{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 119380,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 2.0833333333333333e-05,
      "loss": 7.7598,
      "step": 1000
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.1666666666666665e-05,
      "loss": 5.5454,
      "step": 2000
    },
    {
      "epoch": 0.25,
      "learning_rate": 6.25e-05,
      "loss": 4.9947,
      "step": 3000
    },
    {
      "epoch": 0.34,
      "learning_rate": 8.333333333333333e-05,
      "loss": 4.6428,
      "step": 4000
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00010416666666666667,
      "loss": 4.3201,
      "step": 5000
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.000125,
      "loss": 4.0651,
      "step": 6000
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00014583333333333335,
      "loss": 3.8937,
      "step": 7000
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00016666666666666666,
      "loss": 3.7657,
      "step": 8000
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001875,
      "loss": 3.6866,
      "step": 9000
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00020833333333333335,
      "loss": 3.5962,
      "step": 10000
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00022916666666666666,
      "loss": 3.5129,
      "step": 11000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3523433003120361,
      "eval_loss": 3.4626712799072266,
      "eval_runtime": 8.7029,
      "eval_samples_per_second": 8043.269,
      "eval_steps_per_second": 15.742,
      "step": 11938
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00025,
      "loss": 3.4983,
      "step": 12000
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002708333333333333,
      "loss": 3.4416,
      "step": 13000
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002916666666666667,
      "loss": 3.4233,
      "step": 14000
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0003125,
      "loss": 3.3965,
      "step": 15000
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0003333333333333333,
      "loss": 3.3694,
      "step": 16000
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0003541666666666667,
      "loss": 3.3621,
      "step": 17000
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.000375,
      "loss": 3.333,
      "step": 18000
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0003958333333333333,
      "loss": 3.3604,
      "step": 19000
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0004166666666666667,
      "loss": 3.3273,
      "step": 20000
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0004375,
      "loss": 3.3097,
      "step": 21000
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0004583333333333333,
      "loss": 3.3087,
      "step": 22000
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0004791666666666667,
      "loss": 3.319,
      "step": 23000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.36410225308548383,
      "eval_loss": 3.3322079181671143,
      "eval_runtime": 9.4912,
      "eval_samples_per_second": 7375.242,
      "eval_steps_per_second": 14.434,
      "step": 23876
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0005,
      "loss": 3.3119,
      "step": 24000
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0004947578108618159,
      "loss": 3.304,
      "step": 25000
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0004895156217236318,
      "loss": 3.2935,
      "step": 26000
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0004842734325854477,
      "loss": 3.2888,
      "step": 27000
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0004790312434472636,
      "loss": 3.2313,
      "step": 28000
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0004737890543090795,
      "loss": 3.2641,
      "step": 29000
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0004685468651708954,
      "loss": 3.2296,
      "step": 30000
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.00046330467603271125,
      "loss": 3.2117,
      "step": 31000
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.00045806248689452716,
      "loss": 3.1985,
      "step": 32000
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.000452820297756343,
      "loss": 3.2068,
      "step": 33000
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.00044757810861815897,
      "loss": 3.1658,
      "step": 34000
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.0004423359194799749,
      "loss": 3.1577,
      "step": 35000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.38102238455359716,
      "eval_loss": 3.184136390686035,
      "eval_runtime": 8.8582,
      "eval_samples_per_second": 7902.295,
      "eval_steps_per_second": 15.466,
      "step": 35814
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.00043709373034179073,
      "loss": 3.1581,
      "step": 36000
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.00043185154120360664,
      "loss": 3.1229,
      "step": 37000
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0004266093520654225,
      "loss": 3.1043,
      "step": 38000
    },
    {
      "epoch": 3.27,
      "learning_rate": 0.0004213671629272384,
      "loss": 3.0881,
      "step": 39000
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.00041612497378905436,
      "loss": 3.095,
      "step": 40000
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.0004108827846508702,
      "loss": 3.0888,
      "step": 41000
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.0004056405955126861,
      "loss": 3.0583,
      "step": 42000
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.00040039840637450197,
      "loss": 3.0716,
      "step": 43000
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0003951562172363179,
      "loss": 3.046,
      "step": 44000
    },
    {
      "epoch": 3.77,
      "learning_rate": 0.0003899140280981338,
      "loss": 3.0566,
      "step": 45000
    },
    {
      "epoch": 3.85,
      "learning_rate": 0.0003846718389599497,
      "loss": 3.0571,
      "step": 46000
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.0003794296498217656,
      "loss": 3.0357,
      "step": 47000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3981597974867745,
      "eval_loss": 3.058806896209717,
      "eval_runtime": 9.478,
      "eval_samples_per_second": 7385.506,
      "eval_steps_per_second": 14.454,
      "step": 47752
    },
    {
      "epoch": 4.02,
      "learning_rate": 0.00037418746068358145,
      "loss": 3.0134,
      "step": 48000
    },
    {
      "epoch": 4.1,
      "learning_rate": 0.00036894527154539736,
      "loss": 2.9809,
      "step": 49000
    },
    {
      "epoch": 4.19,
      "learning_rate": 0.00036370308240721327,
      "loss": 2.9708,
      "step": 50000
    },
    {
      "epoch": 4.27,
      "learning_rate": 0.0003584608932690291,
      "loss": 2.9975,
      "step": 51000
    },
    {
      "epoch": 4.36,
      "learning_rate": 0.0003532187041308451,
      "loss": 2.9888,
      "step": 52000
    },
    {
      "epoch": 4.44,
      "learning_rate": 0.00034797651499266093,
      "loss": 2.9647,
      "step": 53000
    },
    {
      "epoch": 4.52,
      "learning_rate": 0.00034273432585447684,
      "loss": 2.9853,
      "step": 54000
    },
    {
      "epoch": 4.61,
      "learning_rate": 0.00033749213671629275,
      "loss": 2.9783,
      "step": 55000
    },
    {
      "epoch": 4.69,
      "learning_rate": 0.0003322499475781086,
      "loss": 2.9553,
      "step": 56000
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.0003270077584399245,
      "loss": 2.9312,
      "step": 57000
    },
    {
      "epoch": 4.86,
      "learning_rate": 0.0003217655693017404,
      "loss": 2.9468,
      "step": 58000
    },
    {
      "epoch": 4.94,
      "learning_rate": 0.0003165233801635563,
      "loss": 2.9606,
      "step": 59000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.41093517308452115,
      "eval_loss": 2.9534778594970703,
      "eval_runtime": 8.7667,
      "eval_samples_per_second": 7984.805,
      "eval_steps_per_second": 15.627,
      "step": 59690
    },
    {
      "epoch": 5.03,
      "learning_rate": 0.00031128119102537223,
      "loss": 2.9322,
      "step": 60000
    },
    {
      "epoch": 5.11,
      "learning_rate": 0.0003060390018871881,
      "loss": 2.9048,
      "step": 61000
    },
    {
      "epoch": 5.19,
      "learning_rate": 0.000300796812749004,
      "loss": 2.8633,
      "step": 62000
    },
    {
      "epoch": 5.28,
      "learning_rate": 0.00029555462361081984,
      "loss": 2.914,
      "step": 63000
    },
    {
      "epoch": 5.36,
      "learning_rate": 0.0002903124344726358,
      "loss": 2.881,
      "step": 64000
    },
    {
      "epoch": 5.44,
      "learning_rate": 0.0002850702453344517,
      "loss": 2.8721,
      "step": 65000
    },
    {
      "epoch": 5.53,
      "learning_rate": 0.00027982805619626756,
      "loss": 2.886,
      "step": 66000
    },
    {
      "epoch": 5.61,
      "learning_rate": 0.00027458586705808347,
      "loss": 2.8801,
      "step": 67000
    },
    {
      "epoch": 5.7,
      "learning_rate": 0.0002693436779198993,
      "loss": 2.87,
      "step": 68000
    },
    {
      "epoch": 5.78,
      "learning_rate": 0.00026410148878171523,
      "loss": 2.8705,
      "step": 69000
    },
    {
      "epoch": 5.86,
      "learning_rate": 0.0002588592996435312,
      "loss": 2.8638,
      "step": 70000
    },
    {
      "epoch": 5.95,
      "learning_rate": 0.00025361711050534704,
      "loss": 2.87,
      "step": 71000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.42207386406938285,
      "eval_loss": 2.8744776248931885,
      "eval_runtime": 8.7882,
      "eval_samples_per_second": 7965.243,
      "eval_steps_per_second": 15.589,
      "step": 71628
    },
    {
      "epoch": 6.03,
      "learning_rate": 0.00024837492136716295,
      "loss": 2.8343,
      "step": 72000
    },
    {
      "epoch": 6.11,
      "learning_rate": 0.00024313273222897883,
      "loss": 2.8347,
      "step": 73000
    },
    {
      "epoch": 6.2,
      "learning_rate": 0.0002378905430907947,
      "loss": 2.8269,
      "step": 74000
    },
    {
      "epoch": 6.28,
      "learning_rate": 0.00023264835395261061,
      "loss": 2.8174,
      "step": 75000
    },
    {
      "epoch": 6.37,
      "learning_rate": 0.0002274061648144265,
      "loss": 2.8174,
      "step": 76000
    },
    {
      "epoch": 6.45,
      "learning_rate": 0.0002221639756762424,
      "loss": 2.8216,
      "step": 77000
    },
    {
      "epoch": 6.53,
      "learning_rate": 0.0002169217865380583,
      "loss": 2.805,
      "step": 78000
    },
    {
      "epoch": 6.62,
      "learning_rate": 0.0002116795973998742,
      "loss": 2.8225,
      "step": 79000
    },
    {
      "epoch": 6.7,
      "learning_rate": 0.00020643740826169007,
      "loss": 2.8059,
      "step": 80000
    },
    {
      "epoch": 6.79,
      "learning_rate": 0.00020119521912350598,
      "loss": 2.8046,
      "step": 81000
    },
    {
      "epoch": 6.87,
      "learning_rate": 0.00019595302998532188,
      "loss": 2.7754,
      "step": 82000
    },
    {
      "epoch": 6.95,
      "learning_rate": 0.00019071084084713776,
      "loss": 2.7817,
      "step": 83000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4283667840107286,
      "eval_loss": 2.8350658416748047,
      "eval_runtime": 8.7803,
      "eval_samples_per_second": 7972.401,
      "eval_steps_per_second": 15.603,
      "step": 83566
    },
    {
      "epoch": 7.04,
      "learning_rate": 0.00018546865170895367,
      "loss": 2.7572,
      "step": 84000
    },
    {
      "epoch": 7.12,
      "learning_rate": 0.00018022646257076955,
      "loss": 2.7627,
      "step": 85000
    },
    {
      "epoch": 7.2,
      "learning_rate": 0.00017498427343258543,
      "loss": 2.7528,
      "step": 86000
    },
    {
      "epoch": 7.29,
      "learning_rate": 0.00016974208429440136,
      "loss": 2.7468,
      "step": 87000
    },
    {
      "epoch": 7.37,
      "learning_rate": 0.00016449989515621724,
      "loss": 2.7556,
      "step": 88000
    },
    {
      "epoch": 7.46,
      "learning_rate": 0.00015925770601803312,
      "loss": 2.7687,
      "step": 89000
    },
    {
      "epoch": 7.54,
      "learning_rate": 0.00015401551687984903,
      "loss": 2.7265,
      "step": 90000
    },
    {
      "epoch": 7.62,
      "learning_rate": 0.0001487733277416649,
      "loss": 2.7375,
      "step": 91000
    },
    {
      "epoch": 7.71,
      "learning_rate": 0.00014353113860348082,
      "loss": 2.7091,
      "step": 92000
    },
    {
      "epoch": 7.79,
      "learning_rate": 0.00013828894946529672,
      "loss": 2.7401,
      "step": 93000
    },
    {
      "epoch": 7.87,
      "learning_rate": 0.0001330467603271126,
      "loss": 2.7341,
      "step": 94000
    },
    {
      "epoch": 7.96,
      "learning_rate": 0.00012780457118892848,
      "loss": 2.7388,
      "step": 95000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.44167649816982785,
      "eval_loss": 2.7535808086395264,
      "eval_runtime": 8.9073,
      "eval_samples_per_second": 7858.746,
      "eval_steps_per_second": 15.381,
      "step": 95504
    },
    {
      "epoch": 8.04,
      "learning_rate": 0.0001225623820507444,
      "loss": 2.7064,
      "step": 96000
    },
    {
      "epoch": 8.13,
      "learning_rate": 0.00011732019291256028,
      "loss": 2.6997,
      "step": 97000
    },
    {
      "epoch": 8.21,
      "learning_rate": 0.00011207800377437619,
      "loss": 2.6857,
      "step": 98000
    },
    {
      "epoch": 8.29,
      "learning_rate": 0.00010683581463619207,
      "loss": 2.6924,
      "step": 99000
    },
    {
      "epoch": 8.38,
      "learning_rate": 0.00010159362549800798,
      "loss": 2.6781,
      "step": 100000
    },
    {
      "epoch": 8.46,
      "learning_rate": 9.635143635982387e-05,
      "loss": 2.6773,
      "step": 101000
    },
    {
      "epoch": 8.54,
      "learning_rate": 9.110924722163975e-05,
      "loss": 2.6722,
      "step": 102000
    },
    {
      "epoch": 8.63,
      "learning_rate": 8.586705808345566e-05,
      "loss": 2.6965,
      "step": 103000
    },
    {
      "epoch": 8.71,
      "learning_rate": 8.062486894527155e-05,
      "loss": 2.6532,
      "step": 104000
    },
    {
      "epoch": 8.8,
      "learning_rate": 7.538267980708744e-05,
      "loss": 2.6804,
      "step": 105000
    },
    {
      "epoch": 8.88,
      "learning_rate": 7.014049066890334e-05,
      "loss": 2.6515,
      "step": 106000
    },
    {
      "epoch": 8.96,
      "learning_rate": 6.489830153071923e-05,
      "loss": 2.6618,
      "step": 107000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.4423983455491013,
      "eval_loss": 2.7308034896850586,
      "eval_runtime": 8.6936,
      "eval_samples_per_second": 8051.873,
      "eval_steps_per_second": 15.759,
      "step": 107442
    },
    {
      "epoch": 9.05,
      "learning_rate": 5.965611239253512e-05,
      "loss": 2.6375,
      "step": 108000
    },
    {
      "epoch": 9.13,
      "learning_rate": 5.441392325435102e-05,
      "loss": 2.6582,
      "step": 109000
    },
    {
      "epoch": 9.21,
      "learning_rate": 4.917173411616691e-05,
      "loss": 2.6559,
      "step": 110000
    },
    {
      "epoch": 9.3,
      "learning_rate": 4.3929544977982805e-05,
      "loss": 2.6294,
      "step": 111000
    },
    {
      "epoch": 9.38,
      "learning_rate": 3.8687355839798705e-05,
      "loss": 2.6339,
      "step": 112000
    },
    {
      "epoch": 9.47,
      "learning_rate": 3.344516670161459e-05,
      "loss": 2.633,
      "step": 113000
    },
    {
      "epoch": 9.55,
      "learning_rate": 2.820297756343049e-05,
      "loss": 2.6274,
      "step": 114000
    },
    {
      "epoch": 9.63,
      "learning_rate": 2.2960788425246382e-05,
      "loss": 2.6228,
      "step": 115000
    },
    {
      "epoch": 9.72,
      "learning_rate": 1.771859928706228e-05,
      "loss": 2.6251,
      "step": 116000
    },
    {
      "epoch": 9.8,
      "learning_rate": 1.2476410148878172e-05,
      "loss": 2.6179,
      "step": 117000
    },
    {
      "epoch": 9.88,
      "learning_rate": 7.234221010694066e-06,
      "loss": 2.6305,
      "step": 118000
    },
    {
      "epoch": 9.97,
      "learning_rate": 1.99203187250996e-06,
      "loss": 2.6258,
      "step": 119000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.45217033845404303,
      "eval_loss": 2.688030242919922,
      "eval_runtime": 8.8102,
      "eval_samples_per_second": 7945.334,
      "eval_steps_per_second": 15.55,
      "step": 119380
    },
    {
      "epoch": 10.0,
      "step": 119380,
      "total_flos": 6558690447189504.0,
      "train_loss": 3.080881611061448,
      "train_runtime": 7359.6812,
      "train_samples_per_second": 1038.074,
      "train_steps_per_second": 16.221
    }
  ],
  "logging_steps": 1000,
  "max_steps": 119380,
  "num_train_epochs": 10,
  "save_steps": 2000,
  "total_flos": 6558690447189504.0,
  "trial_name": null,
  "trial_params": null
}