{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.2,
  "eval_steps": 1000,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002,
      "grad_norm": 0.8063191175460815,
      "learning_rate": 7.920000000000001e-05,
      "loss": 8.2537,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 0.6161019206047058,
      "learning_rate": 0.00015920000000000002,
      "loss": 6.5731,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 0.8347169160842896,
      "learning_rate": 0.00023920000000000001,
      "loss": 6.1639,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 0.9150793552398682,
      "learning_rate": 0.0003192,
      "loss": 5.8682,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.6698479652404785,
      "learning_rate": 0.0003992,
      "loss": 5.5708,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 0.5323518514633179,
      "learning_rate": 0.00047920000000000005,
      "loss": 5.3157,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 0.7472386956214905,
      "learning_rate": 0.0005592,
      "loss": 5.0907,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 0.6033297181129456,
      "learning_rate": 0.0006392,
      "loss": 4.8805,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 0.497113436460495,
      "learning_rate": 0.0007191999999999999,
      "loss": 4.7207,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.47698545455932617,
      "learning_rate": 0.0007992,
      "loss": 4.6068,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_accuracy": 0.2703913894324853,
      "eval_loss": 4.5053935050964355,
      "eval_runtime": 13.0992,
      "eval_samples_per_second": 76.341,
      "eval_steps_per_second": 0.611,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 0.44341349601745605,
      "learning_rate": 0.0008792,
      "loss": 4.4904,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 0.4383724331855774,
      "learning_rate": 0.0009592000000000001,
      "loss": 4.439,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 0.3894674777984619,
      "learning_rate": 0.0010391999999999999,
      "loss": 4.3735,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 0.41903820633888245,
      "learning_rate": 0.0011192,
      "loss": 4.3406,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.36601418256759644,
      "learning_rate": 0.0011992,
      "loss": 4.2879,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.4282030165195465,
      "learning_rate": 0.0012791999999999999,
      "loss": 4.2625,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 0.3431214690208435,
      "learning_rate": 0.0013592,
      "loss": 4.2316,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 0.3303460478782654,
      "learning_rate": 0.0014392,
      "loss": 4.1976,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 0.3030431866645813,
      "learning_rate": 0.0015192,
      "loss": 4.1494,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.3076256513595581,
      "learning_rate": 0.0015992,
      "loss": 4.1688,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.30658512720156555,
      "eval_loss": 4.094621658325195,
      "eval_runtime": 12.4723,
      "eval_samples_per_second": 80.178,
      "eval_steps_per_second": 0.641,
      "step": 2000
    },
    {
      "epoch": 0.042,
      "grad_norm": 0.2820931077003479,
      "learning_rate": 0.0016792,
      "loss": 4.5201,
      "step": 2100
    },
    {
      "epoch": 0.044,
      "grad_norm": 0.23724448680877686,
      "learning_rate": 0.0017592,
      "loss": 4.1333,
      "step": 2200
    },
    {
      "epoch": 0.046,
      "grad_norm": 0.2325439602136612,
      "learning_rate": 0.0018392,
      "loss": 4.1072,
      "step": 2300
    },
    {
      "epoch": 0.048,
      "grad_norm": 0.23736131191253662,
      "learning_rate": 0.0019192,
      "loss": 4.0682,
      "step": 2400
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.2313733696937561,
      "learning_rate": 0.0019992,
      "loss": 4.0571,
      "step": 2500
    },
    {
      "epoch": 0.052,
      "grad_norm": 0.22318924963474274,
      "learning_rate": 0.001999978563623903,
      "loss": 4.0153,
      "step": 2600
    },
    {
      "epoch": 0.054,
      "grad_norm": 0.20275355875492096,
      "learning_rate": 0.0019999133871331223,
      "loss": 4.0163,
      "step": 2700
    },
    {
      "epoch": 0.056,
      "grad_norm": 0.21306632459163666,
      "learning_rate": 0.0019998044711915177,
      "loss": 3.9777,
      "step": 2800
    },
    {
      "epoch": 0.058,
      "grad_norm": 0.19690750539302826,
      "learning_rate": 0.0019996518205634257,
      "loss": 3.9603,
      "step": 2900
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.20464760065078735,
      "learning_rate": 0.0019994554419262797,
      "loss": 3.9566,
      "step": 3000
    },
    {
      "epoch": 0.06,
      "eval_accuracy": 0.3226555772994129,
      "eval_loss": 3.9112932682037354,
      "eval_runtime": 11.7171,
      "eval_samples_per_second": 85.345,
      "eval_steps_per_second": 0.683,
      "step": 3000
    },
    {
      "epoch": 0.062,
      "grad_norm": 0.18151508271694183,
      "learning_rate": 0.001999215343870317,
      "loss": 3.9501,
      "step": 3100
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.18748600780963898,
      "learning_rate": 0.0019989315368982054,
      "loss": 3.9228,
      "step": 3200
    },
    {
      "epoch": 0.066,
      "grad_norm": 0.17895972728729248,
      "learning_rate": 0.0019986040334245797,
      "loss": 3.9041,
      "step": 3300
    },
    {
      "epoch": 0.068,
      "grad_norm": 0.18976759910583496,
      "learning_rate": 0.001998232847775504,
      "loss": 3.9123,
      "step": 3400
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1871727555990219,
      "learning_rate": 0.0019978179961878404,
      "loss": 3.8846,
      "step": 3500
    },
    {
      "epoch": 0.072,
      "grad_norm": 0.17986708879470825,
      "learning_rate": 0.001997359496808541,
      "loss": 3.8794,
      "step": 3600
    },
    {
      "epoch": 0.074,
      "grad_norm": 0.17090782523155212,
      "learning_rate": 0.001996857369693855,
      "loss": 3.8593,
      "step": 3700
    },
    {
      "epoch": 0.076,
      "grad_norm": 0.19015273451805115,
      "learning_rate": 0.0019963116368084486,
      "loss": 3.8642,
      "step": 3800
    },
    {
      "epoch": 0.078,
      "grad_norm": 0.1871633529663086,
      "learning_rate": 0.001995722322024446,
      "loss": 3.8479,
      "step": 3900
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.18049727380275726,
      "learning_rate": 0.001995089451120385,
      "loss": 3.8226,
      "step": 4000
    },
    {
      "epoch": 0.08,
      "eval_accuracy": 0.33194520547945205,
      "eval_loss": 3.8037452697753906,
      "eval_runtime": 12.6853,
      "eval_samples_per_second": 78.832,
      "eval_steps_per_second": 0.631,
      "step": 4000
    },
    {
      "epoch": 0.082,
      "grad_norm": 0.14800746738910675,
      "learning_rate": 0.0019944130517800893,
      "loss": 3.8319,
      "step": 4100
    },
    {
      "epoch": 0.084,
      "grad_norm": 0.14713813364505768,
      "learning_rate": 0.001993693153591457,
      "loss": 3.8319,
      "step": 4200
    },
    {
      "epoch": 0.086,
      "grad_norm": 0.17019635438919067,
      "learning_rate": 0.0019929297880451674,
      "loss": 3.8095,
      "step": 4300
    },
    {
      "epoch": 0.088,
      "grad_norm": 0.15243591368198395,
      "learning_rate": 0.0019921229885333023,
      "loss": 3.8003,
      "step": 4400
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.16134986281394958,
      "learning_rate": 0.001991272790347886,
      "loss": 3.814,
      "step": 4500
    },
    {
      "epoch": 0.092,
      "grad_norm": 0.16883811354637146,
      "learning_rate": 0.0019903792306793415,
      "loss": 3.7822,
      "step": 4600
    },
    {
      "epoch": 0.094,
      "grad_norm": 0.15718473494052887,
      "learning_rate": 0.001989442348614863,
      "loss": 3.7902,
      "step": 4700
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.14631035923957825,
      "learning_rate": 0.0019884621851367075,
      "loss": 3.7805,
      "step": 4800
    },
    {
      "epoch": 0.098,
      "grad_norm": 0.14771369099617004,
      "learning_rate": 0.001987438783120401,
      "loss": 3.7778,
      "step": 4900
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.1536298543214798,
      "learning_rate": 0.001986372187332862,
      "loss": 3.7848,
      "step": 5000
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.340252446183953,
      "eval_loss": 3.7290549278259277,
      "eval_runtime": 11.9375,
      "eval_samples_per_second": 83.77,
      "eval_steps_per_second": 0.67,
      "step": 5000
    },
    {
      "epoch": 0.102,
      "grad_norm": 0.15110976994037628,
      "learning_rate": 0.0019852624444304467,
      "loss": 3.7502,
      "step": 5100
    },
    {
      "epoch": 0.104,
      "grad_norm": 0.15862922370433807,
      "learning_rate": 0.0019841096029569044,
      "loss": 3.761,
      "step": 5200
    },
    {
      "epoch": 0.106,
      "grad_norm": 0.13948658108711243,
      "learning_rate": 0.0019829137133412556,
      "loss": 3.7591,
      "step": 5300
    },
    {
      "epoch": 0.108,
      "grad_norm": 0.148148313164711,
      "learning_rate": 0.001981674827895587,
      "loss": 3.7385,
      "step": 5400
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.13170774281024933,
      "learning_rate": 0.00198039300081276,
      "loss": 3.7424,
      "step": 5500
    },
    {
      "epoch": 0.112,
      "grad_norm": 0.14352479577064514,
      "learning_rate": 0.0019790682881640448,
      "loss": 3.7423,
      "step": 5600
    },
    {
      "epoch": 0.114,
      "grad_norm": 0.12865835428237915,
      "learning_rate": 0.001977700747896664,
      "loss": 3.7263,
      "step": 5700
    },
    {
      "epoch": 0.116,
      "grad_norm": 0.15470698475837708,
      "learning_rate": 0.001976290439831259,
      "loss": 3.7297,
      "step": 5800
    },
    {
      "epoch": 0.118,
      "grad_norm": 0.1353532075881958,
      "learning_rate": 0.0019748374256592736,
      "loss": 3.7218,
      "step": 5900
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.14982788264751434,
      "learning_rate": 0.0019733417689402543,
      "loss": 3.7164,
      "step": 6000
    },
    {
      "epoch": 0.12,
      "eval_accuracy": 0.34496673189823873,
      "eval_loss": 3.6746184825897217,
      "eval_runtime": 11.9904,
      "eval_samples_per_second": 83.4,
      "eval_steps_per_second": 0.667,
      "step": 6000
    },
    {
      "epoch": 0.122,
      "grad_norm": 0.14931881427764893,
      "learning_rate": 0.0019718035350990712,
      "loss": 3.7169,
      "step": 6100
    },
    {
      "epoch": 0.124,
      "grad_norm": 0.14251381158828735,
      "learning_rate": 0.0019702227914230566,
      "loss": 3.6929,
      "step": 6200
    },
    {
      "epoch": 0.126,
      "grad_norm": 0.14251849055290222,
      "learning_rate": 0.001968599607059059,
      "loss": 3.7018,
      "step": 6300
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.15696577727794647,
      "learning_rate": 0.0019669340530104207,
      "loss": 3.7034,
      "step": 6400
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.14737001061439514,
      "learning_rate": 0.001965226202133872,
      "loss": 3.6889,
      "step": 6500
    },
    {
      "epoch": 0.132,
      "grad_norm": 0.13667502999305725,
      "learning_rate": 0.0019634761291363427,
      "loss": 3.6935,
      "step": 6600
    },
    {
      "epoch": 0.134,
      "grad_norm": 0.13390734791755676,
      "learning_rate": 0.0019616839105716954,
      "loss": 3.7032,
      "step": 6700
    },
    {
      "epoch": 0.136,
      "grad_norm": 0.14834214746952057,
      "learning_rate": 0.0019598496248373755,
      "loss": 3.6657,
      "step": 6800
    },
    {
      "epoch": 0.138,
      "grad_norm": 0.1314767599105835,
      "learning_rate": 0.001957973352170984,
      "loss": 3.673,
      "step": 6900
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.15161661803722382,
      "learning_rate": 0.001956055174646765,
      "loss": 3.6713,
      "step": 7000
    },
    {
      "epoch": 0.14,
      "eval_accuracy": 0.3484422700587084,
      "eval_loss": 3.6349546909332275,
      "eval_runtime": 12.1265,
      "eval_samples_per_second": 82.464,
      "eval_steps_per_second": 0.66,
      "step": 7000
    },
    {
      "epoch": 0.142,
      "grad_norm": 0.12710200250148773,
      "learning_rate": 0.0019540951761720174,
      "loss": 3.67,
      "step": 7100
    },
    {
      "epoch": 0.144,
      "grad_norm": 0.18498213589191437,
      "learning_rate": 0.0019520934424834247,
      "loss": 3.675,
      "step": 7200
    },
    {
      "epoch": 0.146,
      "grad_norm": 0.13697299361228943,
      "learning_rate": 0.0019500500611433025,
      "loss": 3.6497,
      "step": 7300
    },
    {
      "epoch": 0.148,
      "grad_norm": 0.13290269672870636,
      "learning_rate": 0.0019479651215357707,
      "loss": 3.662,
      "step": 7400
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.12744054198265076,
      "learning_rate": 0.0019458387148628417,
      "loss": 3.6645,
      "step": 7500
    },
    {
      "epoch": 0.152,
      "grad_norm": 0.13697493076324463,
      "learning_rate": 0.001943670934140432,
      "loss": 3.6374,
      "step": 7600
    },
    {
      "epoch": 0.154,
      "grad_norm": 0.14067181944847107,
      "learning_rate": 0.0019414618741942936,
      "loss": 3.6453,
      "step": 7700
    },
    {
      "epoch": 0.156,
      "grad_norm": 0.1316055953502655,
      "learning_rate": 0.0019392116316558638,
      "loss": 3.6679,
      "step": 7800
    },
    {
      "epoch": 0.158,
      "grad_norm": 0.1261526495218277,
      "learning_rate": 0.001936920304958042,
      "loss": 3.6343,
      "step": 7900
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.13644501566886902,
      "learning_rate": 0.0019345879943308804,
      "loss": 3.6365,
      "step": 8000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.35229158512720155,
      "eval_loss": 3.5980424880981445,
      "eval_runtime": 17.6098,
      "eval_samples_per_second": 56.787,
      "eval_steps_per_second": 0.454,
      "step": 8000
    },
    {
      "epoch": 0.162,
      "grad_norm": 0.13911907374858856,
      "learning_rate": 0.0019322148017972016,
      "loss": 3.646,
      "step": 8100
    },
    {
      "epoch": 0.164,
      "grad_norm": 0.1297680288553238,
      "learning_rate": 0.001929800831168135,
      "loss": 3.6412,
      "step": 8200
    },
    {
      "epoch": 0.166,
      "grad_norm": 0.14803367853164673,
      "learning_rate": 0.001927346188038576,
      "loss": 3.6298,
      "step": 8300
    },
    {
      "epoch": 0.168,
      "grad_norm": 0.1325535923242569,
      "learning_rate": 0.0019248509797825672,
      "loss": 3.6067,
      "step": 8400
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.16948860883712769,
      "learning_rate": 0.0019223153155486009,
      "loss": 3.6284,
      "step": 8500
    },
    {
      "epoch": 0.172,
      "grad_norm": 0.12973898649215698,
      "learning_rate": 0.0019197393062548454,
      "loss": 3.627,
      "step": 8600
    },
    {
      "epoch": 0.174,
      "grad_norm": 0.1576964110136032,
      "learning_rate": 0.0019171230645842923,
      "loss": 3.6027,
      "step": 8700
    },
    {
      "epoch": 0.176,
      "grad_norm": 0.15394999086856842,
      "learning_rate": 0.0019144667049798272,
      "loss": 3.6098,
      "step": 8800
    },
    {
      "epoch": 0.178,
      "grad_norm": 0.12594962120056152,
      "learning_rate": 0.0019117703436392253,
      "loss": 3.6221,
      "step": 8900
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.1420104056596756,
      "learning_rate": 0.001909034098510066,
      "loss": 3.5993,
      "step": 9000
    },
    {
      "epoch": 0.18,
      "eval_accuracy": 0.3549628180039139,
      "eval_loss": 3.570964813232422,
      "eval_runtime": 12.4303,
      "eval_samples_per_second": 80.449,
      "eval_steps_per_second": 0.644,
      "step": 9000
    },
    {
      "epoch": 0.182,
      "grad_norm": 0.12720872461795807,
      "learning_rate": 0.001906258089284576,
      "loss": 3.5913,
      "step": 9100
    },
    {
      "epoch": 0.184,
      "grad_norm": 0.14890190958976746,
      "learning_rate": 0.0019034424373943915,
      "loss": 3.6113,
      "step": 9200
    },
    {
      "epoch": 0.186,
      "grad_norm": 0.15023267269134521,
      "learning_rate": 0.0019005872660052478,
      "loss": 3.6011,
      "step": 9300
    },
    {
      "epoch": 0.188,
      "grad_norm": 0.1637164056301117,
      "learning_rate": 0.001897692700011591,
      "loss": 3.5932,
      "step": 9400
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.13102982938289642,
      "learning_rate": 0.0018947588660311143,
      "loss": 3.5889,
      "step": 9500
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.16267850995063782,
      "learning_rate": 0.0018917858923992211,
      "loss": 3.5939,
      "step": 9600
    },
    {
      "epoch": 0.194,
      "grad_norm": 0.12798982858657837,
      "learning_rate": 0.0018887739091634085,
      "loss": 3.5896,
      "step": 9700
    },
    {
      "epoch": 0.196,
      "grad_norm": 0.13872383534908295,
      "learning_rate": 0.0018857230480775807,
      "loss": 3.574,
      "step": 9800
    },
    {
      "epoch": 0.198,
      "grad_norm": 0.11993825435638428,
      "learning_rate": 0.0018826334425962855,
      "loss": 3.589,
      "step": 9900
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.13074541091918945,
      "learning_rate": 0.0018795052278688753,
      "loss": 3.5828,
      "step": 10000
    },
    {
      "epoch": 0.2,
      "eval_accuracy": 0.35816046966731896,
      "eval_loss": 3.540745973587036,
      "eval_runtime": 17.8387,
      "eval_samples_per_second": 56.058,
      "eval_steps_per_second": 0.448,
      "step": 10000
    }
  ],
  "logging_steps": 100,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}