{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.753886629992825,
  "eval_steps": 500,
  "global_step": 3252,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09567089213106912,
      "grad_norm": 79.64751434326172,
      "learning_rate": 2.5e-06,
      "loss": 10.5359,
      "step": 50
    },
    {
      "epoch": 0.19134178426213824,
      "grad_norm": 57.142730712890625,
      "learning_rate": 5e-06,
      "loss": 9.4534,
      "step": 100
    },
    {
      "epoch": 0.2870126763932074,
      "grad_norm": 56.33840560913086,
      "learning_rate": 7.5e-06,
      "loss": 8.7762,
      "step": 150
    },
    {
      "epoch": 0.3826835685242765,
      "grad_norm": 48.39895248413086,
      "learning_rate": 1e-05,
      "loss": 8.0812,
      "step": 200
    },
    {
      "epoch": 0.4783544606553456,
      "grad_norm": 51.66913604736328,
      "learning_rate": 1.25e-05,
      "loss": 7.4712,
      "step": 250
    },
    {
      "epoch": 0.5740253527864148,
      "grad_norm": 39.3389778137207,
      "learning_rate": 1.5e-05,
      "loss": 6.9249,
      "step": 300
    },
    {
      "epoch": 0.6696962449174838,
      "grad_norm": 42.176029205322266,
      "learning_rate": 1.7500000000000002e-05,
      "loss": 6.3373,
      "step": 350
    },
    {
      "epoch": 0.765367137048553,
      "grad_norm": 27.88520622253418,
      "learning_rate": 2e-05,
      "loss": 5.8156,
      "step": 400
    },
    {
      "epoch": 0.8610380291796221,
      "grad_norm": 27.722829818725586,
      "learning_rate": 2.2499999999999998e-05,
      "loss": 5.362,
      "step": 450
    },
    {
      "epoch": 0.9567089213106912,
      "grad_norm": 28.680429458618164,
      "learning_rate": 2.5e-05,
      "loss": 5.0434,
      "step": 500
    },
    {
      "epoch": 0.9988041138483617,
      "eval_accuracy": 0.0,
      "eval_loss": 8.436083793640137,
      "eval_normalizer": 94012.0,
      "eval_runtime": 81.5272,
      "eval_samples_per_second": 655.695,
      "eval_steps_per_second": 1.288,
      "step": 522
    },
    {
      "epoch": 1.0535756995933987,
      "grad_norm": 25.760271072387695,
      "learning_rate": 2.75e-05,
      "loss": 4.8276,
      "step": 550
    },
    {
      "epoch": 1.1492465917244679,
      "grad_norm": 27.668893814086914,
      "learning_rate": 3e-05,
      "loss": 4.6499,
      "step": 600
    },
    {
      "epoch": 1.244917483855537,
      "grad_norm": 27.842025756835938,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 4.4827,
      "step": 650
    },
    {
      "epoch": 1.3405883759866062,
      "grad_norm": 30.099782943725586,
      "learning_rate": 3.5000000000000004e-05,
      "loss": 4.3643,
      "step": 700
    },
    {
      "epoch": 1.4362592681176751,
      "grad_norm": 33.141204833984375,
      "learning_rate": 3.75e-05,
      "loss": 4.2279,
      "step": 750
    },
    {
      "epoch": 1.5319301602487443,
      "grad_norm": 37.47197723388672,
      "learning_rate": 4e-05,
      "loss": 4.1339,
      "step": 800
    },
    {
      "epoch": 1.6276010523798135,
      "grad_norm": 34.994571685791016,
      "learning_rate": 4.25e-05,
      "loss": 4.0134,
      "step": 850
    },
    {
      "epoch": 1.7232719445108826,
      "grad_norm": 29.075597763061523,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 3.9216,
      "step": 900
    },
    {
      "epoch": 1.8189428366419516,
      "grad_norm": 31.81198501586914,
      "learning_rate": 4.75e-05,
      "loss": 3.8193,
      "step": 950
    },
    {
      "epoch": 1.9146137287730207,
      "grad_norm": 33.48542785644531,
      "learning_rate": 5e-05,
      "loss": 3.7218,
      "step": 1000
    },
    {
      "epoch": 1.9988041138483617,
      "eval_accuracy": 0.0,
      "eval_loss": 8.203332901000977,
      "eval_normalizer": 93558.0,
      "eval_runtime": 83.028,
      "eval_samples_per_second": 643.843,
      "eval_steps_per_second": 1.265,
      "step": 1044
    },
    {
      "epoch": 2.0114805070557282,
      "grad_norm": 84.52313232421875,
      "learning_rate": 5.25e-05,
      "loss": 4.1746,
      "step": 1050
    },
    {
      "epoch": 2.1071513991867974,
      "grad_norm": 32.48819351196289,
      "learning_rate": 5.5e-05,
      "loss": 6.9906,
      "step": 1100
    },
    {
      "epoch": 2.2028222913178666,
      "grad_norm": 20.257213592529297,
      "learning_rate": 5.75e-05,
      "loss": 6.1631,
      "step": 1150
    },
    {
      "epoch": 2.2984931834489357,
      "grad_norm": 20.544626235961914,
      "learning_rate": 6e-05,
      "loss": 5.787,
      "step": 1200
    },
    {
      "epoch": 2.3597225544128198,
      "eval_accuracy": 0.0,
      "eval_loss": 6.199265480041504,
      "eval_normalizer": 94216.0,
      "eval_runtime": 83.3702,
      "eval_samples_per_second": 641.201,
      "eval_steps_per_second": 1.259,
      "step": 1232
    },
    {
      "epoch": 3.0344415211671847,
      "grad_norm": 23.196285247802734,
      "learning_rate": 6.25e-05,
      "loss": 5.59,
      "step": 1250
    },
    {
      "epoch": 3.130112413298254,
      "grad_norm": 19.782928466796875,
      "learning_rate": 6.500000000000001e-05,
      "loss": 5.2666,
      "step": 1300
    },
    {
      "epoch": 3.225783305429323,
      "grad_norm": 20.39004898071289,
      "learning_rate": 6.75e-05,
      "loss": 5.0723,
      "step": 1350
    },
    {
      "epoch": 3.321454197560392,
      "grad_norm": 28.565898895263672,
      "learning_rate": 7.000000000000001e-05,
      "loss": 4.909,
      "step": 1400
    },
    {
      "epoch": 3.3597225544128198,
      "eval_accuracy": 0.0,
      "eval_loss": 5.842504978179932,
      "eval_normalizer": 93665.0,
      "eval_runtime": 82.9992,
      "eval_samples_per_second": 644.067,
      "eval_steps_per_second": 1.265,
      "step": 1420
    },
    {
      "epoch": 4.057402535278642,
      "grad_norm": 27.73941993713379,
      "learning_rate": 7.25e-05,
      "loss": 5.0404,
      "step": 1450
    },
    {
      "epoch": 4.15307342740971,
      "grad_norm": 25.40974998474121,
      "learning_rate": 7.5e-05,
      "loss": 4.6658,
      "step": 1500
    },
    {
      "epoch": 4.24874431954078,
      "grad_norm": 29.910818099975586,
      "learning_rate": 7.75e-05,
      "loss": 4.5082,
      "step": 1550
    },
    {
      "epoch": 4.344415211671849,
      "grad_norm": 26.495317459106445,
      "learning_rate": 8e-05,
      "loss": 4.3851,
      "step": 1600
    },
    {
      "epoch": 4.440086103802918,
      "grad_norm": 28.781644821166992,
      "learning_rate": 8.25e-05,
      "loss": 4.263,
      "step": 1650
    },
    {
      "epoch": 4.535756995933987,
      "grad_norm": 31.134675979614258,
      "learning_rate": 8.5e-05,
      "loss": 4.2074,
      "step": 1700
    },
    {
      "epoch": 4.631427888065057,
      "grad_norm": 24.8262882232666,
      "learning_rate": 8.75e-05,
      "loss": 4.1359,
      "step": 1750
    },
    {
      "epoch": 4.727098780196125,
      "grad_norm": 23.926483154296875,
      "learning_rate": 8.999999999999999e-05,
      "loss": 4.067,
      "step": 1800
    },
    {
      "epoch": 4.822769672327194,
      "grad_norm": 24.742311477661133,
      "learning_rate": 9.25e-05,
      "loss": 4.0219,
      "step": 1850
    },
    {
      "epoch": 4.918440564458264,
      "grad_norm": 27.34615135192871,
      "learning_rate": 9.5e-05,
      "loss": 4.0033,
      "step": 1900
    },
    {
      "epoch": 4.998804113848362,
      "eval_accuracy": 0.0,
      "eval_loss": 5.4360456466674805,
      "eval_normalizer": 94313.0,
      "eval_runtime": 104.459,
      "eval_samples_per_second": 511.751,
      "eval_steps_per_second": 1.005,
      "step": 1942
    },
    {
      "epoch": 5.015307342740971,
      "grad_norm": 24.026140213012695,
      "learning_rate": 9.750000000000001e-05,
      "loss": 3.9511,
      "step": 1950
    },
    {
      "epoch": 5.11097823487204,
      "grad_norm": 26.587495803833008,
      "learning_rate": 0.0001,
      "loss": 3.9194,
      "step": 2000
    },
    {
      "epoch": 5.206649127003109,
      "grad_norm": 23.23008918762207,
      "learning_rate": 0.0001025,
      "loss": 3.8788,
      "step": 2050
    },
    {
      "epoch": 5.302320019134179,
      "grad_norm": 22.73567771911621,
      "learning_rate": 0.000105,
      "loss": 3.8894,
      "step": 2100
    },
    {
      "epoch": 5.397990911265247,
      "grad_norm": 26.093669891357422,
      "learning_rate": 0.0001075,
      "loss": 3.8171,
      "step": 2150
    },
    {
      "epoch": 5.493661803396317,
      "grad_norm": 26.150381088256836,
      "learning_rate": 0.00011,
      "loss": 3.7666,
      "step": 2200
    },
    {
      "epoch": 5.589332695527386,
      "grad_norm": 28.061275482177734,
      "learning_rate": 0.00011250000000000001,
      "loss": 3.7926,
      "step": 2250
    },
    {
      "epoch": 5.685003587658455,
      "grad_norm": 24.63544464111328,
      "learning_rate": 0.000115,
      "loss": 3.7454,
      "step": 2300
    },
    {
      "epoch": 5.780674479789524,
      "grad_norm": 24.13852310180664,
      "learning_rate": 0.0001175,
      "loss": 3.7038,
      "step": 2350
    },
    {
      "epoch": 5.876345371920593,
      "grad_norm": 23.41834259033203,
      "learning_rate": 0.00012,
      "loss": 3.6888,
      "step": 2400
    },
    {
      "epoch": 5.972016264051662,
      "grad_norm": 26.9339656829834,
      "learning_rate": 0.0001225,
      "loss": 3.6689,
      "step": 2450
    },
    {
      "epoch": 5.998804113848362,
      "eval_accuracy": 0.0,
      "eval_loss": 5.285378456115723,
      "eval_normalizer": 93901.0,
      "eval_runtime": 102.6182,
      "eval_samples_per_second": 520.931,
      "eval_steps_per_second": 1.023,
      "step": 2464
    },
    {
      "epoch": 6.068883042334369,
      "grad_norm": 18.920269012451172,
      "learning_rate": 0.000125,
      "loss": 4.4993,
      "step": 2500
    },
    {
      "epoch": 6.164553934465439,
      "grad_norm": 19.811311721801758,
      "learning_rate": 0.0001275,
      "loss": 4.6066,
      "step": 2550
    },
    {
      "epoch": 6.260224826596508,
      "grad_norm": 23.79183578491211,
      "learning_rate": 0.00013000000000000002,
      "loss": 4.4607,
      "step": 2600
    },
    {
      "epoch": 6.355895718727577,
      "grad_norm": 19.32040786743164,
      "learning_rate": 0.00013250000000000002,
      "loss": 4.3629,
      "step": 2650
    },
    {
      "epoch": 6.451566610858646,
      "grad_norm": 18.420616149902344,
      "learning_rate": 0.000135,
      "loss": 4.2841,
      "step": 2700
    },
    {
      "epoch": 6.547237502989716,
      "grad_norm": 24.54909324645996,
      "learning_rate": 0.0001375,
      "loss": 4.2057,
      "step": 2750
    },
    {
      "epoch": 6.642908395120784,
      "grad_norm": 20.581602096557617,
      "learning_rate": 0.00014000000000000001,
      "loss": 4.1175,
      "step": 2800
    },
    {
      "epoch": 6.738579287251854,
      "grad_norm": 19.419635772705078,
      "learning_rate": 0.0001425,
      "loss": 4.0771,
      "step": 2850
    },
    {
      "epoch": 6.753886629992825,
      "eval_accuracy": 0.0,
      "eval_loss": 5.031363487243652,
      "eval_normalizer": 94462.0,
      "eval_runtime": 99.5317,
      "eval_samples_per_second": 537.085,
      "eval_steps_per_second": 1.055,
      "step": 2858
    },
    {
      "epoch": 7.080363549390098,
      "grad_norm": 22.94988250732422,
      "learning_rate": 0.000145,
      "loss": 4.0966,
      "step": 2900
    },
    {
      "epoch": 7.176034441521167,
      "grad_norm": 19.95139503479004,
      "learning_rate": 0.0001475,
      "loss": 3.9988,
      "step": 2950
    },
    {
      "epoch": 7.271705333652236,
      "grad_norm": 20.52120590209961,
      "learning_rate": 0.00015,
      "loss": 3.9693,
      "step": 3000
    },
    {
      "epoch": 7.367376225783305,
      "grad_norm": 20.660249710083008,
      "learning_rate": 0.0001525,
      "loss": 3.8935,
      "step": 3050
    },
    {
      "epoch": 7.463047117914375,
      "grad_norm": 19.104249954223633,
      "learning_rate": 0.000155,
      "loss": 3.8809,
      "step": 3100
    },
    {
      "epoch": 7.558718010045443,
      "grad_norm": 19.59981346130371,
      "learning_rate": 0.0001575,
      "loss": 3.8089,
      "step": 3150
    },
    {
      "epoch": 7.654388902176513,
      "grad_norm": 22.192567825317383,
      "learning_rate": 0.00016,
      "loss": 3.7668,
      "step": 3200
    },
    {
      "epoch": 7.750059794307582,
      "grad_norm": 20.03333282470703,
      "learning_rate": 0.00016250000000000002,
      "loss": 3.7527,
      "step": 3250
    },
    {
      "epoch": 7.753886629992825,
      "eval_accuracy": 0.0,
      "eval_loss": 4.870262145996094,
      "eval_normalizer": 93833.0,
      "eval_runtime": 91.744,
      "eval_samples_per_second": 582.675,
      "eval_steps_per_second": 1.144,
      "step": 3252
    }
  ],
  "logging_steps": 50,
  "max_steps": 5220,
  "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": 2.955536204157861e+17,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}