 0.4269,
      "step": 890
    },
    {
      "epoch": 4.34,
      "grad_norm": 3.8659329414367676,
      "learning_rate": 7.250268528464017e-06,
      "loss": 0.4189,
      "step": 900
    },
    {
      "epoch": 4.39,
      "grad_norm": 2.4085872173309326,
      "learning_rate": 6.713211600429646e-06,
      "loss": 0.4376,
      "step": 910
    },
    {
      "epoch": 4.43,
      "grad_norm": 4.497287750244141,
      "learning_rate": 6.176154672395274e-06,
      "loss": 0.4347,
      "step": 920
    },
    {
      "epoch": 4.48,
      "grad_norm": 5.4745259284973145,
      "learning_rate": 5.639097744360902e-06,
      "loss": 0.3766,
      "step": 930
    },
    {
      "epoch": 4.53,
      "grad_norm": 3.8638925552368164,
      "learning_rate": 5.102040816326531e-06,
      "loss": 0.4002,
      "step": 940
    },
    {
      "epoch": 4.58,
      "grad_norm": 2.668515682220459,
      "learning_rate": 4.564983888292159e-06,
      "loss": 0.4154,
      "step": 950
    },
    {
      "epoch": 4.63,
      "grad_norm": 2.7704601287841797,
      "learning_rate": 4.027926960257788e-06,
      "loss": 0.377,
      "step": 960
    },
    {
      "epoch": 4.67,
      "grad_norm": 4.1865949630737305,
      "learning_rate": 3.490870032223416e-06,
      "loss": 0.375,
      "step": 970
    },
    {
      "epoch": 4.72,
      "grad_norm": 4.669921398162842,
      "learning_rate": 2.9538131041890443e-06,
      "loss": 0.3799,
      "step": 980
    },
    {
      "epoch": 4.77,
      "grad_norm": 4.289856433868408,
      "learning_rate": 2.4167561761546726e-06,
      "loss": 0.3652,
      "step": 990
    },
    {
      "epoch": 4.82,
      "grad_norm": 2.9988179206848145,
      "learning_rate": 1.8796992481203007e-06,
      "loss": 0.389,
      "step": 1000
    },
    {
      "epoch": 4.87,
      "grad_norm": 1.7334405183792114,
      "learning_rate": 1.3426423200859292e-06,
      "loss": 0.3715,
      "step": 1010
    },
    {
      "epoch": 4.92,
      "grad_norm": 4.97418737411499,
      "learning_rate": 8.055853920515575e-07,
      "loss": 0.4106,
      "step": 1020
    },
    {
      "epoch": 4.96,
      "grad_norm": 6.089245319366455,
      "learning_rate": 2.6852846401718585e-07,
      "loss": 0.4526,
      "step": 1030
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.6857386848847139,
      "eval_loss": 0.9283789396286011,
      "eval_runtime": 402.142,
      "eval_samples_per_second": 2.912,
      "eval_steps_per_second": 0.184,
      "step": 1035
    },
    {
      "epoch": 4.99,
      "step": 1035,
      "total_flos": 2.5651227911307264e+18,
      "train_loss": 0.9461189188243111,
      "train_runtime": 36633.1283,
      "train_samples_per_second": 0.906,
      "train_steps_per_second": 0.028
    }
  ],
  "logging_steps": 10,
  "max_steps": 1035,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 2.5651227911307264e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}