{
  "best_metric": 6.409651756286621,
  "best_model_checkpoint": "HBERTv1_emb_compress_48_L10_H64_A2/checkpoint-290000",
  "epoch": 5.0,
  "eval_steps": 10000,
  "global_step": 305145,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 5.000000000000001e-07,
      "loss": 10.3246,
      "step": 500
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 10.286,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.5e-06,
      "loss": 10.2609,
      "step": 1500
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 10.2303,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.5e-06,
      "loss": 10.1912,
      "step": 2500
    },
    {
      "epoch": 0.05,
      "learning_rate": 3e-06,
      "loss": 10.1438,
      "step": 3000
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.5e-06,
      "loss": 10.0874,
      "step": 3500
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.000000000000001e-06,
      "loss": 10.0221,
      "step": 4000
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.5e-06,
      "loss": 9.9475,
      "step": 4500
    },
    {
      "epoch": 0.08,
      "learning_rate": 5e-06,
      "loss": 9.8655,
      "step": 5000
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.500000000000001e-06,
      "loss": 9.7745,
      "step": 5500
    },
    {
      "epoch": 0.1,
      "learning_rate": 6e-06,
      "loss": 9.676,
      "step": 6000
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 9.5701,
      "step": 6500
    },
    {
      "epoch": 0.11,
      "learning_rate": 7e-06,
      "loss": 9.4568,
      "step": 7000
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.500000000000001e-06,
      "loss": 9.3367,
      "step": 7500
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.000000000000001e-06,
      "loss": 9.2106,
      "step": 8000
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.5e-06,
      "loss": 9.0801,
      "step": 8500
    },
    {
      "epoch": 0.15,
      "learning_rate": 9e-06,
      "loss": 8.9456,
      "step": 9000
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.5e-06,
      "loss": 8.8064,
      "step": 9500
    },
    {
      "epoch": 0.16,
      "learning_rate": 1e-05,
      "loss": 8.6666,
      "step": 10000
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.04651231344428305,
      "eval_loss": 8.596183776855469,
      "eval_runtime": 578.6057,
      "eval_samples_per_second": 532.929,
      "eval_steps_per_second": 5.553,
      "step": 10000
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.983059174304157e-06,
      "loss": 8.5283,
      "step": 10500
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.966118348608312e-06,
      "loss": 8.3928,
      "step": 11000
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.949177522912468e-06,
      "loss": 8.2696,
      "step": 11500
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.932236697216624e-06,
      "loss": 8.1507,
      "step": 12000
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.91529587152078e-06,
      "loss": 8.0404,
      "step": 12500
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.898355045824935e-06,
      "loss": 7.94,
      "step": 13000
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.881414220129089e-06,
      "loss": 7.8432,
      "step": 13500
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.864473394433246e-06,
      "loss": 7.7556,
      "step": 14000
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.847532568737402e-06,
      "loss": 7.676,
      "step": 14500
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.830591743041556e-06,
      "loss": 7.6049,
      "step": 15000
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.813650917345712e-06,
      "loss": 7.5389,
      "step": 15500
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.796710091649867e-06,
      "loss": 7.4852,
      "step": 16000
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.779769265954023e-06,
      "loss": 7.4325,
      "step": 16500
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.762828440258179e-06,
      "loss": 7.3862,
      "step": 17000
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.745887614562334e-06,
      "loss": 7.3573,
      "step": 17500
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.72894678886649e-06,
      "loss": 7.3297,
      "step": 18000
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.712005963170646e-06,
      "loss": 7.2966,
      "step": 18500
    },
    {
      "epoch": 0.31,
      "learning_rate": 9.695065137474802e-06,
      "loss": 7.2835,
      "step": 19000
    },
    {
      "epoch": 0.32,
      "learning_rate": 9.678124311778957e-06,
      "loss": 7.2675,
      "step": 19500
    },
    {
      "epoch": 0.33,
      "learning_rate": 9.661183486083113e-06,
      "loss": 7.2486,
      "step": 20000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.046660418954156756,
      "eval_loss": 7.2447123527526855,
      "eval_runtime": 578.3557,
      "eval_samples_per_second": 533.16,
      "eval_steps_per_second": 5.555,
      "step": 20000
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.644242660387269e-06,
      "loss": 7.2384,
      "step": 20500
    },
    {
      "epoch": 0.34,
      "learning_rate": 9.627301834691424e-06,
      "loss": 7.2242,
      "step": 21000
    },
    {
      "epoch": 0.35,
      "learning_rate": 9.610361008995578e-06,
      "loss": 7.2086,
      "step": 21500
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.593420183299736e-06,
      "loss": 7.1973,
      "step": 22000
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.576479357603891e-06,
      "loss": 7.1844,
      "step": 22500
    },
    {
      "epoch": 0.38,
      "learning_rate": 9.559538531908045e-06,
      "loss": 7.1654,
      "step": 23000
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.542597706212201e-06,
      "loss": 7.1566,
      "step": 23500
    },
    {
      "epoch": 0.39,
      "learning_rate": 9.525656880516357e-06,
      "loss": 7.1442,
      "step": 24000
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.508716054820512e-06,
      "loss": 7.1278,
      "step": 24500
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.491775229124668e-06,
      "loss": 7.116,
      "step": 25000
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.474834403428824e-06,
      "loss": 7.1038,
      "step": 25500
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.45789357773298e-06,
      "loss": 7.0909,
      "step": 26000
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.440952752037135e-06,
      "loss": 7.0758,
      "step": 26500
    },
    {
      "epoch": 0.44,
      "learning_rate": 9.424011926341291e-06,
      "loss": 7.0695,
      "step": 27000
    },
    {
      "epoch": 0.45,
      "learning_rate": 9.407071100645447e-06,
      "loss": 7.0572,
      "step": 27500
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.390130274949602e-06,
      "loss": 7.0465,
      "step": 28000
    },
    {
      "epoch": 0.47,
      "learning_rate": 9.373189449253758e-06,
      "loss": 7.0409,
      "step": 28500
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.356248623557914e-06,
      "loss": 7.0307,
      "step": 29000
    },
    {
      "epoch": 0.48,
      "learning_rate": 9.339307797862068e-06,
      "loss": 7.0191,
      "step": 29500
    },
    {
      "epoch": 0.49,
      "learning_rate": 9.322366972166225e-06,
      "loss": 7.0176,
      "step": 30000
    },
    {
      "epoch": 0.49,
      "eval_accuracy": 0.06700670747772926,
      "eval_loss": 7.00886869430542,
      "eval_runtime": 577.0656,
      "eval_samples_per_second": 534.352,
      "eval_steps_per_second": 5.568,
      "step": 30000
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.305426146470379e-06,
      "loss": 7.0064,
      "step": 30500
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.288485320774535e-06,
      "loss": 6.9966,
      "step": 31000
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.27154449507869e-06,
      "loss": 6.9866,
      "step": 31500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.254603669382846e-06,
      "loss": 6.9774,
      "step": 32000
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.237662843687002e-06,
      "loss": 6.9764,
      "step": 32500
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.220722017991158e-06,
      "loss": 6.9761,
      "step": 33000
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.203781192295313e-06,
      "loss": 6.9649,
      "step": 33500
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.186840366599469e-06,
      "loss": 6.9537,
      "step": 34000
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.169899540903625e-06,
      "loss": 6.9483,
      "step": 34500
    },
    {
      "epoch": 0.57,
      "learning_rate": 9.15295871520778e-06,
      "loss": 6.9414,
      "step": 35000
    },
    {
      "epoch": 0.58,
      "learning_rate": 9.136017889511936e-06,
      "loss": 6.9418,
      "step": 35500
    },
    {
      "epoch": 0.59,
      "learning_rate": 9.11907706381609e-06,
      "loss": 6.9344,
      "step": 36000
    },
    {
      "epoch": 0.6,
      "learning_rate": 9.102136238120247e-06,
      "loss": 6.9222,
      "step": 36500
    },
    {
      "epoch": 0.61,
      "learning_rate": 9.085195412424403e-06,
      "loss": 6.9134,
      "step": 37000
    },
    {
      "epoch": 0.61,
      "learning_rate": 9.068254586728557e-06,
      "loss": 6.9102,
      "step": 37500
    },
    {
      "epoch": 0.62,
      "learning_rate": 9.051313761032714e-06,
      "loss": 6.9076,
      "step": 38000
    },
    {
      "epoch": 0.63,
      "learning_rate": 9.034372935336868e-06,
      "loss": 6.9022,
      "step": 38500
    },
    {
      "epoch": 0.64,
      "learning_rate": 9.017432109641024e-06,
      "loss": 6.891,
      "step": 39000
    },
    {
      "epoch": 0.65,
      "learning_rate": 9.00049128394518e-06,
      "loss": 6.8878,
      "step": 39500
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.983550458249335e-06,
      "loss": 6.8859,
      "step": 40000
    },
    {
      "epoch": 0.66,
      "eval_accuracy": 0.08399340434292331,
      "eval_loss": 6.879542827606201,
      "eval_runtime": 577.5975,
      "eval_samples_per_second": 533.86,
      "eval_steps_per_second": 5.563,
      "step": 40000
    },
    {
      "epoch": 0.66,
      "learning_rate": 8.966609632553491e-06,
      "loss": 6.8787,
      "step": 40500
    },
    {
      "epoch": 0.67,
      "learning_rate": 8.949668806857647e-06,
      "loss": 6.8715,
      "step": 41000
    },
    {
      "epoch": 0.68,
      "learning_rate": 8.932727981161803e-06,
      "loss": 6.8664,
      "step": 41500
    },
    {
      "epoch": 0.69,
      "learning_rate": 8.915787155465958e-06,
      "loss": 6.8595,
      "step": 42000
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.898846329770114e-06,
      "loss": 6.8535,
      "step": 42500
    },
    {
      "epoch": 0.7,
      "learning_rate": 8.88190550407427e-06,
      "loss": 6.8547,
      "step": 43000
    },
    {
      "epoch": 0.71,
      "learning_rate": 8.864964678378425e-06,
      "loss": 6.8382,
      "step": 43500
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.84802385268258e-06,
      "loss": 6.8392,
      "step": 44000
    },
    {
      "epoch": 0.73,
      "learning_rate": 8.831083026986737e-06,
      "loss": 6.8347,
      "step": 44500
    },
    {
      "epoch": 0.74,
      "learning_rate": 8.814142201290892e-06,
      "loss": 6.829,
      "step": 45000
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.797201375595046e-06,
      "loss": 6.8266,
      "step": 45500
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.780260549899204e-06,
      "loss": 6.8159,
      "step": 46000
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.763319724203358e-06,
      "loss": 6.8162,
      "step": 46500
    },
    {
      "epoch": 0.77,
      "learning_rate": 8.746378898507513e-06,
      "loss": 6.8145,
      "step": 47000
    },
    {
      "epoch": 0.78,
      "learning_rate": 8.729438072811671e-06,
      "loss": 6.8038,
      "step": 47500
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.712497247115825e-06,
      "loss": 6.807,
      "step": 48000
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.69555642141998e-06,
      "loss": 6.8015,
      "step": 48500
    },
    {
      "epoch": 0.8,
      "learning_rate": 8.678615595724136e-06,
      "loss": 6.7926,
      "step": 49000
    },
    {
      "epoch": 0.81,
      "learning_rate": 8.661674770028292e-06,
      "loss": 6.7957,
      "step": 49500
    },
    {
      "epoch": 0.82,
      "learning_rate": 8.644733944332448e-06,
      "loss": 6.7911,
      "step": 50000
    },
    {
      "epoch": 0.82,
      "eval_accuracy": 0.09175178489384145,
      "eval_loss": 6.785674571990967,
      "eval_runtime": 577.5324,
      "eval_samples_per_second": 533.92,
      "eval_steps_per_second": 5.563,
      "step": 50000
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.627793118636603e-06,
      "loss": 6.7799,
      "step": 50500
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.610852292940759e-06,
      "loss": 6.7777,
      "step": 51000
    },
    {
      "epoch": 0.84,
      "learning_rate": 8.593911467244915e-06,
      "loss": 6.7773,
      "step": 51500
    },
    {
      "epoch": 0.85,
      "learning_rate": 8.576970641549069e-06,
      "loss": 6.7708,
      "step": 52000
    },
    {
      "epoch": 0.86,
      "learning_rate": 8.560029815853226e-06,
      "loss": 6.7657,
      "step": 52500
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.543088990157382e-06,
      "loss": 6.7646,
      "step": 53000
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.526148164461536e-06,
      "loss": 6.7707,
      "step": 53500
    },
    {
      "epoch": 0.88,
      "learning_rate": 8.509207338765693e-06,
      "loss": 6.7593,
      "step": 54000
    },
    {
      "epoch": 0.89,
      "learning_rate": 8.492266513069847e-06,
      "loss": 6.7557,
      "step": 54500
    },
    {
      "epoch": 0.9,
      "learning_rate": 8.475325687374003e-06,
      "loss": 6.7519,
      "step": 55000
    },
    {
      "epoch": 0.91,
      "learning_rate": 8.45838486167816e-06,
      "loss": 6.7464,
      "step": 55500
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.441444035982314e-06,
      "loss": 6.7528,
      "step": 56000
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.42450321028647e-06,
      "loss": 6.7459,
      "step": 56500
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.407562384590626e-06,
      "loss": 6.7459,
      "step": 57000
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.390621558894781e-06,
      "loss": 6.7402,
      "step": 57500
    },
    {
      "epoch": 0.95,
      "learning_rate": 8.373680733198937e-06,
      "loss": 6.7438,
      "step": 58000
    },
    {
      "epoch": 0.96,
      "learning_rate": 8.356739907503093e-06,
      "loss": 6.7325,
      "step": 58500
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.339799081807248e-06,
      "loss": 6.7271,
      "step": 59000
    },
    {
      "epoch": 0.97,
      "learning_rate": 8.322858256111404e-06,
      "loss": 6.73,
      "step": 59500
    },
    {
      "epoch": 0.98,
      "learning_rate": 8.305917430415558e-06,
      "loss": 6.7203,
      "step": 60000
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.09517665157607533,
      "eval_loss": 6.721036434173584,
      "eval_runtime": 577.9155,
      "eval_samples_per_second": 533.566,
      "eval_steps_per_second": 5.56,
      "step": 60000
    },
    {
      "epoch": 0.99,
      "learning_rate": 8.288976604719715e-06,
      "loss": 6.7212,
      "step": 60500
    },
    {
      "epoch": 1.0,
      "learning_rate": 8.272035779023871e-06,
      "loss": 6.7216,
      "step": 61000
    },
    {
      "epoch": 1.01,
      "learning_rate": 8.255094953328025e-06,
      "loss": 6.7161,
      "step": 61500
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.238154127632183e-06,
      "loss": 6.7165,
      "step": 62000
    },
    {
      "epoch": 1.02,
      "learning_rate": 8.221213301936337e-06,
      "loss": 6.7093,
      "step": 62500
    },
    {
      "epoch": 1.03,
      "learning_rate": 8.204272476240492e-06,
      "loss": 6.7039,
      "step": 63000
    },
    {
      "epoch": 1.04,
      "learning_rate": 8.18733165054465e-06,
      "loss": 6.7044,
      "step": 63500
    },
    {
      "epoch": 1.05,
      "learning_rate": 8.170390824848804e-06,
      "loss": 6.7022,
      "step": 64000
    },
    {
      "epoch": 1.06,
      "learning_rate": 8.15344999915296e-06,
      "loss": 6.7041,
      "step": 64500
    },
    {
      "epoch": 1.07,
      "learning_rate": 8.136509173457115e-06,
      "loss": 6.6962,
      "step": 65000
    },
    {
      "epoch": 1.07,
      "learning_rate": 8.11956834776127e-06,
      "loss": 6.6967,
      "step": 65500
    },
    {
      "epoch": 1.08,
      "learning_rate": 8.102627522065426e-06,
      "loss": 6.6925,
      "step": 66000
    },
    {
      "epoch": 1.09,
      "learning_rate": 8.085686696369582e-06,
      "loss": 6.6953,
      "step": 66500
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.068745870673738e-06,
      "loss": 6.6922,
      "step": 67000
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.051805044977893e-06,
      "loss": 6.6839,
      "step": 67500
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.034864219282047e-06,
      "loss": 6.6828,
      "step": 68000
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.017923393586205e-06,
      "loss": 6.6784,
      "step": 68500
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.00098256789036e-06,
      "loss": 6.6716,
      "step": 69000
    },
    {
      "epoch": 1.14,
      "learning_rate": 7.984041742194514e-06,
      "loss": 6.6783,
      "step": 69500
    },
    {
      "epoch": 1.15,
      "learning_rate": 7.967100916498672e-06,
      "loss": 6.6722,
      "step": 70000
    },
    {
      "epoch": 1.15,
      "eval_accuracy": 0.10037379676582102,
      "eval_loss": 6.671488285064697,
      "eval_runtime": 581.7267,
      "eval_samples_per_second": 530.07,
      "eval_steps_per_second": 5.523,
      "step": 70000
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.950160090802826e-06,
      "loss": 6.6706,
      "step": 70500
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.933219265106982e-06,
      "loss": 6.6715,
      "step": 71000
    },
    {
      "epoch": 1.17,
      "learning_rate": 7.916278439411137e-06,
      "loss": 6.6721,
      "step": 71500
    },
    {
      "epoch": 1.18,
      "learning_rate": 7.899337613715293e-06,
      "loss": 6.6669,
      "step": 72000
    },
    {
      "epoch": 1.19,
      "learning_rate": 7.882396788019449e-06,
      "loss": 6.6615,
      "step": 72500
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.865455962323604e-06,
      "loss": 6.6594,
      "step": 73000
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.84851513662776e-06,
      "loss": 6.657,
      "step": 73500
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.831574310931916e-06,
      "loss": 6.6544,
      "step": 74000
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.814633485236071e-06,
      "loss": 6.6507,
      "step": 74500
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.797692659540227e-06,
      "loss": 6.6537,
      "step": 75000
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.780751833844383e-06,
      "loss": 6.6518,
      "step": 75500
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.763811008148537e-06,
      "loss": 6.6443,
      "step": 76000
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.746870182452694e-06,
      "loss": 6.6422,
      "step": 76500
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.729929356756848e-06,
      "loss": 6.6457,
      "step": 77000
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.712988531061004e-06,
      "loss": 6.6414,
      "step": 77500
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.696047705365161e-06,
      "loss": 6.6433,
      "step": 78000
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.679106879669315e-06,
      "loss": 6.6378,
      "step": 78500
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.662166053973471e-06,
      "loss": 6.635,
      "step": 79000
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.645225228277627e-06,
      "loss": 6.6381,
      "step": 79500
    },
    {
      "epoch": 1.31,
      "learning_rate": 7.628284402581782e-06,
      "loss": 6.6362,
      "step": 80000
    },
    {
      "epoch": 1.31,
      "eval_accuracy": 0.10328286349116696,
      "eval_loss": 6.63375997543335,
      "eval_runtime": 579.6365,
      "eval_samples_per_second": 531.982,
      "eval_steps_per_second": 5.543,
      "step": 80000
    },
    {
      "epoch": 1.32,
      "learning_rate": 7.611343576885938e-06,
      "loss": 6.6324,
      "step": 80500
    },
    {
      "epoch": 1.33,
      "learning_rate": 7.5944027511900945e-06,
      "loss": 6.629,
      "step": 81000
    },
    {
      "epoch": 1.34,
      "learning_rate": 7.577461925494249e-06,
      "loss": 6.63,
      "step": 81500
    },
    {
      "epoch": 1.34,
      "learning_rate": 7.560521099798405e-06,
      "loss": 6.6273,
      "step": 82000
    },
    {
      "epoch": 1.35,
      "learning_rate": 7.54358027410256e-06,
      "loss": 6.6237,
      "step": 82500
    },
    {
      "epoch": 1.36,
      "learning_rate": 7.5266394484067164e-06,
      "loss": 6.6262,
      "step": 83000
    },
    {
      "epoch": 1.37,
      "learning_rate": 7.509698622710871e-06,
      "loss": 6.6216,
      "step": 83500
    },
    {
      "epoch": 1.38,
      "learning_rate": 7.492757797015027e-06,
      "loss": 6.6235,
      "step": 84000
    },
    {
      "epoch": 1.38,
      "learning_rate": 7.475816971319183e-06,
      "loss": 6.6192,
      "step": 84500
    },
    {
      "epoch": 1.39,
      "learning_rate": 7.458876145623338e-06,
      "loss": 6.6151,
      "step": 85000
    },
    {
      "epoch": 1.4,
      "learning_rate": 7.441935319927493e-06,
      "loss": 6.6159,
      "step": 85500
    },
    {
      "epoch": 1.41,
      "learning_rate": 7.42499449423165e-06,
      "loss": 6.6108,
      "step": 86000
    },
    {
      "epoch": 1.42,
      "learning_rate": 7.4080536685358054e-06,
      "loss": 6.6109,
      "step": 86500
    },
    {
      "epoch": 1.43,
      "learning_rate": 7.39111284283996e-06,
      "loss": 6.6169,
      "step": 87000
    },
    {
      "epoch": 1.43,
      "learning_rate": 7.374172017144117e-06,
      "loss": 6.6122,
      "step": 87500
    },
    {
      "epoch": 1.44,
      "learning_rate": 7.357231191448272e-06,
      "loss": 6.6108,
      "step": 88000
    },
    {
      "epoch": 1.45,
      "learning_rate": 7.340290365752427e-06,
      "loss": 6.6035,
      "step": 88500
    },
    {
      "epoch": 1.46,
      "learning_rate": 7.323349540056584e-06,
      "loss": 6.6058,
      "step": 89000
    },
    {
      "epoch": 1.47,
      "learning_rate": 7.306408714360739e-06,
      "loss": 6.6057,
      "step": 89500
    },
    {
      "epoch": 1.47,
      "learning_rate": 7.2894678886648936e-06,
      "loss": 6.5995,
      "step": 90000
    },
    {
      "epoch": 1.47,
      "eval_accuracy": 0.106496680198166,
      "eval_loss": 6.6016621589660645,
      "eval_runtime": 577.3676,
      "eval_samples_per_second": 534.072,
      "eval_steps_per_second": 5.565,
      "step": 90000
    },
    {
      "epoch": 1.48,
      "learning_rate": 7.272527062969049e-06,
      "loss": 6.5988,
      "step": 90500
    },
    {
      "epoch": 1.49,
      "learning_rate": 7.255586237273206e-06,
      "loss": 6.5977,
      "step": 91000
    },
    {
      "epoch": 1.5,
      "learning_rate": 7.238645411577361e-06,
      "loss": 6.6024,
      "step": 91500
    },
    {
      "epoch": 1.51,
      "learning_rate": 7.221704585881516e-06,
      "loss": 6.6002,
      "step": 92000
    },
    {
      "epoch": 1.52,
      "learning_rate": 7.204763760185672e-06,
      "loss": 6.5962,
      "step": 92500
    },
    {
      "epoch": 1.52,
      "learning_rate": 7.187822934489828e-06,
      "loss": 6.5949,
      "step": 93000
    },
    {
      "epoch": 1.53,
      "learning_rate": 7.1708821087939826e-06,
      "loss": 6.5965,
      "step": 93500
    },
    {
      "epoch": 1.54,
      "learning_rate": 7.153941283098139e-06,
      "loss": 6.5946,
      "step": 94000
    },
    {
      "epoch": 1.55,
      "learning_rate": 7.137000457402295e-06,
      "loss": 6.5981,
      "step": 94500
    },
    {
      "epoch": 1.56,
      "learning_rate": 7.12005963170645e-06,
      "loss": 6.5909,
      "step": 95000
    },
    {
      "epoch": 1.56,
      "learning_rate": 7.103118806010606e-06,
      "loss": 6.5874,
      "step": 95500
    },
    {
      "epoch": 1.57,
      "learning_rate": 7.086177980314761e-06,
      "loss": 6.5901,
      "step": 96000
    },
    {
      "epoch": 1.58,
      "learning_rate": 7.069237154618917e-06,
      "loss": 6.5871,
      "step": 96500
    },
    {
      "epoch": 1.59,
      "learning_rate": 7.052296328923072e-06,
      "loss": 6.5809,
      "step": 97000
    },
    {
      "epoch": 1.6,
      "learning_rate": 7.035355503227228e-06,
      "loss": 6.5825,
      "step": 97500
    },
    {
      "epoch": 1.61,
      "learning_rate": 7.018414677531383e-06,
      "loss": 6.5806,
      "step": 98000
    },
    {
      "epoch": 1.61,
      "learning_rate": 7.001473851835539e-06,
      "loss": 6.5793,
      "step": 98500
    },
    {
      "epoch": 1.62,
      "learning_rate": 6.984533026139695e-06,
      "loss": 6.5767,
      "step": 99000
    },
    {
      "epoch": 1.63,
      "learning_rate": 6.96759220044385e-06,
      "loss": 6.5799,
      "step": 99500
    },
    {
      "epoch": 1.64,
      "learning_rate": 6.950651374748006e-06,
      "loss": 6.5756,
      "step": 100000
    },
    {
      "epoch": 1.64,
      "eval_accuracy": 0.10915969642166277,
      "eval_loss": 6.5755228996276855,
      "eval_runtime": 577.0168,
      "eval_samples_per_second": 534.397,
      "eval_steps_per_second": 5.568,
      "step": 100000
    },
    {
      "epoch": 1.65,
      "learning_rate": 6.933710549052161e-06,
      "loss": 6.5778,
      "step": 100500
    },
    {
      "epoch": 1.65,
      "learning_rate": 6.916769723356317e-06,
      "loss": 6.5701,
      "step": 101000
    },
    {
      "epoch": 1.66,
      "learning_rate": 6.899828897660472e-06,
      "loss": 6.5787,
      "step": 101500
    },
    {
      "epoch": 1.67,
      "learning_rate": 6.8828880719646285e-06,
      "loss": 6.571,
      "step": 102000
    },
    {
      "epoch": 1.68,
      "learning_rate": 6.865947246268783e-06,
      "loss": 6.5726,
      "step": 102500
    },
    {
      "epoch": 1.69,
      "learning_rate": 6.849006420572939e-06,
      "loss": 6.5655,
      "step": 103000
    },
    {
      "epoch": 1.7,
      "learning_rate": 6.8320655948770955e-06,
      "loss": 6.5669,
      "step": 103500
    },
    {
      "epoch": 1.7,
      "learning_rate": 6.81512476918125e-06,
      "loss": 6.5667,
      "step": 104000
    },
    {
      "epoch": 1.71,
      "learning_rate": 6.798183943485406e-06,
      "loss": 6.5651,
      "step": 104500
    },
    {
      "epoch": 1.72,
      "learning_rate": 6.781243117789562e-06,
      "loss": 6.5635,
      "step": 105000
    },
    {
      "epoch": 1.73,
      "learning_rate": 6.7643022920937174e-06,
      "loss": 6.5636,
      "step": 105500
    },
    {
      "epoch": 1.74,
      "learning_rate": 6.747361466397872e-06,
      "loss": 6.5642,
      "step": 106000
    },
    {
      "epoch": 1.75,
      "learning_rate": 6.730420640702028e-06,
      "loss": 6.5581,
      "step": 106500
    },
    {
      "epoch": 1.75,
      "learning_rate": 6.7134798150061845e-06,
      "loss": 6.5631,
      "step": 107000
    },
    {
      "epoch": 1.76,
      "learning_rate": 6.696538989310339e-06,
      "loss": 6.5596,
      "step": 107500
    },
    {
      "epoch": 1.77,
      "learning_rate": 6.679598163614495e-06,
      "loss": 6.5579,
      "step": 108000
    },
    {
      "epoch": 1.78,
      "learning_rate": 6.662657337918651e-06,
      "loss": 6.5557,
      "step": 108500
    },
    {
      "epoch": 1.79,
      "learning_rate": 6.6457165122228064e-06,
      "loss": 6.5526,
      "step": 109000
    },
    {
      "epoch": 1.79,
      "learning_rate": 6.628775686526961e-06,
      "loss": 6.5576,
      "step": 109500
    },
    {
      "epoch": 1.8,
      "learning_rate": 6.611834860831118e-06,
      "loss": 6.5546,
      "step": 110000
    },
    {
      "epoch": 1.8,
      "eval_accuracy": 0.11176161378336055,
      "eval_loss": 6.552142143249512,
      "eval_runtime": 579.2632,
      "eval_samples_per_second": 532.324,
      "eval_steps_per_second": 5.547,
      "step": 110000
    },
    {
      "epoch": 1.81,
      "learning_rate": 6.594894035135273e-06,
      "loss": 6.5593,
      "step": 110500
    },
    {
      "epoch": 1.82,
      "learning_rate": 6.577953209439428e-06,
      "loss": 6.5531,
      "step": 111000
    },
    {
      "epoch": 1.83,
      "learning_rate": 6.561012383743585e-06,
      "loss": 6.5511,
      "step": 111500
    },
    {
      "epoch": 1.84,
      "learning_rate": 6.54407155804774e-06,
      "loss": 6.5478,
      "step": 112000
    },
    {
      "epoch": 1.84,
      "learning_rate": 6.5271307323518954e-06,
      "loss": 6.5477,
      "step": 112500
    },
    {
      "epoch": 1.85,
      "learning_rate": 6.510189906656051e-06,
      "loss": 6.5487,
      "step": 113000
    },
    {
      "epoch": 1.86,
      "learning_rate": 6.493249080960207e-06,
      "loss": 6.5505,
      "step": 113500
    },
    {
      "epoch": 1.87,
      "learning_rate": 6.476308255264362e-06,
      "loss": 6.5395,
      "step": 114000
    },
    {
      "epoch": 1.88,
      "learning_rate": 6.459367429568517e-06,
      "loss": 6.5459,
      "step": 114500
    },
    {
      "epoch": 1.88,
      "learning_rate": 6.442426603872674e-06,
      "loss": 6.545,
      "step": 115000
    },
    {
      "epoch": 1.89,
      "learning_rate": 6.425485778176829e-06,
      "loss": 6.5459,
      "step": 115500
    },
    {
      "epoch": 1.9,
      "learning_rate": 6.4085449524809836e-06,
      "loss": 6.5438,
      "step": 116000
    },
    {
      "epoch": 1.91,
      "learning_rate": 6.39160412678514e-06,
      "loss": 6.5368,
      "step": 116500
    },
    {
      "epoch": 1.92,
      "learning_rate": 6.374663301089296e-06,
      "loss": 6.5388,
      "step": 117000
    },
    {
      "epoch": 1.93,
      "learning_rate": 6.357722475393451e-06,
      "loss": 6.5326,
      "step": 117500
    },
    {
      "epoch": 1.93,
      "learning_rate": 6.340781649697607e-06,
      "loss": 6.5388,
      "step": 118000
    },
    {
      "epoch": 1.94,
      "learning_rate": 6.323840824001762e-06,
      "loss": 6.5329,
      "step": 118500
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.306899998305918e-06,
      "loss": 6.5384,
      "step": 119000
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.289959172610074e-06,
      "loss": 6.536,
      "step": 119500
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.273018346914229e-06,
      "loss": 6.5302,
      "step": 120000
    },
    {
      "epoch": 1.97,
      "eval_accuracy": 0.11375993998119388,
      "eval_loss": 6.533047199249268,
      "eval_runtime": 582.4997,
      "eval_samples_per_second": 529.367,
      "eval_steps_per_second": 5.516,
      "step": 120000
    },
    {
      "epoch": 1.97,
      "learning_rate": 6.256077521218385e-06,
      "loss": 6.5382,
      "step": 120500
    },
    {
      "epoch": 1.98,
      "learning_rate": 6.2391366955225405e-06,
      "loss": 6.5403,
      "step": 121000
    },
    {
      "epoch": 1.99,
      "learning_rate": 6.222195869826696e-06,
      "loss": 6.5275,
      "step": 121500
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.205255044130851e-06,
      "loss": 6.5277,
      "step": 122000
    },
    {
      "epoch": 2.01,
      "learning_rate": 6.1883142184350076e-06,
      "loss": 6.5236,
      "step": 122500
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.171373392739162e-06,
      "loss": 6.5305,
      "step": 123000
    },
    {
      "epoch": 2.02,
      "learning_rate": 6.154432567043318e-06,
      "loss": 6.5272,
      "step": 123500
    },
    {
      "epoch": 2.03,
      "learning_rate": 6.137491741347473e-06,
      "loss": 6.5298,
      "step": 124000
    },
    {
      "epoch": 2.04,
      "learning_rate": 6.1205509156516295e-06,
      "loss": 6.5216,
      "step": 124500
    },
    {
      "epoch": 2.05,
      "learning_rate": 6.103610089955785e-06,
      "loss": 6.5263,
      "step": 125000
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.08666926425994e-06,
      "loss": 6.5252,
      "step": 125500
    },
    {
      "epoch": 2.06,
      "learning_rate": 6.0697284385640965e-06,
      "loss": 6.5211,
      "step": 126000
    },
    {
      "epoch": 2.07,
      "learning_rate": 6.052787612868251e-06,
      "loss": 6.5225,
      "step": 126500
    },
    {
      "epoch": 2.08,
      "learning_rate": 6.035846787172407e-06,
      "loss": 6.5245,
      "step": 127000
    },
    {
      "epoch": 2.09,
      "learning_rate": 6.018905961476564e-06,
      "loss": 6.5303,
      "step": 127500
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.0019651357807185e-06,
      "loss": 6.5223,
      "step": 128000
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.985024310084873e-06,
      "loss": 6.5242,
      "step": 128500
    },
    {
      "epoch": 2.11,
      "learning_rate": 5.96808348438903e-06,
      "loss": 6.5167,
      "step": 129000
    },
    {
      "epoch": 2.12,
      "learning_rate": 5.9511426586931855e-06,
      "loss": 6.5181,
      "step": 129500
    },
    {
      "epoch": 2.13,
      "learning_rate": 5.93420183299734e-06,
      "loss": 6.5121,
      "step": 130000
    },
    {
      "epoch": 2.13,
      "eval_accuracy": 0.11564648111290464,
      "eval_loss": 6.5166473388671875,
      "eval_runtime": 579.517,
      "eval_samples_per_second": 532.091,
      "eval_steps_per_second": 5.544,
      "step": 130000
    },
    {
      "epoch": 2.14,
      "learning_rate": 5.917261007301497e-06,
      "loss": 6.5156,
      "step": 130500
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.900320181605652e-06,
      "loss": 6.5158,
      "step": 131000
    },
    {
      "epoch": 2.15,
      "learning_rate": 5.8833793559098074e-06,
      "loss": 6.5102,
      "step": 131500
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.866438530213962e-06,
      "loss": 6.5177,
      "step": 132000
    },
    {
      "epoch": 2.17,
      "learning_rate": 5.849497704518119e-06,
      "loss": 6.5134,
      "step": 132500
    },
    {
      "epoch": 2.18,
      "learning_rate": 5.8325568788222745e-06,
      "loss": 6.5118,
      "step": 133000
    },
    {
      "epoch": 2.19,
      "learning_rate": 5.815616053126429e-06,
      "loss": 6.5153,
      "step": 133500
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.798675227430586e-06,
      "loss": 6.5097,
      "step": 134000
    },
    {
      "epoch": 2.2,
      "learning_rate": 5.781734401734741e-06,
      "loss": 6.5077,
      "step": 134500
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.7647935760388964e-06,
      "loss": 6.5074,
      "step": 135000
    },
    {
      "epoch": 2.22,
      "learning_rate": 5.747852750343053e-06,
      "loss": 6.5001,
      "step": 135500
    },
    {
      "epoch": 2.23,
      "learning_rate": 5.730911924647208e-06,
      "loss": 6.5066,
      "step": 136000
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.713971098951363e-06,
      "loss": 6.5088,
      "step": 136500
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.697030273255519e-06,
      "loss": 6.5124,
      "step": 137000
    },
    {
      "epoch": 2.25,
      "learning_rate": 5.680089447559675e-06,
      "loss": 6.5062,
      "step": 137500
    },
    {
      "epoch": 2.26,
      "learning_rate": 5.66314862186383e-06,
      "loss": 6.5039,
      "step": 138000
    },
    {
      "epoch": 2.27,
      "learning_rate": 5.646207796167986e-06,
      "loss": 6.5027,
      "step": 138500
    },
    {
      "epoch": 2.28,
      "learning_rate": 5.629266970472141e-06,
      "loss": 6.5037,
      "step": 139000
    },
    {
      "epoch": 2.29,
      "learning_rate": 5.612326144776297e-06,
      "loss": 6.5029,
      "step": 139500
    },
    {
      "epoch": 2.29,
      "learning_rate": 5.595385319080452e-06,
      "loss": 6.5049,
      "step": 140000
    },
    {
      "epoch": 2.29,
      "eval_accuracy": 0.11736065991506915,
      "eval_loss": 6.500461578369141,
      "eval_runtime": 580.3357,
      "eval_samples_per_second": 531.341,
      "eval_steps_per_second": 5.536,
      "step": 140000
    },
    {
      "epoch": 2.3,
      "learning_rate": 5.578444493384608e-06,
      "loss": 6.5018,
      "step": 140500
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.561503667688764e-06,
      "loss": 6.5009,
      "step": 141000
    },
    {
      "epoch": 2.32,
      "learning_rate": 5.544562841992919e-06,
      "loss": 6.499,
      "step": 141500
    },
    {
      "epoch": 2.33,
      "learning_rate": 5.527622016297075e-06,
      "loss": 6.4954,
      "step": 142000
    },
    {
      "epoch": 2.33,
      "learning_rate": 5.51068119060123e-06,
      "loss": 6.5001,
      "step": 142500
    },
    {
      "epoch": 2.34,
      "learning_rate": 5.493740364905386e-06,
      "loss": 6.4959,
      "step": 143000
    },
    {
      "epoch": 2.35,
      "learning_rate": 5.4767995392095415e-06,
      "loss": 6.4934,
      "step": 143500
    },
    {
      "epoch": 2.36,
      "learning_rate": 5.459858713513697e-06,
      "loss": 6.4967,
      "step": 144000
    },
    {
      "epoch": 2.37,
      "learning_rate": 5.442917887817852e-06,
      "loss": 6.4968,
      "step": 144500
    },
    {
      "epoch": 2.38,
      "learning_rate": 5.4259770621220086e-06,
      "loss": 6.4955,
      "step": 145000
    },
    {
      "epoch": 2.38,
      "learning_rate": 5.409036236426164e-06,
      "loss": 6.4924,
      "step": 145500
    },
    {
      "epoch": 2.39,
      "learning_rate": 5.392095410730319e-06,
      "loss": 6.4879,
      "step": 146000
    },
    {
      "epoch": 2.4,
      "learning_rate": 5.375154585034476e-06,
      "loss": 6.496,
      "step": 146500
    },
    {
      "epoch": 2.41,
      "learning_rate": 5.3582137593386305e-06,
      "loss": 6.4943,
      "step": 147000
    },
    {
      "epoch": 2.42,
      "learning_rate": 5.341272933642786e-06,
      "loss": 6.4918,
      "step": 147500
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.324332107946941e-06,
      "loss": 6.4919,
      "step": 148000
    },
    {
      "epoch": 2.43,
      "learning_rate": 5.3073912822510975e-06,
      "loss": 6.4946,
      "step": 148500
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.290450456555252e-06,
      "loss": 6.4849,
      "step": 149000
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.273509630859408e-06,
      "loss": 6.4904,
      "step": 149500
    },
    {
      "epoch": 2.46,
      "learning_rate": 5.256568805163565e-06,
      "loss": 6.483,
      "step": 150000
    },
    {
      "epoch": 2.46,
      "eval_accuracy": 0.11898823350849988,
      "eval_loss": 6.486917495727539,
      "eval_runtime": 577.9045,
      "eval_samples_per_second": 533.576,
      "eval_steps_per_second": 5.56,
      "step": 150000
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.2396279794677195e-06,
      "loss": 6.4888,
      "step": 150500
    },
    {
      "epoch": 2.47,
      "learning_rate": 5.222687153771875e-06,
      "loss": 6.4936,
      "step": 151000
    },
    {
      "epoch": 2.48,
      "learning_rate": 5.205746328076031e-06,
      "loss": 6.4909,
      "step": 151500
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.1888055023801865e-06,
      "loss": 6.491,
      "step": 152000
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.171864676684341e-06,
      "loss": 6.4817,
      "step": 152500
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.154923850988498e-06,
      "loss": 6.4821,
      "step": 153000
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.137983025292654e-06,
      "loss": 6.4838,
      "step": 153500
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.1210421995968085e-06,
      "loss": 6.4837,
      "step": 154000
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.104101373900965e-06,
      "loss": 6.4828,
      "step": 154500
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.08716054820512e-06,
      "loss": 6.4833,
      "step": 155000
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.0702197225092755e-06,
      "loss": 6.4849,
      "step": 155500
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.053278896813432e-06,
      "loss": 6.4777,
      "step": 156000
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.036338071117587e-06,
      "loss": 6.4816,
      "step": 156500
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.019397245421742e-06,
      "loss": 6.4824,
      "step": 157000
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.0024564197258974e-06,
      "loss": 6.4811,
      "step": 157500
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.985515594030054e-06,
      "loss": 6.4743,
      "step": 158000
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.968574768334209e-06,
      "loss": 6.4761,
      "step": 158500
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.9516339426383645e-06,
      "loss": 6.4798,
      "step": 159000
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.93469311694252e-06,
      "loss": 6.4786,
      "step": 159500
    },
    {
      "epoch": 2.62,
      "learning_rate": 4.917752291246676e-06,
      "loss": 6.4757,
      "step": 160000
    },
    {
      "epoch": 2.62,
      "eval_accuracy": 0.12051287077285619,
      "eval_loss": 6.475541114807129,
      "eval_runtime": 576.3092,
      "eval_samples_per_second": 535.053,
      "eval_steps_per_second": 5.575,
      "step": 160000
    },
    {
      "epoch": 2.63,
      "learning_rate": 4.900811465550832e-06,
      "loss": 6.4783,
      "step": 160500
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.8838706398549864e-06,
      "loss": 6.4758,
      "step": 161000
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.866929814159143e-06,
      "loss": 6.4728,
      "step": 161500
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.849988988463299e-06,
      "loss": 6.4806,
      "step": 162000
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.8330481627674535e-06,
      "loss": 6.4739,
      "step": 162500
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.816107337071609e-06,
      "loss": 6.4728,
      "step": 163000
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.799166511375765e-06,
      "loss": 6.4741,
      "step": 163500
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.782225685679921e-06,
      "loss": 6.4759,
      "step": 164000
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.765284859984076e-06,
      "loss": 6.4751,
      "step": 164500
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.748344034288231e-06,
      "loss": 6.4757,
      "step": 165000
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.731403208592387e-06,
      "loss": 6.4718,
      "step": 165500
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.714462382896543e-06,
      "loss": 6.4718,
      "step": 166000
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.697521557200698e-06,
      "loss": 6.4685,
      "step": 166500
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.680580731504854e-06,
      "loss": 6.4625,
      "step": 167000
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.6636399058090096e-06,
      "loss": 6.4709,
      "step": 167500
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.646699080113165e-06,
      "loss": 6.473,
      "step": 168000
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.629758254417321e-06,
      "loss": 6.4671,
      "step": 168500
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.612817428721476e-06,
      "loss": 6.4675,
      "step": 169000
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.5958766030256315e-06,
      "loss": 6.4631,
      "step": 169500
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.578935777329788e-06,
      "loss": 6.4659,
      "step": 170000
    },
    {
      "epoch": 2.79,
      "eval_accuracy": 0.12186304935455677,
      "eval_loss": 6.4655327796936035,
      "eval_runtime": 579.7308,
      "eval_samples_per_second": 531.895,
      "eval_steps_per_second": 5.542,
      "step": 170000
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.561994951633943e-06,
      "loss": 6.4719,
      "step": 170500
    },
    {
      "epoch": 2.8,
      "learning_rate": 4.5450541259380986e-06,
      "loss": 6.462,
      "step": 171000
    },
    {
      "epoch": 2.81,
      "learning_rate": 4.528113300242254e-06,
      "loss": 6.4635,
      "step": 171500
    },
    {
      "epoch": 2.82,
      "learning_rate": 4.51117247454641e-06,
      "loss": 6.4677,
      "step": 172000
    },
    {
      "epoch": 2.83,
      "learning_rate": 4.494231648850566e-06,
      "loss": 6.4627,
      "step": 172500
    },
    {
      "epoch": 2.83,
      "learning_rate": 4.4772908231547205e-06,
      "loss": 6.4631,
      "step": 173000
    },
    {
      "epoch": 2.84,
      "learning_rate": 4.460349997458876e-06,
      "loss": 6.4643,
      "step": 173500
    },
    {
      "epoch": 2.85,
      "learning_rate": 4.443409171763033e-06,
      "loss": 6.4636,
      "step": 174000
    },
    {
      "epoch": 2.86,
      "learning_rate": 4.4264683460671875e-06,
      "loss": 6.4589,
      "step": 174500
    },
    {
      "epoch": 2.87,
      "learning_rate": 4.409527520371343e-06,
      "loss": 6.4613,
      "step": 175000
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.392586694675499e-06,
      "loss": 6.4595,
      "step": 175500
    },
    {
      "epoch": 2.88,
      "learning_rate": 4.375645868979655e-06,
      "loss": 6.462,
      "step": 176000
    },
    {
      "epoch": 2.89,
      "learning_rate": 4.35870504328381e-06,
      "loss": 6.4578,
      "step": 176500
    },
    {
      "epoch": 2.9,
      "learning_rate": 4.341764217587965e-06,
      "loss": 6.4657,
      "step": 177000
    },
    {
      "epoch": 2.91,
      "learning_rate": 4.324823391892121e-06,
      "loss": 6.4577,
      "step": 177500
    },
    {
      "epoch": 2.92,
      "learning_rate": 4.307882566196277e-06,
      "loss": 6.4585,
      "step": 178000
    },
    {
      "epoch": 2.92,
      "learning_rate": 4.290941740500432e-06,
      "loss": 6.4581,
      "step": 178500
    },
    {
      "epoch": 2.93,
      "learning_rate": 4.274000914804588e-06,
      "loss": 6.4562,
      "step": 179000
    },
    {
      "epoch": 2.94,
      "learning_rate": 4.257060089108744e-06,
      "loss": 6.4584,
      "step": 179500
    },
    {
      "epoch": 2.95,
      "learning_rate": 4.240119263412899e-06,
      "loss": 6.4527,
      "step": 180000
    },
    {
      "epoch": 2.95,
      "eval_accuracy": 0.12271129048456753,
      "eval_loss": 6.456941604614258,
      "eval_runtime": 579.0356,
      "eval_samples_per_second": 532.534,
      "eval_steps_per_second": 5.549,
      "step": 180000
    },
    {
      "epoch": 2.96,
      "learning_rate": 4.223178437717055e-06,
      "loss": 6.4589,
      "step": 180500
    },
    {
      "epoch": 2.97,
      "learning_rate": 4.20623761202121e-06,
      "loss": 6.4564,
      "step": 181000
    },
    {
      "epoch": 2.97,
      "learning_rate": 4.1892967863253655e-06,
      "loss": 6.4594,
      "step": 181500
    },
    {
      "epoch": 2.98,
      "learning_rate": 4.172355960629522e-06,
      "loss": 6.4552,
      "step": 182000
    },
    {
      "epoch": 2.99,
      "learning_rate": 4.155415134933677e-06,
      "loss": 6.4509,
      "step": 182500
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.138474309237833e-06,
      "loss": 6.451,
      "step": 183000
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.121533483541988e-06,
      "loss": 6.451,
      "step": 183500
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.104592657846144e-06,
      "loss": 6.4508,
      "step": 184000
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.0876518321503e-06,
      "loss": 6.4553,
      "step": 184500
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.0707110064544545e-06,
      "loss": 6.4524,
      "step": 185000
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.05377018075861e-06,
      "loss": 6.4533,
      "step": 185500
    },
    {
      "epoch": 3.05,
      "learning_rate": 4.036829355062766e-06,
      "loss": 6.4538,
      "step": 186000
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.019888529366922e-06,
      "loss": 6.4514,
      "step": 186500
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.002947703671077e-06,
      "loss": 6.4476,
      "step": 187000
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.986006877975233e-06,
      "loss": 6.451,
      "step": 187500
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.969066052279389e-06,
      "loss": 6.4492,
      "step": 188000
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.952125226583544e-06,
      "loss": 6.4528,
      "step": 188500
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.935184400887699e-06,
      "loss": 6.4519,
      "step": 189000
    },
    {
      "epoch": 3.11,
      "learning_rate": 3.918243575191855e-06,
      "loss": 6.4451,
      "step": 189500
    },
    {
      "epoch": 3.11,
      "learning_rate": 3.9013027494960106e-06,
      "loss": 6.4517,
      "step": 190000
    },
    {
      "epoch": 3.11,
      "eval_accuracy": 0.12365417396787551,
      "eval_loss": 6.447721481323242,
      "eval_runtime": 580.8795,
      "eval_samples_per_second": 530.843,
      "eval_steps_per_second": 5.531,
      "step": 190000
    },
    {
      "epoch": 3.12,
      "learning_rate": 3.884361923800166e-06,
      "loss": 6.4472,
      "step": 190500
    },
    {
      "epoch": 3.13,
      "learning_rate": 3.867421098104322e-06,
      "loss": 6.4504,
      "step": 191000
    },
    {
      "epoch": 3.14,
      "learning_rate": 3.850480272408478e-06,
      "loss": 6.4455,
      "step": 191500
    },
    {
      "epoch": 3.15,
      "learning_rate": 3.833539446712633e-06,
      "loss": 6.4466,
      "step": 192000
    },
    {
      "epoch": 3.15,
      "learning_rate": 3.816598621016789e-06,
      "loss": 6.4444,
      "step": 192500
    },
    {
      "epoch": 3.16,
      "learning_rate": 3.799657795320944e-06,
      "loss": 6.4472,
      "step": 193000
    },
    {
      "epoch": 3.17,
      "learning_rate": 3.7827169696251e-06,
      "loss": 6.4455,
      "step": 193500
    },
    {
      "epoch": 3.18,
      "learning_rate": 3.7657761439292557e-06,
      "loss": 6.4418,
      "step": 194000
    },
    {
      "epoch": 3.19,
      "learning_rate": 3.748835318233411e-06,
      "loss": 6.4473,
      "step": 194500
    },
    {
      "epoch": 3.2,
      "learning_rate": 3.7318944925375666e-06,
      "loss": 6.4455,
      "step": 195000
    },
    {
      "epoch": 3.2,
      "learning_rate": 3.7149536668417223e-06,
      "loss": 6.4405,
      "step": 195500
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.6980128411458776e-06,
      "loss": 6.4459,
      "step": 196000
    },
    {
      "epoch": 3.22,
      "learning_rate": 3.6810720154500333e-06,
      "loss": 6.4429,
      "step": 196500
    },
    {
      "epoch": 3.23,
      "learning_rate": 3.6641311897541886e-06,
      "loss": 6.4462,
      "step": 197000
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.6471903640583447e-06,
      "loss": 6.4394,
      "step": 197500
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.6302495383625004e-06,
      "loss": 6.4442,
      "step": 198000
    },
    {
      "epoch": 3.25,
      "learning_rate": 3.6133087126666556e-06,
      "loss": 6.4444,
      "step": 198500
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.5963678869708113e-06,
      "loss": 6.4461,
      "step": 199000
    },
    {
      "epoch": 3.27,
      "learning_rate": 3.579427061274967e-06,
      "loss": 6.4399,
      "step": 199500
    },
    {
      "epoch": 3.28,
      "learning_rate": 3.5624862355791223e-06,
      "loss": 6.441,
      "step": 200000
    },
    {
      "epoch": 3.28,
      "eval_accuracy": 0.1244753320060871,
      "eval_loss": 6.442170143127441,
      "eval_runtime": 579.963,
      "eval_samples_per_second": 531.682,
      "eval_steps_per_second": 5.54,
      "step": 200000
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.545545409883278e-06,
      "loss": 6.4407,
      "step": 200500
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.5286045841874332e-06,
      "loss": 6.4387,
      "step": 201000
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.511663758491589e-06,
      "loss": 6.439,
      "step": 201500
    },
    {
      "epoch": 3.31,
      "learning_rate": 3.494722932795745e-06,
      "loss": 6.4377,
      "step": 202000
    },
    {
      "epoch": 3.32,
      "learning_rate": 3.4777821070999003e-06,
      "loss": 6.443,
      "step": 202500
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.460841281404056e-06,
      "loss": 6.4395,
      "step": 203000
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.4439004557082117e-06,
      "loss": 6.4404,
      "step": 203500
    },
    {
      "epoch": 3.34,
      "learning_rate": 3.426959630012367e-06,
      "loss": 6.4401,
      "step": 204000
    },
    {
      "epoch": 3.35,
      "learning_rate": 3.4100188043165226e-06,
      "loss": 6.4374,
      "step": 204500
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.393077978620678e-06,
      "loss": 6.4435,
      "step": 205000
    },
    {
      "epoch": 3.37,
      "learning_rate": 3.3761371529248336e-06,
      "loss": 6.4369,
      "step": 205500
    },
    {
      "epoch": 3.38,
      "learning_rate": 3.3591963272289897e-06,
      "loss": 6.4419,
      "step": 206000
    },
    {
      "epoch": 3.38,
      "learning_rate": 3.342255501533145e-06,
      "loss": 6.4404,
      "step": 206500
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.3253146758373007e-06,
      "loss": 6.4404,
      "step": 207000
    },
    {
      "epoch": 3.4,
      "learning_rate": 3.3083738501414564e-06,
      "loss": 6.4378,
      "step": 207500
    },
    {
      "epoch": 3.41,
      "learning_rate": 3.2914330244456116e-06,
      "loss": 6.4377,
      "step": 208000
    },
    {
      "epoch": 3.42,
      "learning_rate": 3.2744921987497673e-06,
      "loss": 6.4323,
      "step": 208500
    },
    {
      "epoch": 3.42,
      "learning_rate": 3.2575513730539226e-06,
      "loss": 6.4343,
      "step": 209000
    },
    {
      "epoch": 3.43,
      "learning_rate": 3.2406105473580783e-06,
      "loss": 6.4345,
      "step": 209500
    },
    {
      "epoch": 3.44,
      "learning_rate": 3.2236697216622344e-06,
      "loss": 6.4385,
      "step": 210000
    },
    {
      "epoch": 3.44,
      "eval_accuracy": 0.12531759910787985,
      "eval_loss": 6.435288906097412,
      "eval_runtime": 581.2742,
      "eval_samples_per_second": 530.483,
      "eval_steps_per_second": 5.528,
      "step": 210000
    },
    {
      "epoch": 3.45,
      "learning_rate": 3.2067288959663897e-06,
      "loss": 6.4368,
      "step": 210500
    },
    {
      "epoch": 3.46,
      "learning_rate": 3.1897880702705454e-06,
      "loss": 6.4383,
      "step": 211000
    },
    {
      "epoch": 3.47,
      "learning_rate": 3.172847244574701e-06,
      "loss": 6.4348,
      "step": 211500
    },
    {
      "epoch": 3.47,
      "learning_rate": 3.1559064188788563e-06,
      "loss": 6.4386,
      "step": 212000
    },
    {
      "epoch": 3.48,
      "learning_rate": 3.138965593183012e-06,
      "loss": 6.4296,
      "step": 212500
    },
    {
      "epoch": 3.49,
      "learning_rate": 3.1220247674871673e-06,
      "loss": 6.438,
      "step": 213000
    },
    {
      "epoch": 3.5,
      "learning_rate": 3.105083941791323e-06,
      "loss": 6.4325,
      "step": 213500
    },
    {
      "epoch": 3.51,
      "learning_rate": 3.088143116095479e-06,
      "loss": 6.4364,
      "step": 214000
    },
    {
      "epoch": 3.51,
      "learning_rate": 3.071202290399634e-06,
      "loss": 6.4331,
      "step": 214500
    },
    {
      "epoch": 3.52,
      "learning_rate": 3.05426146470379e-06,
      "loss": 6.4295,
      "step": 215000
    },
    {
      "epoch": 3.53,
      "learning_rate": 3.0373206390079457e-06,
      "loss": 6.4375,
      "step": 215500
    },
    {
      "epoch": 3.54,
      "learning_rate": 3.020379813312101e-06,
      "loss": 6.4322,
      "step": 216000
    },
    {
      "epoch": 3.55,
      "learning_rate": 3.0034389876162567e-06,
      "loss": 6.4302,
      "step": 216500
    },
    {
      "epoch": 3.56,
      "learning_rate": 2.986498161920412e-06,
      "loss": 6.4347,
      "step": 217000
    },
    {
      "epoch": 3.56,
      "learning_rate": 2.9695573362245676e-06,
      "loss": 6.4364,
      "step": 217500
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.9526165105287238e-06,
      "loss": 6.4287,
      "step": 218000
    },
    {
      "epoch": 3.58,
      "learning_rate": 2.9356756848328786e-06,
      "loss": 6.4294,
      "step": 218500
    },
    {
      "epoch": 3.59,
      "learning_rate": 2.9187348591370347e-06,
      "loss": 6.4362,
      "step": 219000
    },
    {
      "epoch": 3.6,
      "learning_rate": 2.9017940334411904e-06,
      "loss": 6.4329,
      "step": 219500
    },
    {
      "epoch": 3.6,
      "learning_rate": 2.8848532077453457e-06,
      "loss": 6.4308,
      "step": 220000
    },
    {
      "epoch": 3.6,
      "eval_accuracy": 0.12563817634622382,
      "eval_loss": 6.4294962882995605,
      "eval_runtime": 579.5631,
      "eval_samples_per_second": 532.049,
      "eval_steps_per_second": 5.544,
      "step": 220000
    },
    {
      "epoch": 3.61,
      "learning_rate": 2.8679123820495014e-06,
      "loss": 6.4297,
      "step": 220500
    },
    {
      "epoch": 3.62,
      "learning_rate": 2.8509715563536566e-06,
      "loss": 6.4304,
      "step": 221000
    },
    {
      "epoch": 3.63,
      "learning_rate": 2.8340307306578123e-06,
      "loss": 6.4286,
      "step": 221500
    },
    {
      "epoch": 3.64,
      "learning_rate": 2.817089904961968e-06,
      "loss": 6.4261,
      "step": 222000
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.8001490792661233e-06,
      "loss": 6.4251,
      "step": 222500
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.7832082535702794e-06,
      "loss": 6.4292,
      "step": 223000
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.766267427874435e-06,
      "loss": 6.4283,
      "step": 223500
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.7493266021785904e-06,
      "loss": 6.4298,
      "step": 224000
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.732385776482746e-06,
      "loss": 6.4304,
      "step": 224500
    },
    {
      "epoch": 3.69,
      "learning_rate": 2.7154449507869017e-06,
      "loss": 6.4327,
      "step": 225000
    },
    {
      "epoch": 3.69,
      "learning_rate": 2.698504125091057e-06,
      "loss": 6.4276,
      "step": 225500
    },
    {
      "epoch": 3.7,
      "learning_rate": 2.6815632993952127e-06,
      "loss": 6.4274,
      "step": 226000
    },
    {
      "epoch": 3.71,
      "learning_rate": 2.664622473699368e-06,
      "loss": 6.4275,
      "step": 226500
    },
    {
      "epoch": 3.72,
      "learning_rate": 2.647681648003524e-06,
      "loss": 6.4293,
      "step": 227000
    },
    {
      "epoch": 3.73,
      "learning_rate": 2.6307408223076798e-06,
      "loss": 6.4234,
      "step": 227500
    },
    {
      "epoch": 3.74,
      "learning_rate": 2.613799996611835e-06,
      "loss": 6.4279,
      "step": 228000
    },
    {
      "epoch": 3.74,
      "learning_rate": 2.5968591709159907e-06,
      "loss": 6.4268,
      "step": 228500
    },
    {
      "epoch": 3.75,
      "learning_rate": 2.5799183452201464e-06,
      "loss": 6.4242,
      "step": 229000
    },
    {
      "epoch": 3.76,
      "learning_rate": 2.5629775195243017e-06,
      "loss": 6.4244,
      "step": 229500
    },
    {
      "epoch": 3.77,
      "learning_rate": 2.5460366938284574e-06,
      "loss": 6.4188,
      "step": 230000
    },
    {
      "epoch": 3.77,
      "eval_accuracy": 0.1263907170690432,
      "eval_loss": 6.4250078201293945,
      "eval_runtime": 580.0942,
      "eval_samples_per_second": 531.562,
      "eval_steps_per_second": 5.539,
      "step": 230000
    },
    {
      "epoch": 3.78,
      "learning_rate": 2.5290958681326126e-06,
      "loss": 6.4244,
      "step": 230500
    },
    {
      "epoch": 3.79,
      "learning_rate": 2.5121550424367688e-06,
      "loss": 6.4291,
      "step": 231000
    },
    {
      "epoch": 3.79,
      "learning_rate": 2.495214216740924e-06,
      "loss": 6.4239,
      "step": 231500
    },
    {
      "epoch": 3.8,
      "learning_rate": 2.4782733910450797e-06,
      "loss": 6.4243,
      "step": 232000
    },
    {
      "epoch": 3.81,
      "learning_rate": 2.4613325653492354e-06,
      "loss": 6.4286,
      "step": 232500
    },
    {
      "epoch": 3.82,
      "learning_rate": 2.4443917396533907e-06,
      "loss": 6.4253,
      "step": 233000
    },
    {
      "epoch": 3.83,
      "learning_rate": 2.4274509139575464e-06,
      "loss": 6.4302,
      "step": 233500
    },
    {
      "epoch": 3.83,
      "learning_rate": 2.410510088261702e-06,
      "loss": 6.4235,
      "step": 234000
    },
    {
      "epoch": 3.84,
      "learning_rate": 2.3935692625658577e-06,
      "loss": 6.4265,
      "step": 234500
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.376628436870013e-06,
      "loss": 6.4245,
      "step": 235000
    },
    {
      "epoch": 3.86,
      "learning_rate": 2.3596876111741687e-06,
      "loss": 6.4242,
      "step": 235500
    },
    {
      "epoch": 3.87,
      "learning_rate": 2.3427467854783244e-06,
      "loss": 6.4231,
      "step": 236000
    },
    {
      "epoch": 3.88,
      "learning_rate": 2.32580595978248e-06,
      "loss": 6.42,
      "step": 236500
    },
    {
      "epoch": 3.88,
      "learning_rate": 2.3088651340866354e-06,
      "loss": 6.4292,
      "step": 237000
    },
    {
      "epoch": 3.89,
      "learning_rate": 2.291924308390791e-06,
      "loss": 6.4175,
      "step": 237500
    },
    {
      "epoch": 3.9,
      "learning_rate": 2.2749834826949467e-06,
      "loss": 6.4251,
      "step": 238000
    },
    {
      "epoch": 3.91,
      "learning_rate": 2.2580426569991024e-06,
      "loss": 6.4172,
      "step": 238500
    },
    {
      "epoch": 3.92,
      "learning_rate": 2.2411018313032577e-06,
      "loss": 6.4208,
      "step": 239000
    },
    {
      "epoch": 3.92,
      "learning_rate": 2.2241610056074134e-06,
      "loss": 6.4268,
      "step": 239500
    },
    {
      "epoch": 3.93,
      "learning_rate": 2.207220179911569e-06,
      "loss": 6.422,
      "step": 240000
    },
    {
      "epoch": 3.93,
      "eval_accuracy": 0.1269057358624372,
      "eval_loss": 6.42128849029541,
      "eval_runtime": 579.3125,
      "eval_samples_per_second": 532.279,
      "eval_steps_per_second": 5.546,
      "step": 240000
    },
    {
      "epoch": 3.94,
      "learning_rate": 2.1902793542157248e-06,
      "loss": 6.4199,
      "step": 240500
    },
    {
      "epoch": 3.95,
      "learning_rate": 2.17333852851988e-06,
      "loss": 6.4236,
      "step": 241000
    },
    {
      "epoch": 3.96,
      "learning_rate": 2.1563977028240357e-06,
      "loss": 6.4197,
      "step": 241500
    },
    {
      "epoch": 3.97,
      "learning_rate": 2.1394568771281914e-06,
      "loss": 6.4204,
      "step": 242000
    },
    {
      "epoch": 3.97,
      "learning_rate": 2.122516051432347e-06,
      "loss": 6.4214,
      "step": 242500
    },
    {
      "epoch": 3.98,
      "learning_rate": 2.1055752257365024e-06,
      "loss": 6.4213,
      "step": 243000
    },
    {
      "epoch": 3.99,
      "learning_rate": 2.088634400040658e-06,
      "loss": 6.4188,
      "step": 243500
    },
    {
      "epoch": 4.0,
      "learning_rate": 2.0716935743448138e-06,
      "loss": 6.4194,
      "step": 244000
    },
    {
      "epoch": 4.01,
      "learning_rate": 2.0547527486489694e-06,
      "loss": 6.4167,
      "step": 244500
    },
    {
      "epoch": 4.01,
      "learning_rate": 2.0378119229531247e-06,
      "loss": 6.4198,
      "step": 245000
    },
    {
      "epoch": 4.02,
      "learning_rate": 2.0208710972572804e-06,
      "loss": 6.4244,
      "step": 245500
    },
    {
      "epoch": 4.03,
      "learning_rate": 2.003930271561436e-06,
      "loss": 6.4197,
      "step": 246000
    },
    {
      "epoch": 4.04,
      "learning_rate": 1.986989445865592e-06,
      "loss": 6.4191,
      "step": 246500
    },
    {
      "epoch": 4.05,
      "learning_rate": 1.970048620169747e-06,
      "loss": 6.4194,
      "step": 247000
    },
    {
      "epoch": 4.06,
      "learning_rate": 1.9531077944739027e-06,
      "loss": 6.4171,
      "step": 247500
    },
    {
      "epoch": 4.06,
      "learning_rate": 1.9361669687780584e-06,
      "loss": 6.4143,
      "step": 248000
    },
    {
      "epoch": 4.07,
      "learning_rate": 1.919226143082214e-06,
      "loss": 6.4153,
      "step": 248500
    },
    {
      "epoch": 4.08,
      "learning_rate": 1.9022853173863696e-06,
      "loss": 6.4186,
      "step": 249000
    },
    {
      "epoch": 4.09,
      "learning_rate": 1.885344491690525e-06,
      "loss": 6.4185,
      "step": 249500
    },
    {
      "epoch": 4.1,
      "learning_rate": 1.8684036659946808e-06,
      "loss": 6.416,
      "step": 250000
    },
    {
      "epoch": 4.1,
      "eval_accuracy": 0.12734837097500631,
      "eval_loss": 6.418042182922363,
      "eval_runtime": 585.9741,
      "eval_samples_per_second": 526.228,
      "eval_steps_per_second": 5.483,
      "step": 250000
    },
    {
      "epoch": 4.1,
      "learning_rate": 1.8514628402988363e-06,
      "loss": 6.4176,
      "step": 250500
    },
    {
      "epoch": 4.11,
      "learning_rate": 1.8345220146029917e-06,
      "loss": 6.4132,
      "step": 251000
    },
    {
      "epoch": 4.12,
      "learning_rate": 1.8175811889071474e-06,
      "loss": 6.4166,
      "step": 251500
    },
    {
      "epoch": 4.13,
      "learning_rate": 1.8006403632113031e-06,
      "loss": 6.4178,
      "step": 252000
    },
    {
      "epoch": 4.14,
      "learning_rate": 1.7836995375154586e-06,
      "loss": 6.4156,
      "step": 252500
    },
    {
      "epoch": 4.15,
      "learning_rate": 1.766758711819614e-06,
      "loss": 6.4196,
      "step": 253000
    },
    {
      "epoch": 4.15,
      "learning_rate": 1.7498178861237698e-06,
      "loss": 6.4178,
      "step": 253500
    },
    {
      "epoch": 4.16,
      "learning_rate": 1.7328770604279255e-06,
      "loss": 6.4098,
      "step": 254000
    },
    {
      "epoch": 4.17,
      "learning_rate": 1.715936234732081e-06,
      "loss": 6.4145,
      "step": 254500
    },
    {
      "epoch": 4.18,
      "learning_rate": 1.6989954090362364e-06,
      "loss": 6.4181,
      "step": 255000
    },
    {
      "epoch": 4.19,
      "learning_rate": 1.6820545833403921e-06,
      "loss": 6.4162,
      "step": 255500
    },
    {
      "epoch": 4.19,
      "learning_rate": 1.6651137576445478e-06,
      "loss": 6.4153,
      "step": 256000
    },
    {
      "epoch": 4.2,
      "learning_rate": 1.6481729319487033e-06,
      "loss": 6.4178,
      "step": 256500
    },
    {
      "epoch": 4.21,
      "learning_rate": 1.6312321062528588e-06,
      "loss": 6.4189,
      "step": 257000
    },
    {
      "epoch": 4.22,
      "learning_rate": 1.6142912805570147e-06,
      "loss": 6.4101,
      "step": 257500
    },
    {
      "epoch": 4.23,
      "learning_rate": 1.5973504548611701e-06,
      "loss": 6.4161,
      "step": 258000
    },
    {
      "epoch": 4.24,
      "learning_rate": 1.5804096291653256e-06,
      "loss": 6.4114,
      "step": 258500
    },
    {
      "epoch": 4.24,
      "learning_rate": 1.563468803469481e-06,
      "loss": 6.4155,
      "step": 259000
    },
    {
      "epoch": 4.25,
      "learning_rate": 1.546527977773637e-06,
      "loss": 6.4155,
      "step": 259500
    },
    {
      "epoch": 4.26,
      "learning_rate": 1.5295871520777925e-06,
      "loss": 6.4215,
      "step": 260000
    },
    {
      "epoch": 4.26,
      "eval_accuracy": 0.1275747812907389,
      "eval_loss": 6.415066719055176,
      "eval_runtime": 582.2355,
      "eval_samples_per_second": 529.607,
      "eval_steps_per_second": 5.518,
      "step": 260000
    },
    {
      "epoch": 4.27,
      "learning_rate": 1.512646326381948e-06,
      "loss": 6.4148,
      "step": 260500
    },
    {
      "epoch": 4.28,
      "learning_rate": 1.4957055006861034e-06,
      "loss": 6.4118,
      "step": 261000
    },
    {
      "epoch": 4.28,
      "learning_rate": 1.4787646749902593e-06,
      "loss": 6.4186,
      "step": 261500
    },
    {
      "epoch": 4.29,
      "learning_rate": 1.4618238492944148e-06,
      "loss": 6.4076,
      "step": 262000
    },
    {
      "epoch": 4.3,
      "learning_rate": 1.4448830235985703e-06,
      "loss": 6.4219,
      "step": 262500
    },
    {
      "epoch": 4.31,
      "learning_rate": 1.4279421979027258e-06,
      "loss": 6.414,
      "step": 263000
    },
    {
      "epoch": 4.32,
      "learning_rate": 1.4110013722068817e-06,
      "loss": 6.4137,
      "step": 263500
    },
    {
      "epoch": 4.33,
      "learning_rate": 1.3940605465110372e-06,
      "loss": 6.4136,
      "step": 264000
    },
    {
      "epoch": 4.33,
      "learning_rate": 1.3771197208151926e-06,
      "loss": 6.4129,
      "step": 264500
    },
    {
      "epoch": 4.34,
      "learning_rate": 1.3601788951193481e-06,
      "loss": 6.4167,
      "step": 265000
    },
    {
      "epoch": 4.35,
      "learning_rate": 1.343238069423504e-06,
      "loss": 6.4174,
      "step": 265500
    },
    {
      "epoch": 4.36,
      "learning_rate": 1.3262972437276595e-06,
      "loss": 6.4218,
      "step": 266000
    },
    {
      "epoch": 4.37,
      "learning_rate": 1.309356418031815e-06,
      "loss": 6.4154,
      "step": 266500
    },
    {
      "epoch": 4.37,
      "learning_rate": 1.2924155923359705e-06,
      "loss": 6.4114,
      "step": 267000
    },
    {
      "epoch": 4.38,
      "learning_rate": 1.2754747666401264e-06,
      "loss": 6.4143,
      "step": 267500
    },
    {
      "epoch": 4.39,
      "learning_rate": 1.2585339409442818e-06,
      "loss": 6.4149,
      "step": 268000
    },
    {
      "epoch": 4.4,
      "learning_rate": 1.2415931152484373e-06,
      "loss": 6.4121,
      "step": 268500
    },
    {
      "epoch": 4.41,
      "learning_rate": 1.224652289552593e-06,
      "loss": 6.4181,
      "step": 269000
    },
    {
      "epoch": 4.42,
      "learning_rate": 1.2077114638567485e-06,
      "loss": 6.4134,
      "step": 269500
    },
    {
      "epoch": 4.42,
      "learning_rate": 1.1907706381609042e-06,
      "loss": 6.4135,
      "step": 270000
    },
    {
      "epoch": 4.42,
      "eval_accuracy": 0.1277067626725093,
      "eval_loss": 6.414210319519043,
      "eval_runtime": 581.2376,
      "eval_samples_per_second": 530.516,
      "eval_steps_per_second": 5.528,
      "step": 270000
    },
    {
      "epoch": 4.43,
      "learning_rate": 1.1738298124650597e-06,
      "loss": 6.4139,
      "step": 270500
    },
    {
      "epoch": 4.44,
      "learning_rate": 1.1568889867692153e-06,
      "loss": 6.413,
      "step": 271000
    },
    {
      "epoch": 4.45,
      "learning_rate": 1.1399481610733708e-06,
      "loss": 6.4193,
      "step": 271500
    },
    {
      "epoch": 4.46,
      "learning_rate": 1.1230073353775265e-06,
      "loss": 6.4136,
      "step": 272000
    },
    {
      "epoch": 4.47,
      "learning_rate": 1.106066509681682e-06,
      "loss": 6.4192,
      "step": 272500
    },
    {
      "epoch": 4.47,
      "learning_rate": 1.0891256839858377e-06,
      "loss": 6.4081,
      "step": 273000
    },
    {
      "epoch": 4.48,
      "learning_rate": 1.0721848582899932e-06,
      "loss": 6.4109,
      "step": 273500
    },
    {
      "epoch": 4.49,
      "learning_rate": 1.0552440325941489e-06,
      "loss": 6.4099,
      "step": 274000
    },
    {
      "epoch": 4.5,
      "learning_rate": 1.0383032068983043e-06,
      "loss": 6.4165,
      "step": 274500
    },
    {
      "epoch": 4.51,
      "learning_rate": 1.02136238120246e-06,
      "loss": 6.4108,
      "step": 275000
    },
    {
      "epoch": 4.51,
      "learning_rate": 1.0044215555066155e-06,
      "loss": 6.4143,
      "step": 275500
    },
    {
      "epoch": 4.52,
      "learning_rate": 9.87480729810771e-07,
      "loss": 6.4173,
      "step": 276000
    },
    {
      "epoch": 4.53,
      "learning_rate": 9.705399041149267e-07,
      "loss": 6.4111,
      "step": 276500
    },
    {
      "epoch": 4.54,
      "learning_rate": 9.535990784190823e-07,
      "loss": 6.4102,
      "step": 277000
    },
    {
      "epoch": 4.55,
      "learning_rate": 9.366582527232377e-07,
      "loss": 6.4107,
      "step": 277500
    },
    {
      "epoch": 4.56,
      "learning_rate": 9.197174270273934e-07,
      "loss": 6.4149,
      "step": 278000
    },
    {
      "epoch": 4.56,
      "learning_rate": 9.027766013315489e-07,
      "loss": 6.411,
      "step": 278500
    },
    {
      "epoch": 4.57,
      "learning_rate": 8.858357756357046e-07,
      "loss": 6.4137,
      "step": 279000
    },
    {
      "epoch": 4.58,
      "learning_rate": 8.688949499398601e-07,
      "loss": 6.4162,
      "step": 279500
    },
    {
      "epoch": 4.59,
      "learning_rate": 8.519541242440158e-07,
      "loss": 6.4138,
      "step": 280000
    },
    {
      "epoch": 4.59,
      "eval_accuracy": 0.1280778613427266,
      "eval_loss": 6.411779880523682,
      "eval_runtime": 583.4582,
      "eval_samples_per_second": 528.497,
      "eval_steps_per_second": 5.507,
      "step": 280000
    },
    {
      "epoch": 4.6,
      "learning_rate": 8.350132985481713e-07,
      "loss": 6.4077,
      "step": 280500
    },
    {
      "epoch": 4.6,
      "learning_rate": 8.180724728523269e-07,
      "loss": 6.4138,
      "step": 281000
    },
    {
      "epoch": 4.61,
      "learning_rate": 8.011316471564824e-07,
      "loss": 6.4154,
      "step": 281500
    },
    {
      "epoch": 4.62,
      "learning_rate": 7.841908214606381e-07,
      "loss": 6.4132,
      "step": 282000
    },
    {
      "epoch": 4.63,
      "learning_rate": 7.672499957647936e-07,
      "loss": 6.4178,
      "step": 282500
    },
    {
      "epoch": 4.64,
      "learning_rate": 7.503091700689493e-07,
      "loss": 6.4101,
      "step": 283000
    },
    {
      "epoch": 4.65,
      "learning_rate": 7.333683443731048e-07,
      "loss": 6.4108,
      "step": 283500
    },
    {
      "epoch": 4.65,
      "learning_rate": 7.164275186772605e-07,
      "loss": 6.4117,
      "step": 284000
    },
    {
      "epoch": 4.66,
      "learning_rate": 6.994866929814159e-07,
      "loss": 6.4084,
      "step": 284500
    },
    {
      "epoch": 4.67,
      "learning_rate": 6.825458672855715e-07,
      "loss": 6.405,
      "step": 285000
    },
    {
      "epoch": 4.68,
      "learning_rate": 6.656050415897271e-07,
      "loss": 6.4092,
      "step": 285500
    },
    {
      "epoch": 4.69,
      "learning_rate": 6.486642158938827e-07,
      "loss": 6.4062,
      "step": 286000
    },
    {
      "epoch": 4.69,
      "learning_rate": 6.317233901980383e-07,
      "loss": 6.4138,
      "step": 286500
    },
    {
      "epoch": 4.7,
      "learning_rate": 6.147825645021939e-07,
      "loss": 6.4104,
      "step": 287000
    },
    {
      "epoch": 4.71,
      "learning_rate": 5.978417388063494e-07,
      "loss": 6.4089,
      "step": 287500
    },
    {
      "epoch": 4.72,
      "learning_rate": 5.80900913110505e-07,
      "loss": 6.4151,
      "step": 288000
    },
    {
      "epoch": 4.73,
      "learning_rate": 5.639600874146606e-07,
      "loss": 6.4163,
      "step": 288500
    },
    {
      "epoch": 4.74,
      "learning_rate": 5.470192617188162e-07,
      "loss": 6.4104,
      "step": 289000
    },
    {
      "epoch": 4.74,
      "learning_rate": 5.300784360229718e-07,
      "loss": 6.4114,
      "step": 289500
    },
    {
      "epoch": 4.75,
      "learning_rate": 5.131376103271274e-07,
      "loss": 6.41,
      "step": 290000
    },
    {
      "epoch": 4.75,
      "eval_accuracy": 0.12825801981118998,
      "eval_loss": 6.409651756286621,
      "eval_runtime": 581.1289,
      "eval_samples_per_second": 530.615,
      "eval_steps_per_second": 5.529,
      "step": 290000
    },
    {
      "epoch": 4.76,
      "learning_rate": 4.96196784631283e-07,
      "loss": 6.4122,
      "step": 290500
    },
    {
      "epoch": 4.77,
      "learning_rate": 4.792559589354385e-07,
      "loss": 6.4089,
      "step": 291000
    },
    {
      "epoch": 4.78,
      "learning_rate": 4.623151332395941e-07,
      "loss": 6.4111,
      "step": 291500
    },
    {
      "epoch": 4.78,
      "learning_rate": 4.453743075437497e-07,
      "loss": 6.4113,
      "step": 292000
    },
    {
      "epoch": 4.79,
      "learning_rate": 4.284334818479053e-07,
      "loss": 6.4153,
      "step": 292500
    },
    {
      "epoch": 4.8,
      "learning_rate": 4.114926561520609e-07,
      "loss": 6.4115,
      "step": 293000
    },
    {
      "epoch": 4.81,
      "learning_rate": 3.9455183045621646e-07,
      "loss": 6.4051,
      "step": 293500
    },
    {
      "epoch": 4.82,
      "learning_rate": 3.7761100476037205e-07,
      "loss": 6.4099,
      "step": 294000
    },
    {
      "epoch": 4.83,
      "learning_rate": 3.6067017906452763e-07,
      "loss": 6.4147,
      "step": 294500
    },
    {
      "epoch": 4.83,
      "learning_rate": 3.437293533686832e-07,
      "loss": 6.4115,
      "step": 295000
    },
    {
      "epoch": 4.84,
      "learning_rate": 3.267885276728388e-07,
      "loss": 6.4147,
      "step": 295500
    },
    {
      "epoch": 4.85,
      "learning_rate": 3.098477019769944e-07,
      "loss": 6.4098,
      "step": 296000
    },
    {
      "epoch": 4.86,
      "learning_rate": 2.9290687628114997e-07,
      "loss": 6.4109,
      "step": 296500
    },
    {
      "epoch": 4.87,
      "learning_rate": 2.7596605058530556e-07,
      "loss": 6.4089,
      "step": 297000
    },
    {
      "epoch": 4.87,
      "learning_rate": 2.5902522488946114e-07,
      "loss": 6.4057,
      "step": 297500
    },
    {
      "epoch": 4.88,
      "learning_rate": 2.4208439919361673e-07,
      "loss": 6.4067,
      "step": 298000
    },
    {
      "epoch": 4.89,
      "learning_rate": 2.2514357349777229e-07,
      "loss": 6.4083,
      "step": 298500
    },
    {
      "epoch": 4.9,
      "learning_rate": 2.0820274780192787e-07,
      "loss": 6.4096,
      "step": 299000
    },
    {
      "epoch": 4.91,
      "learning_rate": 1.9126192210608346e-07,
      "loss": 6.413,
      "step": 299500
    },
    {
      "epoch": 4.92,
      "learning_rate": 1.7432109641023904e-07,
      "loss": 6.4114,
      "step": 300000
    },
    {
      "epoch": 4.92,
      "eval_accuracy": 0.12813361905263143,
      "eval_loss": 6.410257339477539,
      "eval_runtime": 580.2144,
      "eval_samples_per_second": 531.452,
      "eval_steps_per_second": 5.538,
      "step": 300000
    },
    {
      "epoch": 4.92,
      "learning_rate": 1.5738027071439465e-07,
      "loss": 6.4054,
      "step": 300500
    },
    {
      "epoch": 4.93,
      "learning_rate": 1.404394450185502e-07,
      "loss": 6.4107,
      "step": 301000
    },
    {
      "epoch": 4.94,
      "learning_rate": 1.234986193227058e-07,
      "loss": 6.4057,
      "step": 301500
    },
    {
      "epoch": 4.95,
      "learning_rate": 1.0655779362686138e-07,
      "loss": 6.4131,
      "step": 302000
    },
    {
      "epoch": 4.96,
      "learning_rate": 8.961696793101697e-08,
      "loss": 6.409,
      "step": 302500
    },
    {
      "epoch": 4.96,
      "learning_rate": 7.267614223517255e-08,
      "loss": 6.4124,
      "step": 303000
    },
    {
      "epoch": 4.97,
      "learning_rate": 5.5735316539328136e-08,
      "loss": 6.4129,
      "step": 303500
    },
    {
      "epoch": 4.98,
      "learning_rate": 3.8794490843483714e-08,
      "loss": 6.4072,
      "step": 304000
    },
    {
      "epoch": 4.99,
      "learning_rate": 2.18536651476393e-08,
      "loss": 6.4046,
      "step": 304500
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.912839451794881e-09,
      "loss": 6.412,
      "step": 305000
    },
    {
      "epoch": 5.0,
      "step": 305145,
      "total_flos": 5.402149293706445e+16,
      "train_loss": 6.686526771036234,
      "train_runtime": 114519.7668,
      "train_samples_per_second": 255.797,
      "train_steps_per_second": 2.665
    }
  ],
  "logging_steps": 500,
  "max_steps": 305145,
  "num_train_epochs": 5,
  "save_steps": 10000,
  "total_flos": 5.402149293706445e+16,
  "trial_name": null,
  "trial_params": null
}