diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,12888 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 7.473575572084416,
+  "global_step": 210000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.999221493020686e-05,
+      "loss": 3.8388,
+      "step": 100
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99844298604137e-05,
+      "loss": 2.944,
+      "step": 200
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.997664479062055e-05,
+      "loss": 2.8362,
+      "step": 300
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.99688597208274e-05,
+      "loss": 2.782,
+      "step": 400
+    },
+    {
+      "epoch": 0.0,
+      "learning_rate": 4.9961074651034254e-05,
+      "loss": 2.7643,
+      "step": 500
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9626307922272054e-05,
+      "loss": 2.7412,
+      "step": 600
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.956402590931739e-05,
+      "loss": 2.7286,
+      "step": 700
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.9501743896362735e-05,
+      "loss": 2.7117,
+      "step": 800
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.943946188340807e-05,
+      "loss": 2.6993,
+      "step": 900
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9377179870453416e-05,
+      "loss": 2.6795,
+      "step": 1000
+    },
+    {
+      "epoch": 0.06,
+      "eval_loss": 2.5517239570617676,
+      "eval_runtime": 3812.4069,
+      "eval_samples_per_second": 838.582,
+      "eval_steps_per_second": 0.819,
+      "step": 1000
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.9314897857498754e-05,
+      "loss": 2.6627,
+      "step": 1100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.92526158445441e-05,
+      "loss": 2.652,
+      "step": 1200
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.919033383158944e-05,
+      "loss": 2.6428,
+      "step": 1300
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.912805181863478e-05,
+      "loss": 2.6378,
+      "step": 1400
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.9065769805680123e-05,
+      "loss": 2.6297,
+      "step": 1500
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.900348779272546e-05,
+      "loss": 2.622,
+      "step": 1600
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.8941205779770805e-05,
+      "loss": 2.6128,
+      "step": 1700
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.887892376681615e-05,
+      "loss": 2.6052,
+      "step": 1800
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.8816641753861486e-05,
+      "loss": 2.594,
+      "step": 1900
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.875435974090683e-05,
+      "loss": 2.5815,
+      "step": 2000
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.869207772795217e-05,
+      "loss": 2.5688,
+      "step": 2100
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.862979571499751e-05,
+      "loss": 2.5553,
+      "step": 2200
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.8567513702042856e-05,
+      "loss": 2.5421,
+      "step": 2300
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.850523168908819e-05,
+      "loss": 2.5244,
+      "step": 2400
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.844294967613354e-05,
+      "loss": 2.5127,
+      "step": 2500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.8380667663178875e-05,
+      "loss": 2.4942,
+      "step": 2600
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.831838565022422e-05,
+      "loss": 2.4787,
+      "step": 2700
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.8256103637269556e-05,
+      "loss": 2.4644,
+      "step": 2800
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.81938216243149e-05,
+      "loss": 2.4422,
+      "step": 2900
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.8131539611360244e-05,
+      "loss": 2.4266,
+      "step": 3000
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.806925759840558e-05,
+      "loss": 2.402,
+      "step": 3100
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.8006975585450926e-05,
+      "loss": 2.388,
+      "step": 3200
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.794469357249626e-05,
+      "loss": 2.3705,
+      "step": 3300
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.788241155954161e-05,
+      "loss": 2.3557,
+      "step": 3400
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.782012954658695e-05,
+      "loss": 2.3361,
+      "step": 3500
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.775784753363229e-05,
+      "loss": 2.3186,
+      "step": 3600
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.769556552067763e-05,
+      "loss": 2.301,
+      "step": 3700
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.763328350772297e-05,
+      "loss": 2.2857,
+      "step": 3800
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.7571001494768314e-05,
+      "loss": 2.2735,
+      "step": 3900
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.750871948181366e-05,
+      "loss": 2.2549,
+      "step": 4000
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.7446437468858996e-05,
+      "loss": 2.245,
+      "step": 4100
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.738415545590434e-05,
+      "loss": 2.2276,
+      "step": 4200
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.732187344294968e-05,
+      "loss": 2.2133,
+      "step": 4300
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.725959142999502e-05,
+      "loss": 2.1995,
+      "step": 4400
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.719730941704036e-05,
+      "loss": 2.1916,
+      "step": 4500
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.71350274040857e-05,
+      "loss": 2.1726,
+      "step": 4600
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.707274539113105e-05,
+      "loss": 2.1621,
+      "step": 4700
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.7010463378176384e-05,
+      "loss": 2.1505,
+      "step": 4800
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.694818136522173e-05,
+      "loss": 2.1406,
+      "step": 4900
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.6885899352267066e-05,
+      "loss": 2.1286,
+      "step": 5000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.682361733931241e-05,
+      "loss": 2.1165,
+      "step": 5100
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.6761335326357754e-05,
+      "loss": 2.1072,
+      "step": 5200
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.669905331340309e-05,
+      "loss": 2.098,
+      "step": 5300
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.6636771300448435e-05,
+      "loss": 2.0874,
+      "step": 5400
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.657448928749377e-05,
+      "loss": 2.0759,
+      "step": 5500
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.651220727453912e-05,
+      "loss": 2.0631,
+      "step": 5600
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.644992526158446e-05,
+      "loss": 2.0547,
+      "step": 5700
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.63876432486298e-05,
+      "loss": 2.0457,
+      "step": 5800
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.632536123567514e-05,
+      "loss": 2.039,
+      "step": 5900
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.626307922272048e-05,
+      "loss": 2.0286,
+      "step": 6000
+    },
+    {
+      "epoch": 0.37,
+      "eval_loss": 1.8759167194366455,
+      "eval_runtime": 3772.4979,
+      "eval_samples_per_second": 847.453,
+      "eval_steps_per_second": 0.111,
+      "step": 6000
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.6200797209765824e-05,
+      "loss": 2.02,
+      "step": 6100
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.613851519681116e-05,
+      "loss": 2.0093,
+      "step": 6200
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.6076233183856505e-05,
+      "loss": 2.0037,
+      "step": 6300
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.601395117090185e-05,
+      "loss": 1.9936,
+      "step": 6400
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.595166915794719e-05,
+      "loss": 1.9883,
+      "step": 6500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.588938714499253e-05,
+      "loss": 1.9801,
+      "step": 6600
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.582710513203787e-05,
+      "loss": 1.9706,
+      "step": 6700
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.576482311908321e-05,
+      "loss": 1.9608,
+      "step": 6800
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.5702541106128556e-05,
+      "loss": 1.9552,
+      "step": 6900
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.5640259093173894e-05,
+      "loss": 1.9484,
+      "step": 7000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.557797708021924e-05,
+      "loss": 1.9403,
+      "step": 7100
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.5515695067264575e-05,
+      "loss": 1.9353,
+      "step": 7200
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.545341305430992e-05,
+      "loss": 1.9266,
+      "step": 7300
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.5391131041355263e-05,
+      "loss": 1.9183,
+      "step": 7400
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.53288490284006e-05,
+      "loss": 1.9128,
+      "step": 7500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.5266567015445945e-05,
+      "loss": 1.9033,
+      "step": 7600
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.520428500249128e-05,
+      "loss": 1.9033,
+      "step": 7700
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.5142002989536626e-05,
+      "loss": 1.8912,
+      "step": 7800
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.5079720976581964e-05,
+      "loss": 1.884,
+      "step": 7900
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.501743896362731e-05,
+      "loss": 1.8806,
+      "step": 8000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.495515695067265e-05,
+      "loss": 1.874,
+      "step": 8100
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.489287493771799e-05,
+      "loss": 1.8705,
+      "step": 8200
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.4830592924763333e-05,
+      "loss": 1.8617,
+      "step": 8300
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.476831091180867e-05,
+      "loss": 1.8551,
+      "step": 8400
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.4706028898854015e-05,
+      "loss": 1.8511,
+      "step": 8500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.464374688589936e-05,
+      "loss": 1.8471,
+      "step": 8600
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.4581464872944696e-05,
+      "loss": 1.8361,
+      "step": 8700
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.451918285999004e-05,
+      "loss": 1.8358,
+      "step": 8800
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.445690084703538e-05,
+      "loss": 1.8295,
+      "step": 8900
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.439461883408072e-05,
+      "loss": 1.8226,
+      "step": 9000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.4332336821126066e-05,
+      "loss": 1.8202,
+      "step": 9100
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.42700548081714e-05,
+      "loss": 1.814,
+      "step": 9200
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.420777279521675e-05,
+      "loss": 1.8097,
+      "step": 9300
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.4145490782262085e-05,
+      "loss": 1.8006,
+      "step": 9400
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.408320876930743e-05,
+      "loss": 1.7972,
+      "step": 9500
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.4020926756352766e-05,
+      "loss": 1.7909,
+      "step": 9600
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.395864474339811e-05,
+      "loss": 1.7888,
+      "step": 9700
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.3896362730443454e-05,
+      "loss": 1.7852,
+      "step": 9800
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.383408071748879e-05,
+      "loss": 1.7806,
+      "step": 9900
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.3771798704534136e-05,
+      "loss": 1.7741,
+      "step": 10000
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.370951669157947e-05,
+      "loss": 1.7699,
+      "step": 10100
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.364723467862482e-05,
+      "loss": 1.7663,
+      "step": 10200
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.358495266567016e-05,
+      "loss": 1.7591,
+      "step": 10300
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.35226706527155e-05,
+      "loss": 1.7592,
+      "step": 10400
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.346038863976084e-05,
+      "loss": 1.7527,
+      "step": 10500
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.339810662680618e-05,
+      "loss": 1.7481,
+      "step": 10600
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.3335824613851524e-05,
+      "loss": 1.7421,
+      "step": 10700
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.327354260089686e-05,
+      "loss": 1.7388,
+      "step": 10800
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.3211260587942206e-05,
+      "loss": 1.7372,
+      "step": 10900
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.314897857498755e-05,
+      "loss": 1.7322,
+      "step": 11000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.308669656203289e-05,
+      "loss": 1.7293,
+      "step": 11100
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.302441454907823e-05,
+      "loss": 1.7228,
+      "step": 11200
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.296213253612357e-05,
+      "loss": 1.7208,
+      "step": 11300
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.289985052316891e-05,
+      "loss": 1.7209,
+      "step": 11400
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.283756851021426e-05,
+      "loss": 1.7147,
+      "step": 11500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.2775286497259594e-05,
+      "loss": 1.709,
+      "step": 11600
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.271300448430494e-05,
+      "loss": 1.7073,
+      "step": 11700
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.2650722471350276e-05,
+      "loss": 1.7032,
+      "step": 11800
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.258844045839562e-05,
+      "loss": 1.6995,
+      "step": 11900
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.2526158445440964e-05,
+      "loss": 1.6961,
+      "step": 12000
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 1.5560799837112427,
+      "eval_runtime": 3764.7391,
+      "eval_samples_per_second": 849.2,
+      "eval_steps_per_second": 0.111,
+      "step": 12000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.24638764324863e-05,
+      "loss": 1.6951,
+      "step": 12100
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.2401594419531645e-05,
+      "loss": 1.6896,
+      "step": 12200
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.233931240657698e-05,
+      "loss": 1.6875,
+      "step": 12300
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.227703039362233e-05,
+      "loss": 1.6824,
+      "step": 12400
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.2214748380667664e-05,
+      "loss": 1.6791,
+      "step": 12500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.215246636771301e-05,
+      "loss": 1.6768,
+      "step": 12600
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.209018435475835e-05,
+      "loss": 1.6697,
+      "step": 12700
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.202790234180369e-05,
+      "loss": 1.6715,
+      "step": 12800
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.1965620328849034e-05,
+      "loss": 1.664,
+      "step": 12900
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.190333831589437e-05,
+      "loss": 1.6636,
+      "step": 13000
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.1841056302939715e-05,
+      "loss": 1.6587,
+      "step": 13100
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.177877428998506e-05,
+      "loss": 1.6541,
+      "step": 13200
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.17164922770304e-05,
+      "loss": 1.6552,
+      "step": 13300
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.165421026407574e-05,
+      "loss": 1.6456,
+      "step": 13400
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.159192825112108e-05,
+      "loss": 1.6481,
+      "step": 13500
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.152964623816642e-05,
+      "loss": 1.6425,
+      "step": 13600
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.1467364225211766e-05,
+      "loss": 1.6406,
+      "step": 13700
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.1405082212257104e-05,
+      "loss": 1.6374,
+      "step": 13800
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.134280019930245e-05,
+      "loss": 1.6321,
+      "step": 13900
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.1280518186347785e-05,
+      "loss": 1.6306,
+      "step": 14000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.121823617339313e-05,
+      "loss": 1.6305,
+      "step": 14100
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.115595416043847e-05,
+      "loss": 1.6254,
+      "step": 14200
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.109367214748381e-05,
+      "loss": 1.6239,
+      "step": 14300
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.1031390134529155e-05,
+      "loss": 1.6219,
+      "step": 14400
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.096910812157449e-05,
+      "loss": 1.6187,
+      "step": 14500
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.0906826108619836e-05,
+      "loss": 1.614,
+      "step": 14600
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.0844544095665174e-05,
+      "loss": 1.6101,
+      "step": 14700
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.078226208271052e-05,
+      "loss": 1.6102,
+      "step": 14800
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.071998006975586e-05,
+      "loss": 1.6081,
+      "step": 14900
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.06576980568012e-05,
+      "loss": 1.6,
+      "step": 15000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.059541604384654e-05,
+      "loss": 1.602,
+      "step": 15100
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.053313403089188e-05,
+      "loss": 1.5951,
+      "step": 15200
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.0470852017937225e-05,
+      "loss": 1.595,
+      "step": 15300
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.040857000498256e-05,
+      "loss": 1.5945,
+      "step": 15400
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.0346287992027906e-05,
+      "loss": 1.5904,
+      "step": 15500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.028400597907325e-05,
+      "loss": 1.5918,
+      "step": 15600
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.022172396611859e-05,
+      "loss": 1.5856,
+      "step": 15700
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.015944195316393e-05,
+      "loss": 1.5831,
+      "step": 15800
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.009715994020927e-05,
+      "loss": 1.5831,
+      "step": 15900
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.003487792725461e-05,
+      "loss": 1.5787,
+      "step": 16000
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 3.997259591429995e-05,
+      "loss": 1.5836,
+      "step": 16100
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 3.9910313901345295e-05,
+      "loss": 1.5703,
+      "step": 16200
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.984803188839064e-05,
+      "loss": 1.5706,
+      "step": 16300
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 3.9785749875435976e-05,
+      "loss": 1.5717,
+      "step": 16400
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.972346786248132e-05,
+      "loss": 1.5688,
+      "step": 16500
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.966118584952666e-05,
+      "loss": 1.5626,
+      "step": 16600
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 3.9598903836572e-05,
+      "loss": 1.5619,
+      "step": 16700
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 3.953662182361734e-05,
+      "loss": 1.5606,
+      "step": 16800
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 3.947433981066268e-05,
+      "loss": 1.5612,
+      "step": 16900
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 3.941205779770803e-05,
+      "loss": 1.5571,
+      "step": 17000
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 3.9349775784753365e-05,
+      "loss": 1.5539,
+      "step": 17100
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 3.928749377179871e-05,
+      "loss": 1.5487,
+      "step": 17200
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.9225211758844046e-05,
+      "loss": 1.5509,
+      "step": 17300
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 3.916292974588939e-05,
+      "loss": 1.5489,
+      "step": 17400
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 3.9100647732934734e-05,
+      "loss": 1.5446,
+      "step": 17500
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.903836571998007e-05,
+      "loss": 1.5426,
+      "step": 17600
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 3.8976083707025416e-05,
+      "loss": 1.5438,
+      "step": 17700
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 3.891380169407075e-05,
+      "loss": 1.5386,
+      "step": 17800
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 3.88515196811161e-05,
+      "loss": 1.5423,
+      "step": 17900
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 3.8789237668161435e-05,
+      "loss": 1.5398,
+      "step": 18000
+    },
+    {
+      "epoch": 1.12,
+      "eval_loss": 1.4072346687316895,
+      "eval_runtime": 3765.404,
+      "eval_samples_per_second": 849.05,
+      "eval_steps_per_second": 0.111,
+      "step": 18000
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 3.872695565520678e-05,
+      "loss": 1.5325,
+      "step": 18100
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 3.866467364225212e-05,
+      "loss": 1.5294,
+      "step": 18200
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 3.860239162929746e-05,
+      "loss": 1.5304,
+      "step": 18300
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 3.8540109616342804e-05,
+      "loss": 1.5273,
+      "step": 18400
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 3.847782760338814e-05,
+      "loss": 1.5241,
+      "step": 18500
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 3.8415545590433486e-05,
+      "loss": 1.5253,
+      "step": 18600
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 3.835326357747882e-05,
+      "loss": 1.5244,
+      "step": 18700
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 3.829098156452417e-05,
+      "loss": 1.5183,
+      "step": 18800
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 3.822869955156951e-05,
+      "loss": 1.5203,
+      "step": 18900
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 3.816641753861485e-05,
+      "loss": 1.5192,
+      "step": 19000
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 3.810413552566019e-05,
+      "loss": 1.5149,
+      "step": 19100
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 3.804185351270553e-05,
+      "loss": 1.5129,
+      "step": 19200
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 3.7979571499750874e-05,
+      "loss": 1.5089,
+      "step": 19300
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 3.791728948679621e-05,
+      "loss": 1.5125,
+      "step": 19400
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 3.7855007473841556e-05,
+      "loss": 1.509,
+      "step": 19500
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 3.77927254608869e-05,
+      "loss": 1.5044,
+      "step": 19600
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 3.773044344793224e-05,
+      "loss": 1.5038,
+      "step": 19700
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 3.766816143497758e-05,
+      "loss": 1.5041,
+      "step": 19800
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 3.760587942202292e-05,
+      "loss": 1.5035,
+      "step": 19900
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 3.754359740906826e-05,
+      "loss": 1.5,
+      "step": 20000
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 3.74813153961136e-05,
+      "loss": 1.4976,
+      "step": 20100
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 3.7419033383158944e-05,
+      "loss": 1.4956,
+      "step": 20200
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 3.735675137020429e-05,
+      "loss": 1.4972,
+      "step": 20300
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 3.7294469357249626e-05,
+      "loss": 1.4933,
+      "step": 20400
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 3.723218734429497e-05,
+      "loss": 1.4929,
+      "step": 20500
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 3.716990533134031e-05,
+      "loss": 1.4885,
+      "step": 20600
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 3.710762331838565e-05,
+      "loss": 1.4897,
+      "step": 20700
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 3.704534130543099e-05,
+      "loss": 1.4888,
+      "step": 20800
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 3.698305929247633e-05,
+      "loss": 1.4871,
+      "step": 20900
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 3.692077727952168e-05,
+      "loss": 1.4816,
+      "step": 21000
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 3.6858495266567014e-05,
+      "loss": 1.4798,
+      "step": 21100
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 3.679621325361236e-05,
+      "loss": 1.4803,
+      "step": 21200
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.6733931240657695e-05,
+      "loss": 1.477,
+      "step": 21300
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 3.667164922770304e-05,
+      "loss": 1.4763,
+      "step": 21400
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 3.660936721474838e-05,
+      "loss": 1.4753,
+      "step": 21500
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 3.654708520179372e-05,
+      "loss": 1.4732,
+      "step": 21600
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 3.6484803188839065e-05,
+      "loss": 1.4735,
+      "step": 21700
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 3.64225211758844e-05,
+      "loss": 1.4739,
+      "step": 21800
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 3.6360239162929747e-05,
+      "loss": 1.4713,
+      "step": 21900
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 3.6297957149975084e-05,
+      "loss": 1.4708,
+      "step": 22000
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 3.623567513702043e-05,
+      "loss": 1.4662,
+      "step": 22100
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 3.617339312406577e-05,
+      "loss": 1.4667,
+      "step": 22200
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 3.611111111111111e-05,
+      "loss": 1.4632,
+      "step": 22300
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 3.6048829098156454e-05,
+      "loss": 1.4614,
+      "step": 22400
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 3.598654708520179e-05,
+      "loss": 1.4566,
+      "step": 22500
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 3.5924265072247135e-05,
+      "loss": 1.4599,
+      "step": 22600
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 3.586198305929247e-05,
+      "loss": 1.46,
+      "step": 22700
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 3.5799701046337816e-05,
+      "loss": 1.4546,
+      "step": 22800
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 3.573741903338316e-05,
+      "loss": 1.4555,
+      "step": 22900
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 3.56751370204285e-05,
+      "loss": 1.4561,
+      "step": 23000
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.561285500747384e-05,
+      "loss": 1.4547,
+      "step": 23100
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 3.555057299451918e-05,
+      "loss": 1.4519,
+      "step": 23200
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 3.5488290981564523e-05,
+      "loss": 1.4516,
+      "step": 23300
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 3.542600896860987e-05,
+      "loss": 1.4482,
+      "step": 23400
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 3.5363726955655205e-05,
+      "loss": 1.4505,
+      "step": 23500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 3.530144494270055e-05,
+      "loss": 1.4449,
+      "step": 23600
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 3.5239162929745886e-05,
+      "loss": 1.4458,
+      "step": 23700
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 3.517688091679123e-05,
+      "loss": 1.4447,
+      "step": 23800
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 3.5114598903836575e-05,
+      "loss": 1.4413,
+      "step": 23900
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 3.505231689088191e-05,
+      "loss": 1.4424,
+      "step": 24000
+    },
+    {
+      "epoch": 1.49,
+      "eval_loss": 1.3201874494552612,
+      "eval_runtime": 3766.4667,
+      "eval_samples_per_second": 848.81,
+      "eval_steps_per_second": 0.111,
+      "step": 24000
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 3.4990034877927256e-05,
+      "loss": 1.4434,
+      "step": 24100
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 3.4927752864972593e-05,
+      "loss": 1.4395,
+      "step": 24200
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 3.486547085201794e-05,
+      "loss": 1.4401,
+      "step": 24300
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 3.4803188839063275e-05,
+      "loss": 1.4324,
+      "step": 24400
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 3.474090682610862e-05,
+      "loss": 1.4342,
+      "step": 24500
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 3.467862481315396e-05,
+      "loss": 1.4349,
+      "step": 24600
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 3.46163428001993e-05,
+      "loss": 1.4306,
+      "step": 24700
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 3.4554060787244645e-05,
+      "loss": 1.4331,
+      "step": 24800
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 3.449177877428998e-05,
+      "loss": 1.4294,
+      "step": 24900
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 3.4429496761335326e-05,
+      "loss": 1.4288,
+      "step": 25000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 3.436721474838067e-05,
+      "loss": 1.4266,
+      "step": 25100
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 3.430493273542601e-05,
+      "loss": 1.4267,
+      "step": 25200
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 3.424265072247135e-05,
+      "loss": 1.4264,
+      "step": 25300
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 3.418036870951669e-05,
+      "loss": 1.4262,
+      "step": 25400
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 3.411808669656203e-05,
+      "loss": 1.4219,
+      "step": 25500
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 3.405580468360738e-05,
+      "loss": 1.4211,
+      "step": 25600
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 3.3993522670652714e-05,
+      "loss": 1.4216,
+      "step": 25700
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 3.393124065769806e-05,
+      "loss": 1.419,
+      "step": 25800
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 3.3868958644743396e-05,
+      "loss": 1.4209,
+      "step": 25900
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 3.380667663178874e-05,
+      "loss": 1.4182,
+      "step": 26000
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.374439461883408e-05,
+      "loss": 1.417,
+      "step": 26100
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 3.368211260587942e-05,
+      "loss": 1.4154,
+      "step": 26200
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.3619830592924766e-05,
+      "loss": 1.4106,
+      "step": 26300
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.35575485799701e-05,
+      "loss": 1.4133,
+      "step": 26400
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.349526656701545e-05,
+      "loss": 1.4118,
+      "step": 26500
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.3432984554060784e-05,
+      "loss": 1.4102,
+      "step": 26600
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.337070254110613e-05,
+      "loss": 1.4133,
+      "step": 26700
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.330842052815147e-05,
+      "loss": 1.4065,
+      "step": 26800
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.324613851519681e-05,
+      "loss": 1.41,
+      "step": 26900
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.3183856502242154e-05,
+      "loss": 1.4057,
+      "step": 27000
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.312157448928749e-05,
+      "loss": 1.4048,
+      "step": 27100
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 3.3059292476332835e-05,
+      "loss": 1.4031,
+      "step": 27200
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 3.299701046337818e-05,
+      "loss": 1.4039,
+      "step": 27300
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 3.293472845042352e-05,
+      "loss": 1.4053,
+      "step": 27400
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 3.287244643746886e-05,
+      "loss": 1.4039,
+      "step": 27500
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 3.28101644245142e-05,
+      "loss": 1.3993,
+      "step": 27600
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 3.274788241155954e-05,
+      "loss": 1.4015,
+      "step": 27700
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 3.268560039860488e-05,
+      "loss": 1.3998,
+      "step": 27800
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 3.2623318385650224e-05,
+      "loss": 1.3974,
+      "step": 27900
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 3.256103637269557e-05,
+      "loss": 1.3942,
+      "step": 28000
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 3.2498754359740905e-05,
+      "loss": 1.3995,
+      "step": 28100
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 3.243647234678625e-05,
+      "loss": 1.3976,
+      "step": 28200
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 3.237419033383159e-05,
+      "loss": 1.3932,
+      "step": 28300
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 3.231190832087693e-05,
+      "loss": 1.3897,
+      "step": 28400
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 3.2249626307922275e-05,
+      "loss": 1.3919,
+      "step": 28500
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 3.218734429496761e-05,
+      "loss": 1.3912,
+      "step": 28600
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 3.2125062282012957e-05,
+      "loss": 1.3895,
+      "step": 28700
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 3.2062780269058294e-05,
+      "loss": 1.3893,
+      "step": 28800
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 3.200049825610364e-05,
+      "loss": 1.3901,
+      "step": 28900
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 3.193821624314898e-05,
+      "loss": 1.3839,
+      "step": 29000
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 3.187593423019432e-05,
+      "loss": 1.3853,
+      "step": 29100
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 3.1813652217239664e-05,
+      "loss": 1.3853,
+      "step": 29200
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 3.1751370204285e-05,
+      "loss": 1.3829,
+      "step": 29300
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 3.1689088191330345e-05,
+      "loss": 1.3833,
+      "step": 29400
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 3.162680617837568e-05,
+      "loss": 1.3822,
+      "step": 29500
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 3.1564524165421026e-05,
+      "loss": 1.3796,
+      "step": 29600
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 3.150224215246637e-05,
+      "loss": 1.3807,
+      "step": 29700
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 3.143996013951171e-05,
+      "loss": 1.3764,
+      "step": 29800
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 3.137767812655705e-05,
+      "loss": 1.3786,
+      "step": 29900
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 3.131539611360239e-05,
+      "loss": 1.3776,
+      "step": 30000
+    },
+    {
+      "epoch": 1.87,
+      "eval_loss": 1.2612946033477783,
+      "eval_runtime": 3765.9008,
+      "eval_samples_per_second": 848.938,
+      "eval_steps_per_second": 0.111,
+      "step": 30000
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 3.1253114100647733e-05,
+      "loss": 1.3748,
+      "step": 30100
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 3.119083208769308e-05,
+      "loss": 1.3744,
+      "step": 30200
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.1128550074738415e-05,
+      "loss": 1.3776,
+      "step": 30300
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.106626806178376e-05,
+      "loss": 1.3751,
+      "step": 30400
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 3.1003986048829096e-05,
+      "loss": 1.3728,
+      "step": 30500
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 3.094170403587444e-05,
+      "loss": 1.3721,
+      "step": 30600
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 3.087942202291978e-05,
+      "loss": 1.3759,
+      "step": 30700
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 3.081714000996512e-05,
+      "loss": 1.3731,
+      "step": 30800
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 3.0754857997010466e-05,
+      "loss": 1.3736,
+      "step": 30900
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 3.06925759840558e-05,
+      "loss": 1.3705,
+      "step": 31000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 3.063029397110115e-05,
+      "loss": 1.3671,
+      "step": 31100
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 3.0568011958146485e-05,
+      "loss": 1.3671,
+      "step": 31200
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 3.050572994519183e-05,
+      "loss": 1.3647,
+      "step": 31300
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 3.0443447932237173e-05,
+      "loss": 1.3654,
+      "step": 31400
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 3.038116591928251e-05,
+      "loss": 1.3648,
+      "step": 31500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.0318883906327854e-05,
+      "loss": 1.3665,
+      "step": 31600
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.0256601893373192e-05,
+      "loss": 1.361,
+      "step": 31700
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 3.0194319880418536e-05,
+      "loss": 1.3635,
+      "step": 31800
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 3.013203786746388e-05,
+      "loss": 1.3646,
+      "step": 31900
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 3.0069755854509217e-05,
+      "loss": 1.3592,
+      "step": 32000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 3.000747384155456e-05,
+      "loss": 1.3619,
+      "step": 32100
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.99451918285999e-05,
+      "loss": 1.3653,
+      "step": 32200
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 2.9882909815645243e-05,
+      "loss": 1.3559,
+      "step": 32300
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.982062780269058e-05,
+      "loss": 1.3582,
+      "step": 32400
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 2.9758345789735924e-05,
+      "loss": 1.3544,
+      "step": 32500
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 2.969606377678127e-05,
+      "loss": 1.3574,
+      "step": 32600
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 2.9633781763826606e-05,
+      "loss": 1.3538,
+      "step": 32700
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 2.957149975087195e-05,
+      "loss": 1.3537,
+      "step": 32800
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 2.9509217737917287e-05,
+      "loss": 1.3509,
+      "step": 32900
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 2.944693572496263e-05,
+      "loss": 1.3511,
+      "step": 33000
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 2.9384653712007976e-05,
+      "loss": 1.3508,
+      "step": 33100
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 2.9322371699053313e-05,
+      "loss": 1.3503,
+      "step": 33200
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 2.9260089686098657e-05,
+      "loss": 1.3487,
+      "step": 33300
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 2.9197807673143994e-05,
+      "loss": 1.3453,
+      "step": 33400
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 2.913552566018934e-05,
+      "loss": 1.345,
+      "step": 33500
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 2.9073243647234683e-05,
+      "loss": 1.3482,
+      "step": 33600
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 2.901096163428002e-05,
+      "loss": 1.3499,
+      "step": 33700
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 2.8948679621325364e-05,
+      "loss": 1.3463,
+      "step": 33800
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 2.88863976083707e-05,
+      "loss": 1.3464,
+      "step": 33900
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 2.8824115595416045e-05,
+      "loss": 1.3478,
+      "step": 34000
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 2.8761833582461383e-05,
+      "loss": 1.3403,
+      "step": 34100
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 2.8699551569506727e-05,
+      "loss": 1.3387,
+      "step": 34200
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 2.863726955655207e-05,
+      "loss": 1.3411,
+      "step": 34300
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 2.857498754359741e-05,
+      "loss": 1.3408,
+      "step": 34400
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 2.8512705530642752e-05,
+      "loss": 1.3422,
+      "step": 34500
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 2.845042351768809e-05,
+      "loss": 1.3438,
+      "step": 34600
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 2.8388141504733434e-05,
+      "loss": 1.3375,
+      "step": 34700
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 2.8325859491778778e-05,
+      "loss": 1.3398,
+      "step": 34800
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 2.8263577478824115e-05,
+      "loss": 1.3363,
+      "step": 34900
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 2.820129546586946e-05,
+      "loss": 1.3354,
+      "step": 35000
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 2.8139013452914797e-05,
+      "loss": 1.3382,
+      "step": 35100
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 2.807673143996014e-05,
+      "loss": 1.3353,
+      "step": 35200
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 2.8014449427005485e-05,
+      "loss": 1.3369,
+      "step": 35300
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 2.7952167414050822e-05,
+      "loss": 1.3373,
+      "step": 35400
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 2.7889885401096166e-05,
+      "loss": 1.338,
+      "step": 35500
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 2.7827603388141504e-05,
+      "loss": 1.3325,
+      "step": 35600
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 2.7765321375186848e-05,
+      "loss": 1.3343,
+      "step": 35700
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 2.7703039362232185e-05,
+      "loss": 1.336,
+      "step": 35800
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 2.764075734927753e-05,
+      "loss": 1.3327,
+      "step": 35900
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 2.7578475336322873e-05,
+      "loss": 1.3319,
+      "step": 36000
+    },
+    {
+      "epoch": 2.24,
+      "eval_loss": 1.2188748121261597,
+      "eval_runtime": 3765.9112,
+      "eval_samples_per_second": 848.935,
+      "eval_steps_per_second": 0.111,
+      "step": 36000
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 2.751619332336821e-05,
+      "loss": 1.3288,
+      "step": 36100
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 2.7453911310413555e-05,
+      "loss": 1.3313,
+      "step": 36200
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 2.7391629297458892e-05,
+      "loss": 1.33,
+      "step": 36300
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 2.7329347284504236e-05,
+      "loss": 1.3285,
+      "step": 36400
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 2.726706527154958e-05,
+      "loss": 1.3286,
+      "step": 36500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.7204783258594918e-05,
+      "loss": 1.3282,
+      "step": 36600
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 2.7142501245640262e-05,
+      "loss": 1.3262,
+      "step": 36700
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 2.70802192326856e-05,
+      "loss": 1.3266,
+      "step": 36800
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.7017937219730943e-05,
+      "loss": 1.3257,
+      "step": 36900
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 2.6955655206776288e-05,
+      "loss": 1.3256,
+      "step": 37000
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 2.6893373193821625e-05,
+      "loss": 1.3226,
+      "step": 37100
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.683109118086697e-05,
+      "loss": 1.3211,
+      "step": 37200
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.6768809167912306e-05,
+      "loss": 1.3228,
+      "step": 37300
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 2.670652715495765e-05,
+      "loss": 1.3241,
+      "step": 37400
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 2.6644245142002988e-05,
+      "loss": 1.3201,
+      "step": 37500
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 2.6581963129048332e-05,
+      "loss": 1.3232,
+      "step": 37600
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 2.6519681116093676e-05,
+      "loss": 1.3221,
+      "step": 37700
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 2.6457399103139013e-05,
+      "loss": 1.3229,
+      "step": 37800
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 2.6395117090184357e-05,
+      "loss": 1.3188,
+      "step": 37900
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.6332835077229695e-05,
+      "loss": 1.3175,
+      "step": 38000
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 2.627055306427504e-05,
+      "loss": 1.3183,
+      "step": 38100
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 2.6208271051320383e-05,
+      "loss": 1.3192,
+      "step": 38200
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 2.614598903836572e-05,
+      "loss": 1.3184,
+      "step": 38300
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 2.6083707025411064e-05,
+      "loss": 1.3155,
+      "step": 38400
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.6021425012456402e-05,
+      "loss": 1.3178,
+      "step": 38500
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 2.5959142999501746e-05,
+      "loss": 1.3187,
+      "step": 38600
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 2.589686098654709e-05,
+      "loss": 1.3131,
+      "step": 38700
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 2.5834578973592427e-05,
+      "loss": 1.3139,
+      "step": 38800
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 2.577229696063777e-05,
+      "loss": 1.3152,
+      "step": 38900
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 2.571001494768311e-05,
+      "loss": 1.3124,
+      "step": 39000
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 2.5647732934728453e-05,
+      "loss": 1.3139,
+      "step": 39100
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 2.558545092177379e-05,
+      "loss": 1.3136,
+      "step": 39200
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 2.5523168908819134e-05,
+      "loss": 1.3107,
+      "step": 39300
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 2.546088689586448e-05,
+      "loss": 1.3136,
+      "step": 39400
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 2.5398604882909816e-05,
+      "loss": 1.3118,
+      "step": 39500
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 2.533632286995516e-05,
+      "loss": 1.308,
+      "step": 39600
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 2.5274040857000497e-05,
+      "loss": 1.3139,
+      "step": 39700
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 2.521175884404584e-05,
+      "loss": 1.3063,
+      "step": 39800
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 2.5149476831091185e-05,
+      "loss": 1.3063,
+      "step": 39900
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 2.5087194818136523e-05,
+      "loss": 1.305,
+      "step": 40000
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 2.5024912805181867e-05,
+      "loss": 1.3059,
+      "step": 40100
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 2.4962630792227208e-05,
+      "loss": 1.3081,
+      "step": 40200
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 2.490034877927255e-05,
+      "loss": 1.3041,
+      "step": 40300
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 2.483806676631789e-05,
+      "loss": 1.3044,
+      "step": 40400
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 2.477578475336323e-05,
+      "loss": 1.3076,
+      "step": 40500
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 2.471350274040857e-05,
+      "loss": 1.3018,
+      "step": 40600
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 2.4651220727453915e-05,
+      "loss": 1.3061,
+      "step": 40700
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 2.4588938714499255e-05,
+      "loss": 1.307,
+      "step": 40800
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 2.4526656701544596e-05,
+      "loss": 1.3049,
+      "step": 40900
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 2.4464374688589937e-05,
+      "loss": 1.3042,
+      "step": 41000
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 2.4402092675635278e-05,
+      "loss": 1.2987,
+      "step": 41100
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 2.4339810662680618e-05,
+      "loss": 1.3039,
+      "step": 41200
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 2.4277528649725962e-05,
+      "loss": 1.3028,
+      "step": 41300
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 2.4215246636771303e-05,
+      "loss": 1.2998,
+      "step": 41400
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 2.4152964623816644e-05,
+      "loss": 1.3006,
+      "step": 41500
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 2.4090682610861985e-05,
+      "loss": 1.3027,
+      "step": 41600
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 2.4028400597907325e-05,
+      "loss": 1.3016,
+      "step": 41700
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 2.396611858495267e-05,
+      "loss": 1.3001,
+      "step": 41800
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 2.390383657199801e-05,
+      "loss": 1.2998,
+      "step": 41900
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 2.384155455904335e-05,
+      "loss": 1.3011,
+      "step": 42000
+    },
+    {
+      "epoch": 2.62,
+      "eval_loss": 1.189267635345459,
+      "eval_runtime": 3765.6956,
+      "eval_samples_per_second": 848.984,
+      "eval_steps_per_second": 0.111,
+      "step": 42000
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 2.377927254608869e-05,
+      "loss": 1.2986,
+      "step": 42100
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 2.3716990533134032e-05,
+      "loss": 1.2972,
+      "step": 42200
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 2.3654708520179373e-05,
+      "loss": 1.2977,
+      "step": 42300
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 2.3592426507224717e-05,
+      "loss": 1.2945,
+      "step": 42400
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 2.3530144494270058e-05,
+      "loss": 1.2955,
+      "step": 42500
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 2.34678624813154e-05,
+      "loss": 1.2931,
+      "step": 42600
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 2.340558046836074e-05,
+      "loss": 1.2941,
+      "step": 42700
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 2.334329845540608e-05,
+      "loss": 1.2977,
+      "step": 42800
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 2.328101644245142e-05,
+      "loss": 1.2974,
+      "step": 42900
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 2.3218734429496765e-05,
+      "loss": 1.2967,
+      "step": 43000
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 2.3156452416542106e-05,
+      "loss": 1.2919,
+      "step": 43100
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 2.3094170403587446e-05,
+      "loss": 1.2928,
+      "step": 43200
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 2.3031888390632787e-05,
+      "loss": 1.293,
+      "step": 43300
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 2.2969606377678128e-05,
+      "loss": 1.2947,
+      "step": 43400
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 2.2907324364723472e-05,
+      "loss": 1.2926,
+      "step": 43500
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 2.2845042351768813e-05,
+      "loss": 1.2892,
+      "step": 43600
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 2.2782760338814153e-05,
+      "loss": 1.2944,
+      "step": 43700
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 2.2720478325859494e-05,
+      "loss": 1.2907,
+      "step": 43800
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 2.2658196312904835e-05,
+      "loss": 1.2888,
+      "step": 43900
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 2.2595914299950176e-05,
+      "loss": 1.2882,
+      "step": 44000
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 2.253363228699552e-05,
+      "loss": 1.2868,
+      "step": 44100
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 2.247135027404086e-05,
+      "loss": 1.2885,
+      "step": 44200
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 2.24090682610862e-05,
+      "loss": 1.2885,
+      "step": 44300
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 2.2346786248131542e-05,
+      "loss": 1.2855,
+      "step": 44400
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 2.2284504235176883e-05,
+      "loss": 1.2869,
+      "step": 44500
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 1.2854,
+      "step": 44600
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 2.2159940209267564e-05,
+      "loss": 1.2869,
+      "step": 44700
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 2.2097658196312908e-05,
+      "loss": 1.2861,
+      "step": 44800
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.203537618335825e-05,
+      "loss": 1.2858,
+      "step": 44900
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.197309417040359e-05,
+      "loss": 1.2863,
+      "step": 45000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 2.191081215744893e-05,
+      "loss": 1.2839,
+      "step": 45100
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 2.184853014449427e-05,
+      "loss": 1.2812,
+      "step": 45200
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 2.1786248131539612e-05,
+      "loss": 1.2837,
+      "step": 45300
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 2.1723966118584952e-05,
+      "loss": 1.2828,
+      "step": 45400
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 2.1661684105630297e-05,
+      "loss": 1.281,
+      "step": 45500
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 2.1599402092675637e-05,
+      "loss": 1.2833,
+      "step": 45600
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 2.1537120079720978e-05,
+      "loss": 1.2823,
+      "step": 45700
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 2.147483806676632e-05,
+      "loss": 1.283,
+      "step": 45800
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 2.141255605381166e-05,
+      "loss": 1.2774,
+      "step": 45900
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 2.1350274040857e-05,
+      "loss": 1.2794,
+      "step": 46000
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 2.128799202790234e-05,
+      "loss": 1.2784,
+      "step": 46100
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 2.1225710014947685e-05,
+      "loss": 1.2797,
+      "step": 46200
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 2.1163428001993026e-05,
+      "loss": 1.2771,
+      "step": 46300
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 2.1101145989038367e-05,
+      "loss": 1.2746,
+      "step": 46400
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 2.1038863976083707e-05,
+      "loss": 1.2825,
+      "step": 46500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 2.0976581963129048e-05,
+      "loss": 1.2783,
+      "step": 46600
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 2.091429995017439e-05,
+      "loss": 1.2773,
+      "step": 46700
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 2.085201793721973e-05,
+      "loss": 1.2773,
+      "step": 46800
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 2.0789735924265074e-05,
+      "loss": 1.2795,
+      "step": 46900
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 2.0727453911310414e-05,
+      "loss": 1.2756,
+      "step": 47000
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 2.0665171898355755e-05,
+      "loss": 1.2732,
+      "step": 47100
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 2.0602889885401096e-05,
+      "loss": 1.2799,
+      "step": 47200
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 2.0540607872446436e-05,
+      "loss": 1.2752,
+      "step": 47300
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 2.0478325859491777e-05,
+      "loss": 1.2769,
+      "step": 47400
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 2.041604384653712e-05,
+      "loss": 1.2744,
+      "step": 47500
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 2.0353761833582462e-05,
+      "loss": 1.2759,
+      "step": 47600
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 2.0291479820627803e-05,
+      "loss": 1.2773,
+      "step": 47700
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 2.0229197807673143e-05,
+      "loss": 1.2718,
+      "step": 47800
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 2.0166915794718484e-05,
+      "loss": 1.2753,
+      "step": 47900
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 2.0104633781763825e-05,
+      "loss": 1.2737,
+      "step": 48000
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 1.1658238172531128,
+      "eval_runtime": 3765.1156,
+      "eval_samples_per_second": 849.115,
+      "eval_steps_per_second": 0.111,
+      "step": 48000
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 2.004235176880917e-05,
+      "loss": 1.2731,
+      "step": 48100
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 1.998006975585451e-05,
+      "loss": 1.2779,
+      "step": 48200
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 1.991778774289985e-05,
+      "loss": 1.2682,
+      "step": 48300
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 1.985550572994519e-05,
+      "loss": 1.2672,
+      "step": 48400
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 1.9793223716990532e-05,
+      "loss": 1.2717,
+      "step": 48500
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 1.9730941704035873e-05,
+      "loss": 1.2675,
+      "step": 48600
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 1.9668659691081217e-05,
+      "loss": 1.2708,
+      "step": 48700
+    },
+    {
+      "epoch": 3.04,
+      "learning_rate": 1.9606377678126557e-05,
+      "loss": 1.2687,
+      "step": 48800
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 1.9544095665171898e-05,
+      "loss": 1.2709,
+      "step": 48900
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 1.948181365221724e-05,
+      "loss": 1.2707,
+      "step": 49000
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 1.941953163926258e-05,
+      "loss": 1.2693,
+      "step": 49100
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 1.9357249626307924e-05,
+      "loss": 1.2653,
+      "step": 49200
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 1.9294967613353264e-05,
+      "loss": 1.267,
+      "step": 49300
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 1.9232685600398605e-05,
+      "loss": 1.2651,
+      "step": 49400
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 1.9170403587443946e-05,
+      "loss": 1.2661,
+      "step": 49500
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 1.9108121574489287e-05,
+      "loss": 1.2656,
+      "step": 49600
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 1.9045839561534627e-05,
+      "loss": 1.2665,
+      "step": 49700
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 1.898355754857997e-05,
+      "loss": 1.2666,
+      "step": 49800
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 1.8921275535625312e-05,
+      "loss": 1.267,
+      "step": 49900
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 1.8858993522670653e-05,
+      "loss": 1.2609,
+      "step": 50000
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 1.8796711509715994e-05,
+      "loss": 1.2659,
+      "step": 50100
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 1.8734429496761334e-05,
+      "loss": 1.2663,
+      "step": 50200
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 1.8672147483806675e-05,
+      "loss": 1.2657,
+      "step": 50300
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 1.860986547085202e-05,
+      "loss": 1.2639,
+      "step": 50400
+    },
+    {
+      "epoch": 3.15,
+      "learning_rate": 1.854758345789736e-05,
+      "loss": 1.2662,
+      "step": 50500
+    },
+    {
+      "epoch": 3.15,
+      "learning_rate": 1.84853014449427e-05,
+      "loss": 1.2608,
+      "step": 50600
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 1.842301943198804e-05,
+      "loss": 1.2623,
+      "step": 50700
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 1.8360737419033382e-05,
+      "loss": 1.2631,
+      "step": 50800
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 1.8298455406078726e-05,
+      "loss": 1.2613,
+      "step": 50900
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 1.8236173393124067e-05,
+      "loss": 1.2637,
+      "step": 51000
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 1.8173891380169408e-05,
+      "loss": 1.2591,
+      "step": 51100
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 1.811160936721475e-05,
+      "loss": 1.2594,
+      "step": 51200
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 1.804932735426009e-05,
+      "loss": 1.2596,
+      "step": 51300
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 1.798704534130543e-05,
+      "loss": 1.2633,
+      "step": 51400
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 1.7924763328350774e-05,
+      "loss": 1.2586,
+      "step": 51500
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 1.7862481315396115e-05,
+      "loss": 1.2593,
+      "step": 51600
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 1.7800199302441455e-05,
+      "loss": 1.2585,
+      "step": 51700
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 1.7737917289486796e-05,
+      "loss": 1.2578,
+      "step": 51800
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 1.7675635276532137e-05,
+      "loss": 1.2583,
+      "step": 51900
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 1.7613353263577478e-05,
+      "loss": 1.2601,
+      "step": 52000
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 1.7551071250622822e-05,
+      "loss": 1.2603,
+      "step": 52100
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 1.7488789237668162e-05,
+      "loss": 1.2571,
+      "step": 52200
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 1.7426507224713503e-05,
+      "loss": 1.2565,
+      "step": 52300
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 1.7364225211758844e-05,
+      "loss": 1.2592,
+      "step": 52400
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 1.7301943198804185e-05,
+      "loss": 1.2577,
+      "step": 52500
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 1.723966118584953e-05,
+      "loss": 1.2554,
+      "step": 52600
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 1.717737917289487e-05,
+      "loss": 1.2581,
+      "step": 52700
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 1.711509715994021e-05,
+      "loss": 1.2549,
+      "step": 52800
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 1.705281514698555e-05,
+      "loss": 1.2565,
+      "step": 52900
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 1.699053313403089e-05,
+      "loss": 1.2552,
+      "step": 53000
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 1.6928251121076232e-05,
+      "loss": 1.2579,
+      "step": 53100
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 1.6865969108121576e-05,
+      "loss": 1.2526,
+      "step": 53200
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 1.6803687095166917e-05,
+      "loss": 1.2573,
+      "step": 53300
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 1.6741405082212258e-05,
+      "loss": 1.2559,
+      "step": 53400
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 1.66791230692576e-05,
+      "loss": 1.2574,
+      "step": 53500
+    },
+    {
+      "epoch": 3.34,
+      "learning_rate": 1.661684105630294e-05,
+      "loss": 1.2537,
+      "step": 53600
+    },
+    {
+      "epoch": 3.34,
+      "learning_rate": 1.655455904334828e-05,
+      "loss": 1.2549,
+      "step": 53700
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 1.6492277030393624e-05,
+      "loss": 1.2552,
+      "step": 53800
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 1.6429995017438965e-05,
+      "loss": 1.2515,
+      "step": 53900
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 1.6367713004484306e-05,
+      "loss": 1.2524,
+      "step": 54000
+    },
+    {
+      "epoch": 3.36,
+      "eval_loss": 1.1479804515838623,
+      "eval_runtime": 3765.2288,
+      "eval_samples_per_second": 849.089,
+      "eval_steps_per_second": 0.111,
+      "step": 54000
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 1.6305430991529646e-05,
+      "loss": 1.2528,
+      "step": 54100
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 1.6243148978574987e-05,
+      "loss": 1.2518,
+      "step": 54200
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 1.618086696562033e-05,
+      "loss": 1.2482,
+      "step": 54300
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 1.6118584952665672e-05,
+      "loss": 1.2502,
+      "step": 54400
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 1.6056302939711013e-05,
+      "loss": 1.2515,
+      "step": 54500
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 1.5994020926756353e-05,
+      "loss": 1.2514,
+      "step": 54600
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 1.5931738913801694e-05,
+      "loss": 1.251,
+      "step": 54700
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 1.5869456900847035e-05,
+      "loss": 1.2519,
+      "step": 54800
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 1.580717488789238e-05,
+      "loss": 1.2503,
+      "step": 54900
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 1.574489287493772e-05,
+      "loss": 1.252,
+      "step": 55000
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 1.568261086198306e-05,
+      "loss": 1.2499,
+      "step": 55100
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 1.56203288490284e-05,
+      "loss": 1.2458,
+      "step": 55200
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 1.5558046836073742e-05,
+      "loss": 1.2474,
+      "step": 55300
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 1.5495764823119083e-05,
+      "loss": 1.2494,
+      "step": 55400
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 1.5433482810164427e-05,
+      "loss": 1.2502,
+      "step": 55500
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 1.5371200797209767e-05,
+      "loss": 1.2502,
+      "step": 55600
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 1.5308918784255108e-05,
+      "loss": 1.2509,
+      "step": 55700
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 1.5246636771300449e-05,
+      "loss": 1.2491,
+      "step": 55800
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 1.518435475834579e-05,
+      "loss": 1.2499,
+      "step": 55900
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 1.512207274539113e-05,
+      "loss": 1.244,
+      "step": 56000
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 1.5059790732436474e-05,
+      "loss": 1.2428,
+      "step": 56100
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 1.4997508719481815e-05,
+      "loss": 1.2492,
+      "step": 56200
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 1.4935226706527156e-05,
+      "loss": 1.2488,
+      "step": 56300
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 1.4872944693572497e-05,
+      "loss": 1.2476,
+      "step": 56400
+    },
+    {
+      "epoch": 3.52,
+      "learning_rate": 1.4810662680617837e-05,
+      "loss": 1.2468,
+      "step": 56500
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 1.4748380667663181e-05,
+      "loss": 1.2453,
+      "step": 56600
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 1.4686098654708522e-05,
+      "loss": 1.2441,
+      "step": 56700
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 1.4623816641753863e-05,
+      "loss": 1.2474,
+      "step": 56800
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 1.4561534628799204e-05,
+      "loss": 1.2443,
+      "step": 56900
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 1.4499252615844544e-05,
+      "loss": 1.2447,
+      "step": 57000
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 1.4436970602889885e-05,
+      "loss": 1.2432,
+      "step": 57100
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 1.437468858993523e-05,
+      "loss": 1.2434,
+      "step": 57200
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 1.431240657698057e-05,
+      "loss": 1.2447,
+      "step": 57300
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 1.425012456402591e-05,
+      "loss": 1.2442,
+      "step": 57400
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 1.4187842551071251e-05,
+      "loss": 1.2462,
+      "step": 57500
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 1.4125560538116592e-05,
+      "loss": 1.245,
+      "step": 57600
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 1.4063278525161933e-05,
+      "loss": 1.2461,
+      "step": 57700
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 1.4000996512207277e-05,
+      "loss": 1.2462,
+      "step": 57800
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 1.3938714499252618e-05,
+      "loss": 1.2435,
+      "step": 57900
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 1.3876432486297958e-05,
+      "loss": 1.2461,
+      "step": 58000
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 1.3814150473343299e-05,
+      "loss": 1.2398,
+      "step": 58100
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 1.375186846038864e-05,
+      "loss": 1.2402,
+      "step": 58200
+    },
+    {
+      "epoch": 3.63,
+      "learning_rate": 1.3689586447433982e-05,
+      "loss": 1.2427,
+      "step": 58300
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 1.3627304434479323e-05,
+      "loss": 1.2396,
+      "step": 58400
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 1.3565022421524665e-05,
+      "loss": 1.2373,
+      "step": 58500
+    },
+    {
+      "epoch": 3.65,
+      "learning_rate": 1.3502740408570006e-05,
+      "loss": 1.2422,
+      "step": 58600
+    },
+    {
+      "epoch": 3.66,
+      "learning_rate": 1.3440458395615347e-05,
+      "loss": 1.2388,
+      "step": 58700
+    },
+    {
+      "epoch": 3.66,
+      "learning_rate": 1.3378176382660688e-05,
+      "loss": 1.2424,
+      "step": 58800
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 1.331589436970603e-05,
+      "loss": 1.2402,
+      "step": 58900
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 1.325361235675137e-05,
+      "loss": 1.2377,
+      "step": 59000
+    },
+    {
+      "epoch": 3.68,
+      "learning_rate": 1.3191330343796713e-05,
+      "loss": 1.2408,
+      "step": 59100
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 1.3129048330842054e-05,
+      "loss": 1.2405,
+      "step": 59200
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 1.3066766317887395e-05,
+      "loss": 1.2373,
+      "step": 59300
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 1.3004484304932735e-05,
+      "loss": 1.2391,
+      "step": 59400
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 1.2942202291978078e-05,
+      "loss": 1.2409,
+      "step": 59500
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 1.2879920279023418e-05,
+      "loss": 1.2403,
+      "step": 59600
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 1.281763826606876e-05,
+      "loss": 1.2382,
+      "step": 59700
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 1.2755356253114102e-05,
+      "loss": 1.2396,
+      "step": 59800
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 1.2693074240159442e-05,
+      "loss": 1.2376,
+      "step": 59900
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 1.2630792227204785e-05,
+      "loss": 1.2388,
+      "step": 60000
+    },
+    {
+      "epoch": 3.74,
+      "eval_loss": 1.135990023612976,
+      "eval_runtime": 3765.93,
+      "eval_samples_per_second": 848.931,
+      "eval_steps_per_second": 0.111,
+      "step": 60000
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 1.2568510214250125e-05,
+      "loss": 1.2378,
+      "step": 60100
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 1.2506228201295466e-05,
+      "loss": 1.2371,
+      "step": 60200
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 1.2443946188340807e-05,
+      "loss": 1.2379,
+      "step": 60300
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 1.2381664175386148e-05,
+      "loss": 1.2392,
+      "step": 60400
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 1.231938216243149e-05,
+      "loss": 1.2374,
+      "step": 60500
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 1.225710014947683e-05,
+      "loss": 1.2385,
+      "step": 60600
+    },
+    {
+      "epoch": 3.78,
+      "learning_rate": 1.2194818136522173e-05,
+      "loss": 1.2375,
+      "step": 60700
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 1.2132536123567514e-05,
+      "loss": 1.2386,
+      "step": 60800
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 1.2070254110612855e-05,
+      "loss": 1.2368,
+      "step": 60900
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 1.2007972097658197e-05,
+      "loss": 1.2335,
+      "step": 61000
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 1.1945690084703538e-05,
+      "loss": 1.2341,
+      "step": 61100
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 1.1883408071748879e-05,
+      "loss": 1.2366,
+      "step": 61200
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 1.1821126058794221e-05,
+      "loss": 1.2334,
+      "step": 61300
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 1.1758844045839562e-05,
+      "loss": 1.2355,
+      "step": 61400
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 1.1696562032884902e-05,
+      "loss": 1.2353,
+      "step": 61500
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 1.1634280019930245e-05,
+      "loss": 1.2363,
+      "step": 61600
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 1.1571998006975586e-05,
+      "loss": 1.2348,
+      "step": 61700
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 1.1509715994020926e-05,
+      "loss": 1.2334,
+      "step": 61800
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 1.1447433981066269e-05,
+      "loss": 1.2383,
+      "step": 61900
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 1.138515196811161e-05,
+      "loss": 1.2331,
+      "step": 62000
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 1.132286995515695e-05,
+      "loss": 1.2358,
+      "step": 62100
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 1.1260587942202293e-05,
+      "loss": 1.2378,
+      "step": 62200
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 1.1198305929247633e-05,
+      "loss": 1.2354,
+      "step": 62300
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 1.1136023916292976e-05,
+      "loss": 1.2345,
+      "step": 62400
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 1.1073741903338316e-05,
+      "loss": 1.2308,
+      "step": 62500
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 1.1011459890383657e-05,
+      "loss": 1.2326,
+      "step": 62600
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 1.0949177877429e-05,
+      "loss": 1.2352,
+      "step": 62700
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 1.088689586447434e-05,
+      "loss": 1.2328,
+      "step": 62800
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 1.0824613851519681e-05,
+      "loss": 1.2305,
+      "step": 62900
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 1.0762331838565023e-05,
+      "loss": 1.2327,
+      "step": 63000
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 1.0700049825610364e-05,
+      "loss": 1.2314,
+      "step": 63100
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 1.0637767812655705e-05,
+      "loss": 1.2349,
+      "step": 63200
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 1.0575485799701047e-05,
+      "loss": 1.2324,
+      "step": 63300
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 1.0513203786746388e-05,
+      "loss": 1.2324,
+      "step": 63400
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 1.0450921773791729e-05,
+      "loss": 1.2311,
+      "step": 63500
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 1.0388639760837071e-05,
+      "loss": 1.2321,
+      "step": 63600
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 1.0326357747882412e-05,
+      "loss": 1.2327,
+      "step": 63700
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 1.0264075734927753e-05,
+      "loss": 1.2342,
+      "step": 63800
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 1.0201793721973095e-05,
+      "loss": 1.2334,
+      "step": 63900
+    },
+    {
+      "epoch": 3.99,
+      "learning_rate": 1.0139511709018436e-05,
+      "loss": 1.2321,
+      "step": 64000
+    },
+    {
+      "epoch": 3.99,
+      "learning_rate": 1.0077229696063776e-05,
+      "loss": 1.2332,
+      "step": 64100
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 1.0014947683109119e-05,
+      "loss": 1.2331,
+      "step": 64200
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 9.95266567015446e-06,
+      "loss": 1.2375,
+      "step": 64300
+    },
+    {
+      "epoch": 4.01,
+      "learning_rate": 9.890383657199802e-06,
+      "loss": 1.2312,
+      "step": 64400
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 9.828101644245143e-06,
+      "loss": 1.228,
+      "step": 64500
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 9.765819631290484e-06,
+      "loss": 1.229,
+      "step": 64600
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 9.703537618335826e-06,
+      "loss": 1.2275,
+      "step": 64700
+    },
+    {
+      "epoch": 4.04,
+      "learning_rate": 9.641255605381167e-06,
+      "loss": 1.228,
+      "step": 64800
+    },
+    {
+      "epoch": 4.04,
+      "learning_rate": 9.578973592426507e-06,
+      "loss": 1.2293,
+      "step": 64900
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 9.51669157947185e-06,
+      "loss": 1.2284,
+      "step": 65000
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 9.45440956651719e-06,
+      "loss": 1.231,
+      "step": 65100
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 9.392127553562531e-06,
+      "loss": 1.2283,
+      "step": 65200
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 9.329845540607874e-06,
+      "loss": 1.2283,
+      "step": 65300
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 9.267563527653214e-06,
+      "loss": 1.2274,
+      "step": 65400
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 9.205281514698555e-06,
+      "loss": 1.2284,
+      "step": 65500
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 9.142999501743898e-06,
+      "loss": 1.2257,
+      "step": 65600
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 9.080717488789238e-06,
+      "loss": 1.2309,
+      "step": 65700
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 9.018435475834579e-06,
+      "loss": 1.2262,
+      "step": 65800
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 8.956153462879921e-06,
+      "loss": 1.2296,
+      "step": 65900
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 8.893871449925262e-06,
+      "loss": 1.2269,
+      "step": 66000
+    },
+    {
+      "epoch": 4.11,
+      "eval_loss": 1.1256771087646484,
+      "eval_runtime": 3766.7778,
+      "eval_samples_per_second": 848.74,
+      "eval_steps_per_second": 0.111,
+      "step": 66000
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 8.831589436970605e-06,
+      "loss": 1.226,
+      "step": 66100
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 8.769307424015945e-06,
+      "loss": 1.225,
+      "step": 66200
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 8.707025411061286e-06,
+      "loss": 1.2287,
+      "step": 66300
+    },
+    {
+      "epoch": 4.14,
+      "learning_rate": 8.644743398106628e-06,
+      "loss": 1.2269,
+      "step": 66400
+    },
+    {
+      "epoch": 4.14,
+      "learning_rate": 8.582461385151969e-06,
+      "loss": 1.2251,
+      "step": 66500
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 8.52017937219731e-06,
+      "loss": 1.2281,
+      "step": 66600
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 8.457897359242652e-06,
+      "loss": 1.2278,
+      "step": 66700
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 8.395615346287993e-06,
+      "loss": 1.2263,
+      "step": 66800
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 1.2243,
+      "step": 66900
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 8.271051320378676e-06,
+      "loss": 1.2246,
+      "step": 67000
+    },
+    {
+      "epoch": 4.18,
+      "learning_rate": 8.208769307424017e-06,
+      "loss": 1.228,
+      "step": 67100
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 8.146487294469358e-06,
+      "loss": 1.225,
+      "step": 67200
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 8.0842052815147e-06,
+      "loss": 1.2274,
+      "step": 67300
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 8.02192326856004e-06,
+      "loss": 1.2275,
+      "step": 67400
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 7.959641255605381e-06,
+      "loss": 1.2278,
+      "step": 67500
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 7.897359242650724e-06,
+      "loss": 1.2289,
+      "step": 67600
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 7.835077229696065e-06,
+      "loss": 1.2256,
+      "step": 67700
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 7.772795216741405e-06,
+      "loss": 1.223,
+      "step": 67800
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 7.710513203786748e-06,
+      "loss": 1.2265,
+      "step": 67900
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 7.648231190832088e-06,
+      "loss": 1.2261,
+      "step": 68000
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 7.58594917787743e-06,
+      "loss": 1.2246,
+      "step": 68100
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 7.523667164922771e-06,
+      "loss": 1.2243,
+      "step": 68200
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 7.4613851519681115e-06,
+      "loss": 1.2227,
+      "step": 68300
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 7.399103139013454e-06,
+      "loss": 1.2258,
+      "step": 68400
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 7.336821126058795e-06,
+      "loss": 1.2239,
+      "step": 68500
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 7.274539113104135e-06,
+      "loss": 1.2241,
+      "step": 68600
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 7.212257100149478e-06,
+      "loss": 1.2247,
+      "step": 68700
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 7.1499750871948185e-06,
+      "loss": 1.2269,
+      "step": 68800
+    },
+    {
+      "epoch": 4.29,
+      "learning_rate": 7.087693074240159e-06,
+      "loss": 1.2227,
+      "step": 68900
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 7.025411061285502e-06,
+      "loss": 1.2259,
+      "step": 69000
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 6.963129048330842e-06,
+      "loss": 1.2231,
+      "step": 69100
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 6.900847035376183e-06,
+      "loss": 1.2225,
+      "step": 69200
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 6.8385650224215255e-06,
+      "loss": 1.2232,
+      "step": 69300
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 6.776283009466866e-06,
+      "loss": 1.2267,
+      "step": 69400
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 6.714000996512207e-06,
+      "loss": 1.2207,
+      "step": 69500
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 6.651718983557549e-06,
+      "loss": 1.2246,
+      "step": 69600
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 6.58943697060289e-06,
+      "loss": 1.2199,
+      "step": 69700
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 6.5271549576482325e-06,
+      "loss": 1.222,
+      "step": 69800
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 6.464872944693573e-06,
+      "loss": 1.2247,
+      "step": 69900
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 6.402590931738914e-06,
+      "loss": 1.2255,
+      "step": 70000
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 6.3403089187842556e-06,
+      "loss": 1.221,
+      "step": 70100
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 6.278026905829597e-06,
+      "loss": 1.2253,
+      "step": 70200
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 6.215744892874939e-06,
+      "loss": 1.2222,
+      "step": 70300
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 6.1534628799202794e-06,
+      "loss": 1.2211,
+      "step": 70400
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 6.091180866965621e-06,
+      "loss": 1.2221,
+      "step": 70500
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 6.028898854010962e-06,
+      "loss": 1.2206,
+      "step": 70600
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 5.966616841056303e-06,
+      "loss": 1.2213,
+      "step": 70700
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 5.904334828101644e-06,
+      "loss": 1.2236,
+      "step": 70800
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 5.842052815146986e-06,
+      "loss": 1.223,
+      "step": 70900
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 5.779770802192327e-06,
+      "loss": 1.2184,
+      "step": 71000
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 5.717488789237668e-06,
+      "loss": 1.2201,
+      "step": 71100
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 5.6552067762830095e-06,
+      "loss": 1.221,
+      "step": 71200
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 5.592924763328351e-06,
+      "loss": 1.2204,
+      "step": 71300
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 5.530642750373692e-06,
+      "loss": 1.2206,
+      "step": 71400
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 5.468360737419033e-06,
+      "loss": 1.2195,
+      "step": 71500
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 5.406078724464375e-06,
+      "loss": 1.2231,
+      "step": 71600
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 5.343796711509716e-06,
+      "loss": 1.2222,
+      "step": 71700
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 5.281514698555057e-06,
+      "loss": 1.2194,
+      "step": 71800
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 5.219232685600399e-06,
+      "loss": 1.2203,
+      "step": 71900
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 5.15695067264574e-06,
+      "loss": 1.2226,
+      "step": 72000
+    },
+    {
+      "epoch": 4.48,
+      "eval_loss": 1.1193122863769531,
+      "eval_runtime": 3765.4633,
+      "eval_samples_per_second": 849.036,
+      "eval_steps_per_second": 0.111,
+      "step": 72000
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 5.094668659691081e-06,
+      "loss": 1.2224,
+      "step": 72100
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 5.032386646736423e-06,
+      "loss": 1.2179,
+      "step": 72200
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 4.970104633781764e-06,
+      "loss": 1.2197,
+      "step": 72300
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 4.907822620827105e-06,
+      "loss": 1.2199,
+      "step": 72400
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 4.8455406078724465e-06,
+      "loss": 1.2201,
+      "step": 72500
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 4.783258594917788e-06,
+      "loss": 1.2208,
+      "step": 72600
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 4.720976581963129e-06,
+      "loss": 1.2179,
+      "step": 72700
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 4.65869456900847e-06,
+      "loss": 1.2177,
+      "step": 72800
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 4.596412556053812e-06,
+      "loss": 1.22,
+      "step": 72900
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 4.5341305430991535e-06,
+      "loss": 1.2178,
+      "step": 73000
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 4.471848530144494e-06,
+      "loss": 1.2224,
+      "step": 73100
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 4.409566517189836e-06,
+      "loss": 1.2202,
+      "step": 73200
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 4.347284504235177e-06,
+      "loss": 1.2215,
+      "step": 73300
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 4.285002491280518e-06,
+      "loss": 1.2227,
+      "step": 73400
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 4.22272047832586e-06,
+      "loss": 1.2212,
+      "step": 73500
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 4.160438465371201e-06,
+      "loss": 1.2197,
+      "step": 73600
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 4.098156452416543e-06,
+      "loss": 1.2196,
+      "step": 73700
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 4.0358744394618836e-06,
+      "loss": 1.2189,
+      "step": 73800
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 3.973592426507225e-06,
+      "loss": 1.2191,
+      "step": 73900
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 3.911310413552567e-06,
+      "loss": 1.2191,
+      "step": 74000
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.8490284005979074e-06,
+      "loss": 1.221,
+      "step": 74100
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.7867463876432486e-06,
+      "loss": 1.2192,
+      "step": 74200
+    },
+    {
+      "epoch": 4.63,
+      "learning_rate": 3.72446437468859e-06,
+      "loss": 1.2207,
+      "step": 74300
+    },
+    {
+      "epoch": 4.63,
+      "learning_rate": 3.6621823617339313e-06,
+      "loss": 1.2151,
+      "step": 74400
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 3.5999003487792725e-06,
+      "loss": 1.2182,
+      "step": 74500
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.537618335824614e-06,
+      "loss": 1.2195,
+      "step": 74600
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.4753363228699556e-06,
+      "loss": 1.2201,
+      "step": 74700
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 3.4130543099152963e-06,
+      "loss": 1.2188,
+      "step": 74800
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 3.350772296960638e-06,
+      "loss": 1.2234,
+      "step": 74900
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.2884902840059795e-06,
+      "loss": 1.2169,
+      "step": 75000
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 3.22620827105132e-06,
+      "loss": 1.2174,
+      "step": 75100
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 3.1639262580966618e-06,
+      "loss": 1.219,
+      "step": 75200
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.1016442451420033e-06,
+      "loss": 1.2176,
+      "step": 75300
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 3.0393622321873445e-06,
+      "loss": 1.219,
+      "step": 75400
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 2.9770802192326856e-06,
+      "loss": 1.2197,
+      "step": 75500
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 2.9147982062780272e-06,
+      "loss": 1.216,
+      "step": 75600
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 2.8525161933233684e-06,
+      "loss": 1.2199,
+      "step": 75700
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 2.79023418036871e-06,
+      "loss": 1.2165,
+      "step": 75800
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 2.727952167414051e-06,
+      "loss": 1.2172,
+      "step": 75900
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 2.6656701544593922e-06,
+      "loss": 1.2168,
+      "step": 76000
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 2.603388141504734e-06,
+      "loss": 1.2162,
+      "step": 76100
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 2.541106128550075e-06,
+      "loss": 1.2179,
+      "step": 76200
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 2.478824115595416e-06,
+      "loss": 1.2227,
+      "step": 76300
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 2.4165421026407577e-06,
+      "loss": 1.2172,
+      "step": 76400
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 2.354260089686099e-06,
+      "loss": 1.2153,
+      "step": 76500
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 2.29197807673144e-06,
+      "loss": 1.2184,
+      "step": 76600
+    },
+    {
+      "epoch": 4.78,
+      "learning_rate": 2.229696063776781e-06,
+      "loss": 1.2154,
+      "step": 76700
+    },
+    {
+      "epoch": 4.78,
+      "learning_rate": 2.1674140508221227e-06,
+      "loss": 1.2177,
+      "step": 76800
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 2.105132037867464e-06,
+      "loss": 1.2191,
+      "step": 76900
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 2.042850024912805e-06,
+      "loss": 1.2147,
+      "step": 77000
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 1.9805680119581466e-06,
+      "loss": 1.2168,
+      "step": 77100
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 1.9182859990034877e-06,
+      "loss": 1.2164,
+      "step": 77200
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 1.8560039860488293e-06,
+      "loss": 1.2125,
+      "step": 77300
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 1.7937219730941704e-06,
+      "loss": 1.2177,
+      "step": 77400
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 1.7314399601395116e-06,
+      "loss": 1.2161,
+      "step": 77500
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 1.6691579471848532e-06,
+      "loss": 1.2186,
+      "step": 77600
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 1.6068759342301943e-06,
+      "loss": 1.2223,
+      "step": 77700
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 1.5445939212755357e-06,
+      "loss": 1.2181,
+      "step": 77800
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 1.482311908320877e-06,
+      "loss": 1.2183,
+      "step": 77900
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 1.4200298953662184e-06,
+      "loss": 1.215,
+      "step": 78000
+    },
+    {
+      "epoch": 4.86,
+      "eval_loss": 1.116109848022461,
+      "eval_runtime": 3771.6482,
+      "eval_samples_per_second": 847.644,
+      "eval_steps_per_second": 0.111,
+      "step": 78000
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 1.3577478824115595e-06,
+      "loss": 1.2191,
+      "step": 78100
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 1.2954658694569009e-06,
+      "loss": 1.2197,
+      "step": 78200
+    },
+    {
+      "epoch": 4.88,
+      "learning_rate": 1.233183856502242e-06,
+      "loss": 1.2189,
+      "step": 78300
+    },
+    {
+      "epoch": 4.88,
+      "learning_rate": 1.1709018435475834e-06,
+      "loss": 1.2182,
+      "step": 78400
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 1.1086198305929248e-06,
+      "loss": 1.216,
+      "step": 78500
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 1.0463378176382661e-06,
+      "loss": 1.2162,
+      "step": 78600
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 9.840558046836075e-07,
+      "loss": 1.2144,
+      "step": 78700
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 9.217737917289488e-07,
+      "loss": 1.2152,
+      "step": 78800
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 8.5949177877429e-07,
+      "loss": 1.2146,
+      "step": 78900
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 7.972097658196313e-07,
+      "loss": 1.2185,
+      "step": 79000
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 7.349277528649726e-07,
+      "loss": 1.2179,
+      "step": 79100
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 6.72645739910314e-07,
+      "loss": 1.2167,
+      "step": 79200
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 6.103637269556552e-07,
+      "loss": 1.2201,
+      "step": 79300
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 5.480817140009966e-07,
+      "loss": 1.2175,
+      "step": 79400
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 4.857997010463378e-07,
+      "loss": 1.2193,
+      "step": 79500
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 1.90146985550573e-05,
+      "loss": 1.2172,
+      "step": 79600
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 1.897577229696064e-05,
+      "loss": 1.2198,
+      "step": 79700
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 1.8936846038863976e-05,
+      "loss": 1.2203,
+      "step": 79800
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 1.8897919780767316e-05,
+      "loss": 1.2197,
+      "step": 79900
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 1.8858993522670653e-05,
+      "loss": 1.2189,
+      "step": 80000
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 1.8820067264573993e-05,
+      "loss": 1.2181,
+      "step": 80100
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 1.878114100647733e-05,
+      "loss": 1.2196,
+      "step": 80200
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 1.874221474838067e-05,
+      "loss": 1.2217,
+      "step": 80300
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 1.8703288490284006e-05,
+      "loss": 1.2172,
+      "step": 80400
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 1.8664362232187347e-05,
+      "loss": 1.2162,
+      "step": 80500
+    },
+    {
+      "epoch": 5.02,
+      "learning_rate": 1.8625435974090683e-05,
+      "loss": 1.2177,
+      "step": 80600
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 1.8586509715994023e-05,
+      "loss": 1.2178,
+      "step": 80700
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 1.854758345789736e-05,
+      "loss": 1.218,
+      "step": 80800
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 1.85086571998007e-05,
+      "loss": 1.2185,
+      "step": 80900
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 1.8469730941704037e-05,
+      "loss": 1.2157,
+      "step": 81000
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 1.8430804683607373e-05,
+      "loss": 1.2126,
+      "step": 81100
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 1.8391878425510713e-05,
+      "loss": 1.2129,
+      "step": 81200
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 1.835295216741405e-05,
+      "loss": 1.2163,
+      "step": 81300
+    },
+    {
+      "epoch": 5.07,
+      "learning_rate": 1.831402590931739e-05,
+      "loss": 1.2145,
+      "step": 81400
+    },
+    {
+      "epoch": 5.08,
+      "learning_rate": 1.8275099651220727e-05,
+      "loss": 1.2161,
+      "step": 81500
+    },
+    {
+      "epoch": 5.08,
+      "learning_rate": 1.8236173393124067e-05,
+      "loss": 1.2158,
+      "step": 81600
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 1.8197247135027404e-05,
+      "loss": 1.2124,
+      "step": 81700
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 1.8158320876930744e-05,
+      "loss": 1.2137,
+      "step": 81800
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 1.811939461883408e-05,
+      "loss": 1.2161,
+      "step": 81900
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 1.808046836073742e-05,
+      "loss": 1.2128,
+      "step": 82000
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 1.8041542102640757e-05,
+      "loss": 1.2185,
+      "step": 82100
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 1.8002615844544097e-05,
+      "loss": 1.2123,
+      "step": 82200
+    },
+    {
+      "epoch": 5.13,
+      "learning_rate": 1.7963689586447434e-05,
+      "loss": 1.2143,
+      "step": 82300
+    },
+    {
+      "epoch": 5.13,
+      "learning_rate": 1.7924763328350774e-05,
+      "loss": 1.2138,
+      "step": 82400
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 1.788583707025411e-05,
+      "loss": 1.2151,
+      "step": 82500
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 1.784691081215745e-05,
+      "loss": 1.215,
+      "step": 82600
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 1.7807984554060787e-05,
+      "loss": 1.2125,
+      "step": 82700
+    },
+    {
+      "epoch": 5.16,
+      "learning_rate": 1.7769058295964127e-05,
+      "loss": 1.2151,
+      "step": 82800
+    },
+    {
+      "epoch": 5.16,
+      "learning_rate": 1.7730132037867464e-05,
+      "loss": 1.2114,
+      "step": 82900
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 1.7691205779770804e-05,
+      "loss": 1.2088,
+      "step": 83000
+    },
+    {
+      "epoch": 5.18,
+      "learning_rate": 1.765227952167414e-05,
+      "loss": 1.2114,
+      "step": 83100
+    },
+    {
+      "epoch": 5.18,
+      "learning_rate": 1.7613353263577478e-05,
+      "loss": 1.2137,
+      "step": 83200
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 1.7574427005480818e-05,
+      "loss": 1.2117,
+      "step": 83300
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 1.7535500747384158e-05,
+      "loss": 1.2145,
+      "step": 83400
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 1.7496574489287494e-05,
+      "loss": 1.2149,
+      "step": 83500
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 1.745764823119083e-05,
+      "loss": 1.2105,
+      "step": 83600
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 1.741872197309417e-05,
+      "loss": 1.2117,
+      "step": 83700
+    },
+    {
+      "epoch": 5.22,
+      "learning_rate": 1.737979571499751e-05,
+      "loss": 1.2114,
+      "step": 83800
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 1.7340869456900848e-05,
+      "loss": 1.2121,
+      "step": 83900
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 1.7301943198804185e-05,
+      "loss": 1.2107,
+      "step": 84000
+    },
+    {
+      "epoch": 5.23,
+      "eval_loss": 1.1098637580871582,
+      "eval_runtime": 3772.0412,
+      "eval_samples_per_second": 847.556,
+      "eval_steps_per_second": 0.111,
+      "step": 84000
+    },
+    {
+      "epoch": 5.24,
+      "learning_rate": 1.7263016940707525e-05,
+      "loss": 1.2076,
+      "step": 84100
+    },
+    {
+      "epoch": 5.24,
+      "learning_rate": 1.7224090682610865e-05,
+      "loss": 1.2118,
+      "step": 84200
+    },
+    {
+      "epoch": 5.25,
+      "learning_rate": 1.71851644245142e-05,
+      "loss": 1.2069,
+      "step": 84300
+    },
+    {
+      "epoch": 5.26,
+      "learning_rate": 1.7146238166417538e-05,
+      "loss": 1.2099,
+      "step": 84400
+    },
+    {
+      "epoch": 5.26,
+      "learning_rate": 1.7107311908320878e-05,
+      "loss": 1.2106,
+      "step": 84500
+    },
+    {
+      "epoch": 5.27,
+      "learning_rate": 1.7068385650224218e-05,
+      "loss": 1.207,
+      "step": 84600
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 1.702945939212755e-05,
+      "loss": 1.2082,
+      "step": 84700
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 1.699053313403089e-05,
+      "loss": 1.2087,
+      "step": 84800
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 1.6951606875934232e-05,
+      "loss": 1.2076,
+      "step": 84900
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 1.6912680617837572e-05,
+      "loss": 1.2071,
+      "step": 85000
+    },
+    {
+      "epoch": 5.3,
+      "learning_rate": 1.6873754359740905e-05,
+      "loss": 1.2058,
+      "step": 85100
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 1.6834828101644245e-05,
+      "loss": 1.2086,
+      "step": 85200
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 1.6795901843547585e-05,
+      "loss": 1.2044,
+      "step": 85300
+    },
+    {
+      "epoch": 5.32,
+      "learning_rate": 1.6756975585450925e-05,
+      "loss": 1.2075,
+      "step": 85400
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 1.671804932735426e-05,
+      "loss": 1.2076,
+      "step": 85500
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 1.66791230692576e-05,
+      "loss": 1.2092,
+      "step": 85600
+    },
+    {
+      "epoch": 5.34,
+      "learning_rate": 1.664019681116094e-05,
+      "loss": 1.2069,
+      "step": 85700
+    },
+    {
+      "epoch": 5.34,
+      "learning_rate": 1.6601270553064275e-05,
+      "loss": 1.2044,
+      "step": 85800
+    },
+    {
+      "epoch": 5.35,
+      "learning_rate": 1.6562344294967612e-05,
+      "loss": 1.2063,
+      "step": 85900
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 1.6523418036870952e-05,
+      "loss": 1.2023,
+      "step": 86000
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 1.6484491778774292e-05,
+      "loss": 1.2016,
+      "step": 86100
+    },
+    {
+      "epoch": 5.37,
+      "learning_rate": 1.644556552067763e-05,
+      "loss": 1.2022,
+      "step": 86200
+    },
+    {
+      "epoch": 5.37,
+      "learning_rate": 1.6406639262580966e-05,
+      "loss": 1.2043,
+      "step": 86300
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 1.6367713004484306e-05,
+      "loss": 1.2067,
+      "step": 86400
+    },
+    {
+      "epoch": 5.39,
+      "learning_rate": 1.6328786746387646e-05,
+      "loss": 1.2051,
+      "step": 86500
+    },
+    {
+      "epoch": 5.39,
+      "learning_rate": 1.6289860488290982e-05,
+      "loss": 1.2033,
+      "step": 86600
+    },
+    {
+      "epoch": 5.4,
+      "learning_rate": 1.625093423019432e-05,
+      "loss": 1.2025,
+      "step": 86700
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 1.621200797209766e-05,
+      "loss": 1.2046,
+      "step": 86800
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 1.6173081714001e-05,
+      "loss": 1.2057,
+      "step": 86900
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 1.6134155455904336e-05,
+      "loss": 1.2048,
+      "step": 87000
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 1.6095229197807673e-05,
+      "loss": 1.205,
+      "step": 87100
+    },
+    {
+      "epoch": 5.43,
+      "learning_rate": 1.6056302939711013e-05,
+      "loss": 1.205,
+      "step": 87200
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 1.6017376681614353e-05,
+      "loss": 1.2071,
+      "step": 87300
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 1.597845042351769e-05,
+      "loss": 1.2007,
+      "step": 87400
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 1.5939524165421026e-05,
+      "loss": 1.2041,
+      "step": 87500
+    },
+    {
+      "epoch": 5.46,
+      "learning_rate": 1.5900597907324366e-05,
+      "loss": 1.2033,
+      "step": 87600
+    },
+    {
+      "epoch": 5.46,
+      "learning_rate": 1.5861671649227706e-05,
+      "loss": 1.2014,
+      "step": 87700
+    },
+    {
+      "epoch": 5.47,
+      "learning_rate": 1.5822745391131043e-05,
+      "loss": 1.2038,
+      "step": 87800
+    },
+    {
+      "epoch": 5.47,
+      "learning_rate": 1.578381913303438e-05,
+      "loss": 1.2035,
+      "step": 87900
+    },
+    {
+      "epoch": 5.48,
+      "learning_rate": 1.574489287493772e-05,
+      "loss": 1.2017,
+      "step": 88000
+    },
+    {
+      "epoch": 5.49,
+      "learning_rate": 1.5705966616841056e-05,
+      "loss": 1.201,
+      "step": 88100
+    },
+    {
+      "epoch": 5.49,
+      "learning_rate": 1.5667040358744396e-05,
+      "loss": 1.1996,
+      "step": 88200
+    },
+    {
+      "epoch": 5.5,
+      "learning_rate": 1.5628114100647733e-05,
+      "loss": 1.1999,
+      "step": 88300
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 1.5589187842551073e-05,
+      "loss": 1.2003,
+      "step": 88400
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 1.555026158445441e-05,
+      "loss": 1.2,
+      "step": 88500
+    },
+    {
+      "epoch": 5.52,
+      "learning_rate": 1.5511335326357747e-05,
+      "loss": 1.2009,
+      "step": 88600
+    },
+    {
+      "epoch": 5.52,
+      "learning_rate": 1.5472409068261087e-05,
+      "loss": 1.2003,
+      "step": 88700
+    },
+    {
+      "epoch": 5.53,
+      "learning_rate": 1.5433482810164427e-05,
+      "loss": 1.1988,
+      "step": 88800
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 1.5394556552067763e-05,
+      "loss": 1.1993,
+      "step": 88900
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 1.53556302939711e-05,
+      "loss": 1.1983,
+      "step": 89000
+    },
+    {
+      "epoch": 5.55,
+      "learning_rate": 1.531670403587444e-05,
+      "loss": 1.2016,
+      "step": 89100
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 1.527777777777778e-05,
+      "loss": 1.2031,
+      "step": 89200
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 1.5238851519681115e-05,
+      "loss": 1.201,
+      "step": 89300
+    },
+    {
+      "epoch": 5.57,
+      "learning_rate": 1.5199925261584455e-05,
+      "loss": 1.1997,
+      "step": 89400
+    },
+    {
+      "epoch": 5.57,
+      "learning_rate": 1.5160999003487794e-05,
+      "loss": 1.2019,
+      "step": 89500
+    },
+    {
+      "epoch": 5.58,
+      "learning_rate": 1.512207274539113e-05,
+      "loss": 1.1986,
+      "step": 89600
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 1.5083146487294469e-05,
+      "loss": 1.1964,
+      "step": 89700
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 1.5044220229197809e-05,
+      "loss": 1.1958,
+      "step": 89800
+    },
+    {
+      "epoch": 5.6,
+      "learning_rate": 1.5005293971101147e-05,
+      "loss": 1.1984,
+      "step": 89900
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 1.4966367713004484e-05,
+      "loss": 1.1975,
+      "step": 90000
+    },
+    {
+      "epoch": 5.61,
+      "eval_loss": 1.0976917743682861,
+      "eval_runtime": 3764.7544,
+      "eval_samples_per_second": 849.196,
+      "eval_steps_per_second": 0.111,
+      "step": 90000
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 1.4927441454907822e-05,
+      "loss": 1.1948,
+      "step": 90100
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 1.4888515196811162e-05,
+      "loss": 1.1941,
+      "step": 90200
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 1.48495889387145e-05,
+      "loss": 1.1972,
+      "step": 90300
+    },
+    {
+      "epoch": 5.63,
+      "learning_rate": 1.4810662680617837e-05,
+      "loss": 1.1975,
+      "step": 90400
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 1.4771736422521176e-05,
+      "loss": 1.1966,
+      "step": 90500
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 1.4732810164424516e-05,
+      "loss": 1.1958,
+      "step": 90600
+    },
+    {
+      "epoch": 5.65,
+      "learning_rate": 1.4693883906327854e-05,
+      "loss": 1.1966,
+      "step": 90700
+    },
+    {
+      "epoch": 5.66,
+      "learning_rate": 1.465495764823119e-05,
+      "loss": 1.1971,
+      "step": 90800
+    },
+    {
+      "epoch": 5.66,
+      "learning_rate": 1.461603139013453e-05,
+      "loss": 1.1962,
+      "step": 90900
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 1.457710513203787e-05,
+      "loss": 1.1951,
+      "step": 91000
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 1.4538178873941208e-05,
+      "loss": 1.1944,
+      "step": 91100
+    },
+    {
+      "epoch": 5.68,
+      "learning_rate": 1.4499252615844544e-05,
+      "loss": 1.1968,
+      "step": 91200
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 1.4460326357747883e-05,
+      "loss": 1.1943,
+      "step": 91300
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 1.4421400099651223e-05,
+      "loss": 1.1968,
+      "step": 91400
+    },
+    {
+      "epoch": 5.7,
+      "learning_rate": 1.4382473841554558e-05,
+      "loss": 1.1949,
+      "step": 91500
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 1.4343547583457898e-05,
+      "loss": 1.196,
+      "step": 91600
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 1.4304621325361236e-05,
+      "loss": 1.1967,
+      "step": 91700
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 1.4265695067264576e-05,
+      "loss": 1.1923,
+      "step": 91800
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 1.4226768809167911e-05,
+      "loss": 1.1952,
+      "step": 91900
+    },
+    {
+      "epoch": 5.73,
+      "learning_rate": 1.4187842551071251e-05,
+      "loss": 1.1949,
+      "step": 92000
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 1.414891629297459e-05,
+      "loss": 1.1939,
+      "step": 92100
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 1.410999003487793e-05,
+      "loss": 1.1905,
+      "step": 92200
+    },
+    {
+      "epoch": 5.75,
+      "learning_rate": 1.4071063776781265e-05,
+      "loss": 1.1931,
+      "step": 92300
+    },
+    {
+      "epoch": 5.75,
+      "learning_rate": 1.4032137518684605e-05,
+      "loss": 1.192,
+      "step": 92400
+    },
+    {
+      "epoch": 5.76,
+      "learning_rate": 1.3993211260587943e-05,
+      "loss": 1.1925,
+      "step": 92500
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 1.3954285002491282e-05,
+      "loss": 1.194,
+      "step": 92600
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 1.3915358744394618e-05,
+      "loss": 1.19,
+      "step": 92700
+    },
+    {
+      "epoch": 5.78,
+      "learning_rate": 1.3876432486297958e-05,
+      "loss": 1.1951,
+      "step": 92800
+    },
+    {
+      "epoch": 5.79,
+      "learning_rate": 1.3837506228201297e-05,
+      "loss": 1.1893,
+      "step": 92900
+    },
+    {
+      "epoch": 5.79,
+      "learning_rate": 1.3798579970104633e-05,
+      "loss": 1.1939,
+      "step": 93000
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 1.3759653712007972e-05,
+      "loss": 1.1904,
+      "step": 93100
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 1.3720727453911312e-05,
+      "loss": 1.1902,
+      "step": 93200
+    },
+    {
+      "epoch": 5.81,
+      "learning_rate": 1.368180119581465e-05,
+      "loss": 1.191,
+      "step": 93300
+    },
+    {
+      "epoch": 5.82,
+      "learning_rate": 1.3642874937717987e-05,
+      "loss": 1.1911,
+      "step": 93400
+    },
+    {
+      "epoch": 5.82,
+      "learning_rate": 1.3603948679621325e-05,
+      "loss": 1.1933,
+      "step": 93500
+    },
+    {
+      "epoch": 5.83,
+      "learning_rate": 1.3565022421524665e-05,
+      "loss": 1.1905,
+      "step": 93600
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 1.3526096163428004e-05,
+      "loss": 1.1917,
+      "step": 93700
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 1.348716990533134e-05,
+      "loss": 1.1926,
+      "step": 93800
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 1.3448243647234679e-05,
+      "loss": 1.1913,
+      "step": 93900
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 1.3409317389138017e-05,
+      "loss": 1.1902,
+      "step": 94000
+    },
+    {
+      "epoch": 5.86,
+      "learning_rate": 1.3370391131041357e-05,
+      "loss": 1.1858,
+      "step": 94100
+    },
+    {
+      "epoch": 5.87,
+      "learning_rate": 1.3331464872944694e-05,
+      "loss": 1.1908,
+      "step": 94200
+    },
+    {
+      "epoch": 5.87,
+      "learning_rate": 1.3292538614848032e-05,
+      "loss": 1.1901,
+      "step": 94300
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 1.325361235675137e-05,
+      "loss": 1.1905,
+      "step": 94400
+    },
+    {
+      "epoch": 5.89,
+      "learning_rate": 1.321468609865471e-05,
+      "loss": 1.19,
+      "step": 94500
+    },
+    {
+      "epoch": 5.89,
+      "learning_rate": 1.3175759840558047e-05,
+      "loss": 1.1905,
+      "step": 94600
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 1.3136833582461386e-05,
+      "loss": 1.1913,
+      "step": 94700
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 1.3097907324364724e-05,
+      "loss": 1.1904,
+      "step": 94800
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 1.3058981066268061e-05,
+      "loss": 1.1876,
+      "step": 94900
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 1.3020054808171401e-05,
+      "loss": 1.1909,
+      "step": 95000
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 1.298112855007474e-05,
+      "loss": 1.1894,
+      "step": 95100
+    },
+    {
+      "epoch": 5.93,
+      "learning_rate": 1.2942202291978078e-05,
+      "loss": 1.1899,
+      "step": 95200
+    },
+    {
+      "epoch": 5.94,
+      "learning_rate": 1.2903276033881414e-05,
+      "loss": 1.1898,
+      "step": 95300
+    },
+    {
+      "epoch": 5.94,
+      "learning_rate": 1.2864349775784754e-05,
+      "loss": 1.1879,
+      "step": 95400
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 1.2825423517688093e-05,
+      "loss": 1.1888,
+      "step": 95500
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 1.2786497259591431e-05,
+      "loss": 1.1926,
+      "step": 95600
+    },
+    {
+      "epoch": 5.96,
+      "learning_rate": 1.2747571001494768e-05,
+      "loss": 1.1878,
+      "step": 95700
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 1.2708644743398106e-05,
+      "loss": 1.1874,
+      "step": 95800
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 1.2669718485301446e-05,
+      "loss": 1.1888,
+      "step": 95900
+    },
+    {
+      "epoch": 5.98,
+      "learning_rate": 1.2630792227204785e-05,
+      "loss": 1.1865,
+      "step": 96000
+    },
+    {
+      "epoch": 5.98,
+      "eval_loss": 1.088141918182373,
+      "eval_runtime": 3765.4987,
+      "eval_samples_per_second": 849.028,
+      "eval_steps_per_second": 0.111,
+      "step": 96000
+    },
+    {
+      "epoch": 5.99,
+      "learning_rate": 1.2591865969108121e-05,
+      "loss": 1.1876,
+      "step": 96100
+    },
+    {
+      "epoch": 5.99,
+      "learning_rate": 1.255293971101146e-05,
+      "loss": 1.1875,
+      "step": 96200
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 1.25140134529148e-05,
+      "loss": 1.1865,
+      "step": 96300
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 1.2475087194818137e-05,
+      "loss": 1.1897,
+      "step": 96400
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 1.2436160936721475e-05,
+      "loss": 1.1864,
+      "step": 96500
+    },
+    {
+      "epoch": 6.02,
+      "learning_rate": 1.2397234678624813e-05,
+      "loss": 1.1863,
+      "step": 96600
+    },
+    {
+      "epoch": 6.02,
+      "learning_rate": 1.2358308420528152e-05,
+      "loss": 1.1852,
+      "step": 96700
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 1.231938216243149e-05,
+      "loss": 1.1857,
+      "step": 96800
+    },
+    {
+      "epoch": 6.04,
+      "learning_rate": 1.2280455904334828e-05,
+      "loss": 1.1899,
+      "step": 96900
+    },
+    {
+      "epoch": 6.04,
+      "learning_rate": 1.2241529646238167e-05,
+      "loss": 1.1851,
+      "step": 97000
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 1.2202603388141505e-05,
+      "loss": 1.1846,
+      "step": 97100
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 1.2163677130044844e-05,
+      "loss": 1.1887,
+      "step": 97200
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 1.2124750871948182e-05,
+      "loss": 1.1853,
+      "step": 97300
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 1.208582461385152e-05,
+      "loss": 1.1863,
+      "step": 97400
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 1.2046898355754859e-05,
+      "loss": 1.1871,
+      "step": 97500
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 1.2007972097658197e-05,
+      "loss": 1.1867,
+      "step": 97600
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 1.1969045839561535e-05,
+      "loss": 1.1865,
+      "step": 97700
+    },
+    {
+      "epoch": 6.09,
+      "learning_rate": 1.1930119581464874e-05,
+      "loss": 1.1821,
+      "step": 97800
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 1.1891193323368212e-05,
+      "loss": 1.1866,
+      "step": 97900
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 1.1852267065271549e-05,
+      "loss": 1.1868,
+      "step": 98000
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 1.1813340807174889e-05,
+      "loss": 1.1832,
+      "step": 98100
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 1.1774414549078226e-05,
+      "loss": 1.1829,
+      "step": 98200
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 1.1735488290981566e-05,
+      "loss": 1.1857,
+      "step": 98300
+    },
+    {
+      "epoch": 6.13,
+      "learning_rate": 1.1696562032884902e-05,
+      "loss": 1.1836,
+      "step": 98400
+    },
+    {
+      "epoch": 6.13,
+      "learning_rate": 1.1657635774788242e-05,
+      "loss": 1.1825,
+      "step": 98500
+    },
+    {
+      "epoch": 6.14,
+      "learning_rate": 1.1618709516691579e-05,
+      "loss": 1.1847,
+      "step": 98600
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 1.157978325859492e-05,
+      "loss": 1.1832,
+      "step": 98700
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 1.1540857000498256e-05,
+      "loss": 1.1861,
+      "step": 98800
+    },
+    {
+      "epoch": 6.16,
+      "learning_rate": 1.1501930742401596e-05,
+      "loss": 1.1837,
+      "step": 98900
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 1.1463004484304933e-05,
+      "loss": 1.1821,
+      "step": 99000
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 1.1424078226208273e-05,
+      "loss": 1.1804,
+      "step": 99100
+    },
+    {
+      "epoch": 6.18,
+      "learning_rate": 1.138515196811161e-05,
+      "loss": 1.1828,
+      "step": 99200
+    },
+    {
+      "epoch": 6.18,
+      "learning_rate": 1.134622571001495e-05,
+      "loss": 1.1824,
+      "step": 99300
+    },
+    {
+      "epoch": 6.19,
+      "learning_rate": 1.1307299451918286e-05,
+      "loss": 1.1831,
+      "step": 99400
+    },
+    {
+      "epoch": 6.2,
+      "learning_rate": 1.1268373193821626e-05,
+      "loss": 1.1832,
+      "step": 99500
+    },
+    {
+      "epoch": 6.2,
+      "learning_rate": 1.1229446935724963e-05,
+      "loss": 1.1833,
+      "step": 99600
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 1.1190520677628301e-05,
+      "loss": 1.1842,
+      "step": 99700
+    },
+    {
+      "epoch": 6.22,
+      "learning_rate": 1.115159441953164e-05,
+      "loss": 1.1842,
+      "step": 99800
+    },
+    {
+      "epoch": 6.22,
+      "learning_rate": 1.1112668161434978e-05,
+      "loss": 1.1823,
+      "step": 99900
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 1.1073741903338316e-05,
+      "loss": 1.1813,
+      "step": 100000
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 1.1034815645241655e-05,
+      "loss": 1.1827,
+      "step": 100100
+    },
+    {
+      "epoch": 6.24,
+      "learning_rate": 1.0995889387144993e-05,
+      "loss": 1.181,
+      "step": 100200
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 1.0956963129048332e-05,
+      "loss": 1.1834,
+      "step": 100300
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 1.091803687095167e-05,
+      "loss": 1.183,
+      "step": 100400
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 1.0879110612855008e-05,
+      "loss": 1.1766,
+      "step": 100500
+    },
+    {
+      "epoch": 6.27,
+      "learning_rate": 1.0840184354758347e-05,
+      "loss": 1.1772,
+      "step": 100600
+    },
+    {
+      "epoch": 6.27,
+      "learning_rate": 1.0801258096661685e-05,
+      "loss": 1.1813,
+      "step": 100700
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 1.0762331838565023e-05,
+      "loss": 1.1808,
+      "step": 100800
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 1.0723405580468362e-05,
+      "loss": 1.1793,
+      "step": 100900
+    },
+    {
+      "epoch": 6.29,
+      "learning_rate": 1.06844793223717e-05,
+      "loss": 1.1791,
+      "step": 101000
+    },
+    {
+      "epoch": 6.3,
+      "learning_rate": 1.0645553064275039e-05,
+      "loss": 1.1807,
+      "step": 101100
+    },
+    {
+      "epoch": 6.3,
+      "learning_rate": 1.0606626806178375e-05,
+      "loss": 1.1769,
+      "step": 101200
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 1.0567700548081715e-05,
+      "loss": 1.1802,
+      "step": 101300
+    },
+    {
+      "epoch": 6.32,
+      "learning_rate": 1.0528774289985052e-05,
+      "loss": 1.1784,
+      "step": 101400
+    },
+    {
+      "epoch": 6.32,
+      "learning_rate": 1.0489848031888392e-05,
+      "loss": 1.1782,
+      "step": 101500
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 1.0450921773791729e-05,
+      "loss": 1.1773,
+      "step": 101600
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 1.0411995515695069e-05,
+      "loss": 1.1783,
+      "step": 101700
+    },
+    {
+      "epoch": 6.34,
+      "learning_rate": 1.0373069257598406e-05,
+      "loss": 1.179,
+      "step": 101800
+    },
+    {
+      "epoch": 6.35,
+      "learning_rate": 1.0334142999501744e-05,
+      "loss": 1.1765,
+      "step": 101900
+    },
+    {
+      "epoch": 6.35,
+      "learning_rate": 1.0295216741405082e-05,
+      "loss": 1.1764,
+      "step": 102000
+    },
+    {
+      "epoch": 6.35,
+      "eval_loss": 1.0806279182434082,
+      "eval_runtime": 3770.9356,
+      "eval_samples_per_second": 847.804,
+      "eval_steps_per_second": 0.111,
+      "step": 102000
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 1.025629048330842e-05,
+      "loss": 1.177,
+      "step": 102100
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 1.0217364225211759e-05,
+      "loss": 1.1794,
+      "step": 102200
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 1.0178437967115097e-05,
+      "loss": 1.1783,
+      "step": 102300
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 1.0139511709018436e-05,
+      "loss": 1.1807,
+      "step": 102400
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 1.0100585450921774e-05,
+      "loss": 1.1807,
+      "step": 102500
+    },
+    {
+      "epoch": 6.39,
+      "learning_rate": 1.0061659192825113e-05,
+      "loss": 1.1769,
+      "step": 102600
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 1.0022732934728451e-05,
+      "loss": 1.1777,
+      "step": 102700
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 9.98380667663179e-06,
+      "loss": 1.1766,
+      "step": 102800
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 9.944880418535128e-06,
+      "loss": 1.1775,
+      "step": 102900
+    },
+    {
+      "epoch": 6.42,
+      "learning_rate": 9.905954160438466e-06,
+      "loss": 1.1743,
+      "step": 103000
+    },
+    {
+      "epoch": 6.42,
+      "learning_rate": 9.867027902341804e-06,
+      "loss": 1.1771,
+      "step": 103100
+    },
+    {
+      "epoch": 6.43,
+      "learning_rate": 9.828101644245143e-06,
+      "loss": 1.1765,
+      "step": 103200
+    },
+    {
+      "epoch": 6.43,
+      "learning_rate": 9.78917538614848e-06,
+      "loss": 1.1774,
+      "step": 103300
+    },
+    {
+      "epoch": 6.44,
+      "learning_rate": 9.75024912805182e-06,
+      "loss": 1.177,
+      "step": 103400
+    },
+    {
+      "epoch": 6.45,
+      "learning_rate": 9.711322869955156e-06,
+      "loss": 1.1752,
+      "step": 103500
+    },
+    {
+      "epoch": 6.45,
+      "learning_rate": 9.672396611858496e-06,
+      "loss": 1.1776,
+      "step": 103600
+    },
+    {
+      "epoch": 6.46,
+      "learning_rate": 9.633470353761833e-06,
+      "loss": 1.1792,
+      "step": 103700
+    },
+    {
+      "epoch": 6.46,
+      "learning_rate": 9.594544095665173e-06,
+      "loss": 1.1748,
+      "step": 103800
+    },
+    {
+      "epoch": 6.47,
+      "learning_rate": 9.55561783756851e-06,
+      "loss": 1.1761,
+      "step": 103900
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 9.51669157947185e-06,
+      "loss": 1.1766,
+      "step": 104000
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 9.477765321375186e-06,
+      "loss": 1.1788,
+      "step": 104100
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 9.438839063278527e-06,
+      "loss": 1.1755,
+      "step": 104200
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 9.399912805181863e-06,
+      "loss": 1.1764,
+      "step": 104300
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 9.360986547085203e-06,
+      "loss": 1.1726,
+      "step": 104400
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 9.32206028898854e-06,
+      "loss": 1.1769,
+      "step": 104500
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 9.283134030891878e-06,
+      "loss": 1.1775,
+      "step": 104600
+    },
+    {
+      "epoch": 6.52,
+      "learning_rate": 9.244207772795217e-06,
+      "loss": 1.1732,
+      "step": 104700
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 9.205281514698555e-06,
+      "loss": 1.175,
+      "step": 104800
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 9.166355256601893e-06,
+      "loss": 1.1744,
+      "step": 104900
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 9.127428998505232e-06,
+      "loss": 1.1795,
+      "step": 105000
+    },
+    {
+      "epoch": 6.55,
+      "learning_rate": 9.08850274040857e-06,
+      "loss": 1.1779,
+      "step": 105100
+    },
+    {
+      "epoch": 6.55,
+      "learning_rate": 9.049576482311909e-06,
+      "loss": 1.1736,
+      "step": 105200
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 9.010650224215247e-06,
+      "loss": 1.1744,
+      "step": 105300
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 8.971723966118585e-06,
+      "loss": 1.1723,
+      "step": 105400
+    },
+    {
+      "epoch": 6.57,
+      "learning_rate": 8.932797708021924e-06,
+      "loss": 1.1779,
+      "step": 105500
+    },
+    {
+      "epoch": 6.58,
+      "learning_rate": 8.893871449925262e-06,
+      "loss": 1.1728,
+      "step": 105600
+    },
+    {
+      "epoch": 6.58,
+      "learning_rate": 8.8549451918286e-06,
+      "loss": 1.1724,
+      "step": 105700
+    },
+    {
+      "epoch": 6.59,
+      "learning_rate": 8.816018933731939e-06,
+      "loss": 1.175,
+      "step": 105800
+    },
+    {
+      "epoch": 6.6,
+      "learning_rate": 8.777092675635277e-06,
+      "loss": 1.1758,
+      "step": 105900
+    },
+    {
+      "epoch": 6.6,
+      "learning_rate": 8.738166417538616e-06,
+      "loss": 1.1737,
+      "step": 106000
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 8.699240159441954e-06,
+      "loss": 1.174,
+      "step": 106100
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 8.660313901345292e-06,
+      "loss": 1.1754,
+      "step": 106200
+    },
+    {
+      "epoch": 6.62,
+      "learning_rate": 8.621387643248629e-06,
+      "loss": 1.171,
+      "step": 106300
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 8.582461385151969e-06,
+      "loss": 1.1725,
+      "step": 106400
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 8.543535127055306e-06,
+      "loss": 1.1707,
+      "step": 106500
+    },
+    {
+      "epoch": 6.64,
+      "learning_rate": 8.504608868958646e-06,
+      "loss": 1.1731,
+      "step": 106600
+    },
+    {
+      "epoch": 6.65,
+      "learning_rate": 8.465682610861983e-06,
+      "loss": 1.1725,
+      "step": 106700
+    },
+    {
+      "epoch": 6.65,
+      "learning_rate": 8.426756352765323e-06,
+      "loss": 1.1718,
+      "step": 106800
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 8.38783009466866e-06,
+      "loss": 1.1735,
+      "step": 106900
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 8.348903836572e-06,
+      "loss": 1.1749,
+      "step": 107000
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 8.309977578475336e-06,
+      "loss": 1.1734,
+      "step": 107100
+    },
+    {
+      "epoch": 6.68,
+      "learning_rate": 8.271051320378676e-06,
+      "loss": 1.1726,
+      "step": 107200
+    },
+    {
+      "epoch": 6.68,
+      "learning_rate": 8.232125062282013e-06,
+      "loss": 1.1735,
+      "step": 107300
+    },
+    {
+      "epoch": 6.69,
+      "learning_rate": 8.193198804185353e-06,
+      "loss": 1.1727,
+      "step": 107400
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 8.15427254608869e-06,
+      "loss": 1.1721,
+      "step": 107500
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 8.11534628799203e-06,
+      "loss": 1.172,
+      "step": 107600
+    },
+    {
+      "epoch": 6.71,
+      "learning_rate": 8.076420029895366e-06,
+      "loss": 1.1748,
+      "step": 107700
+    },
+    {
+      "epoch": 6.71,
+      "learning_rate": 8.037493771798706e-06,
+      "loss": 1.1746,
+      "step": 107800
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 7.998567513702043e-06,
+      "loss": 1.1713,
+      "step": 107900
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 7.959641255605381e-06,
+      "loss": 1.1751,
+      "step": 108000
+    },
+    {
+      "epoch": 6.73,
+      "eval_loss": 1.0753653049468994,
+      "eval_runtime": 3765.5362,
+      "eval_samples_per_second": 849.02,
+      "eval_steps_per_second": 0.111,
+      "step": 108000
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 7.92071499750872e-06,
+      "loss": 1.1738,
+      "step": 108100
+    },
+    {
+      "epoch": 6.74,
+      "learning_rate": 7.881788739412058e-06,
+      "loss": 1.1715,
+      "step": 108200
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 7.842862481315397e-06,
+      "loss": 1.1737,
+      "step": 108300
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 7.803936223218735e-06,
+      "loss": 1.1703,
+      "step": 108400
+    },
+    {
+      "epoch": 6.76,
+      "learning_rate": 7.765009965122073e-06,
+      "loss": 1.1739,
+      "step": 108500
+    },
+    {
+      "epoch": 6.76,
+      "learning_rate": 7.726083707025412e-06,
+      "loss": 1.167,
+      "step": 108600
+    },
+    {
+      "epoch": 6.77,
+      "learning_rate": 7.68715744892875e-06,
+      "loss": 1.1735,
+      "step": 108700
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 7.648231190832088e-06,
+      "loss": 1.169,
+      "step": 108800
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 7.609304932735427e-06,
+      "loss": 1.1709,
+      "step": 108900
+    },
+    {
+      "epoch": 6.79,
+      "learning_rate": 7.570378674638764e-06,
+      "loss": 1.1723,
+      "step": 109000
+    },
+    {
+      "epoch": 6.79,
+      "learning_rate": 7.531452416542104e-06,
+      "loss": 1.1733,
+      "step": 109100
+    },
+    {
+      "epoch": 6.8,
+      "learning_rate": 7.492526158445441e-06,
+      "loss": 1.1717,
+      "step": 109200
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 7.45359990034878e-06,
+      "loss": 1.1724,
+      "step": 109300
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 7.414673642252118e-06,
+      "loss": 1.1715,
+      "step": 109400
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 7.3757473841554554e-06,
+      "loss": 1.1696,
+      "step": 109500
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 7.336821126058795e-06,
+      "loss": 1.1719,
+      "step": 109600
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 7.297894867962132e-06,
+      "loss": 1.1712,
+      "step": 109700
+    },
+    {
+      "epoch": 6.84,
+      "learning_rate": 7.258968609865471e-06,
+      "loss": 1.1728,
+      "step": 109800
+    },
+    {
+      "epoch": 6.84,
+      "learning_rate": 7.220042351768809e-06,
+      "loss": 1.1714,
+      "step": 109900
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 7.181116093672148e-06,
+      "loss": 1.1714,
+      "step": 110000
+    },
+    {
+      "epoch": 6.86,
+      "learning_rate": 7.142189835575486e-06,
+      "loss": 1.1723,
+      "step": 110100
+    },
+    {
+      "epoch": 6.86,
+      "learning_rate": 7.103263577478825e-06,
+      "loss": 1.1681,
+      "step": 110200
+    },
+    {
+      "epoch": 6.87,
+      "learning_rate": 7.0643373193821624e-06,
+      "loss": 1.1716,
+      "step": 110300
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 7.025411061285502e-06,
+      "loss": 1.169,
+      "step": 110400
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 6.986484803188839e-06,
+      "loss": 1.168,
+      "step": 110500
+    },
+    {
+      "epoch": 6.89,
+      "learning_rate": 6.947558545092178e-06,
+      "loss": 1.1717,
+      "step": 110600
+    },
+    {
+      "epoch": 6.89,
+      "learning_rate": 6.908632286995516e-06,
+      "loss": 1.172,
+      "step": 110700
+    },
+    {
+      "epoch": 6.9,
+      "learning_rate": 6.869706028898855e-06,
+      "loss": 1.1704,
+      "step": 110800
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 6.830779770802193e-06,
+      "loss": 1.1689,
+      "step": 110900
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 6.791853512705532e-06,
+      "loss": 1.1655,
+      "step": 111000
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 6.7529272546088695e-06,
+      "loss": 1.1639,
+      "step": 111100
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 6.714000996512207e-06,
+      "loss": 1.1693,
+      "step": 111200
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 6.675074738415546e-06,
+      "loss": 1.1669,
+      "step": 111300
+    },
+    {
+      "epoch": 6.94,
+      "learning_rate": 6.636148480318884e-06,
+      "loss": 1.1692,
+      "step": 111400
+    },
+    {
+      "epoch": 6.94,
+      "learning_rate": 6.597222222222223e-06,
+      "loss": 1.1657,
+      "step": 111500
+    },
+    {
+      "epoch": 6.95,
+      "learning_rate": 6.5582959641255605e-06,
+      "loss": 1.1698,
+      "step": 111600
+    },
+    {
+      "epoch": 6.96,
+      "learning_rate": 6.5193697060289e-06,
+      "loss": 1.1721,
+      "step": 111700
+    },
+    {
+      "epoch": 6.96,
+      "learning_rate": 6.480443447932237e-06,
+      "loss": 1.1694,
+      "step": 111800
+    },
+    {
+      "epoch": 6.97,
+      "learning_rate": 6.4415171898355765e-06,
+      "loss": 1.1709,
+      "step": 111900
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 6.402590931738914e-06,
+      "loss": 1.1702,
+      "step": 112000
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 6.363664673642253e-06,
+      "loss": 1.1655,
+      "step": 112100
+    },
+    {
+      "epoch": 6.99,
+      "learning_rate": 6.324738415545591e-06,
+      "loss": 1.1695,
+      "step": 112200
+    },
+    {
+      "epoch": 6.99,
+      "learning_rate": 6.285812157448929e-06,
+      "loss": 1.1689,
+      "step": 112300
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 6.2468858993522675e-06,
+      "loss": 1.1781,
+      "step": 112400
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 6.207959641255606e-06,
+      "loss": 1.1675,
+      "step": 112500
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 6.169033383158944e-06,
+      "loss": 1.1665,
+      "step": 112600
+    },
+    {
+      "epoch": 7.02,
+      "learning_rate": 6.130107125062283e-06,
+      "loss": 1.1679,
+      "step": 112700
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 6.091180866965621e-06,
+      "loss": 1.17,
+      "step": 112800
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 6.052254608868959e-06,
+      "loss": 1.1653,
+      "step": 112900
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 6.013328350772297e-06,
+      "loss": 1.1712,
+      "step": 113000
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 5.974402092675635e-06,
+      "loss": 1.1653,
+      "step": 113100
+    },
+    {
+      "epoch": 7.05,
+      "learning_rate": 5.935475834578974e-06,
+      "loss": 1.1676,
+      "step": 113200
+    },
+    {
+      "epoch": 7.06,
+      "learning_rate": 5.896549576482312e-06,
+      "loss": 1.1679,
+      "step": 113300
+    },
+    {
+      "epoch": 7.06,
+      "learning_rate": 5.8576233183856504e-06,
+      "loss": 1.1671,
+      "step": 113400
+    },
+    {
+      "epoch": 7.07,
+      "learning_rate": 5.818697060288989e-06,
+      "loss": 1.1654,
+      "step": 113500
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 5.779770802192327e-06,
+      "loss": 1.1718,
+      "step": 113600
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 5.7408445440956656e-06,
+      "loss": 1.1703,
+      "step": 113700
+    },
+    {
+      "epoch": 7.09,
+      "learning_rate": 5.701918285999003e-06,
+      "loss": 1.165,
+      "step": 113800
+    },
+    {
+      "epoch": 7.09,
+      "learning_rate": 5.6629920279023415e-06,
+      "loss": 1.167,
+      "step": 113900
+    },
+    {
+      "epoch": 7.1,
+      "learning_rate": 5.62406576980568e-06,
+      "loss": 1.1679,
+      "step": 114000
+    },
+    {
+      "epoch": 7.1,
+      "eval_loss": 1.070519208908081,
+      "eval_runtime": 3765.1139,
+      "eval_samples_per_second": 849.115,
+      "eval_steps_per_second": 0.111,
+      "step": 114000
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 3.308065530208667e-05,
+      "loss": 2.3823,
+      "step": 114100
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 3.306582677912618e-05,
+      "loss": 2.1578,
+      "step": 114200
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 3.30509982561657e-05,
+      "loss": 2.14,
+      "step": 114300
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 3.303616973320522e-05,
+      "loss": 2.1186,
+      "step": 114400
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 3.302134121024473e-05,
+      "loss": 2.1023,
+      "step": 114500
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 3.3006512687284244e-05,
+      "loss": 2.0899,
+      "step": 114600
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 3.299168416432376e-05,
+      "loss": 2.0709,
+      "step": 114700
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 3.2976855641363274e-05,
+      "loss": 2.0573,
+      "step": 114800
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 3.296202711840279e-05,
+      "loss": 2.0477,
+      "step": 114900
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 3.2947198595442305e-05,
+      "loss": 2.0338,
+      "step": 115000
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 3.2932370072481824e-05,
+      "loss": 2.0172,
+      "step": 115100
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 3.2917541549521335e-05,
+      "loss": 2.0062,
+      "step": 115200
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 3.290271302656085e-05,
+      "loss": 1.9953,
+      "step": 115300
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 3.2887884503600366e-05,
+      "loss": 1.9829,
+      "step": 115400
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 3.2873055980639885e-05,
+      "loss": 1.9714,
+      "step": 115500
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 3.2858227457679397e-05,
+      "loss": 1.9585,
+      "step": 115600
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 3.284339893471891e-05,
+      "loss": 1.9509,
+      "step": 115700
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 3.282857041175843e-05,
+      "loss": 1.9385,
+      "step": 115800
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 3.2813741888797946e-05,
+      "loss": 1.932,
+      "step": 115900
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 3.279891336583746e-05,
+      "loss": 1.9274,
+      "step": 116000
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 3.278408484287697e-05,
+      "loss": 1.9128,
+      "step": 116100
+    },
+    {
+      "epoch": 4.14,
+      "learning_rate": 3.276925631991649e-05,
+      "loss": 1.9078,
+      "step": 116200
+    },
+    {
+      "epoch": 4.14,
+      "learning_rate": 3.2754427796956e-05,
+      "loss": 1.8988,
+      "step": 116300
+    },
+    {
+      "epoch": 4.14,
+      "learning_rate": 3.273959927399552e-05,
+      "loss": 1.8916,
+      "step": 116400
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 3.272477075103503e-05,
+      "loss": 1.8821,
+      "step": 116500
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 3.270994222807455e-05,
+      "loss": 1.878,
+      "step": 116600
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 3.269511370511406e-05,
+      "loss": 1.8717,
+      "step": 116700
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 3.268028518215357e-05,
+      "loss": 1.8593,
+      "step": 116800
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 3.266545665919309e-05,
+      "loss": 1.8599,
+      "step": 116900
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 3.265062813623261e-05,
+      "loss": 1.8474,
+      "step": 117000
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 3.263579961327212e-05,
+      "loss": 1.8443,
+      "step": 117100
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 3.2620971090311634e-05,
+      "loss": 1.8391,
+      "step": 117200
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 3.260614256735115e-05,
+      "loss": 1.8372,
+      "step": 117300
+    },
+    {
+      "epoch": 4.18,
+      "learning_rate": 3.259131404439067e-05,
+      "loss": 1.8278,
+      "step": 117400
+    },
+    {
+      "epoch": 4.18,
+      "learning_rate": 3.2576485521430184e-05,
+      "loss": 1.8248,
+      "step": 117500
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 3.2561656998469695e-05,
+      "loss": 1.8199,
+      "step": 117600
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 3.2546828475509214e-05,
+      "loss": 1.814,
+      "step": 117700
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 3.253199995254873e-05,
+      "loss": 1.813,
+      "step": 117800
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 3.2517171429588245e-05,
+      "loss": 1.8052,
+      "step": 117900
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 3.2502342906627757e-05,
+      "loss": 1.8019,
+      "step": 118000
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 3.2487514383667275e-05,
+      "loss": 1.7977,
+      "step": 118100
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 3.247268586070679e-05,
+      "loss": 1.7908,
+      "step": 118200
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 3.2457857337746306e-05,
+      "loss": 1.7882,
+      "step": 118300
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 3.244302881478582e-05,
+      "loss": 1.7808,
+      "step": 118400
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 3.2428200291825336e-05,
+      "loss": 1.7811,
+      "step": 118500
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 3.241337176886485e-05,
+      "loss": 1.7708,
+      "step": 118600
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 3.239854324590436e-05,
+      "loss": 1.7708,
+      "step": 118700
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 3.238371472294388e-05,
+      "loss": 1.7681,
+      "step": 118800
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 3.23688861999834e-05,
+      "loss": 1.7638,
+      "step": 118900
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 3.235405767702291e-05,
+      "loss": 1.763,
+      "step": 119000
+    },
+    {
+      "epoch": 4.24,
+      "eval_loss": 1.660704255104065,
+      "eval_runtime": 5267.7174,
+      "eval_samples_per_second": 606.907,
+      "eval_steps_per_second": 0.395,
+      "step": 119000
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 3.233922915406242e-05,
+      "loss": 1.7514,
+      "step": 119100
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 3.232440063110193e-05,
+      "loss": 1.7539,
+      "step": 119200
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 3.230957210814146e-05,
+      "loss": 1.748,
+      "step": 119300
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 3.229474358518097e-05,
+      "loss": 1.7446,
+      "step": 119400
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 3.227991506222048e-05,
+      "loss": 1.7415,
+      "step": 119500
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 3.226508653926e-05,
+      "loss": 1.7379,
+      "step": 119600
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 3.225025801629951e-05,
+      "loss": 1.7366,
+      "step": 119700
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 3.223542949333903e-05,
+      "loss": 1.7303,
+      "step": 119800
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 3.2220600970378543e-05,
+      "loss": 1.7306,
+      "step": 119900
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 3.220577244741806e-05,
+      "loss": 1.7261,
+      "step": 120000
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 3.2190943924457574e-05,
+      "loss": 1.7264,
+      "step": 120100
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 3.2176115401497086e-05,
+      "loss": 1.7197,
+      "step": 120200
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 3.2161286878536605e-05,
+      "loss": 1.7138,
+      "step": 120300
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 3.214645835557612e-05,
+      "loss": 1.7129,
+      "step": 120400
+    },
+    {
+      "epoch": 4.29,
+      "learning_rate": 3.2131629832615635e-05,
+      "loss": 1.712,
+      "step": 120500
+    },
+    {
+      "epoch": 4.29,
+      "learning_rate": 3.211680130965515e-05,
+      "loss": 1.708,
+      "step": 120600
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 3.210197278669466e-05,
+      "loss": 1.7059,
+      "step": 120700
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 3.2087144263734184e-05,
+      "loss": 1.7045,
+      "step": 120800
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 3.2072315740773696e-05,
+      "loss": 1.6968,
+      "step": 120900
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 3.205748721781321e-05,
+      "loss": 1.6983,
+      "step": 121000
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 3.204265869485272e-05,
+      "loss": 1.6883,
+      "step": 121100
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 3.2027830171892246e-05,
+      "loss": 1.6899,
+      "step": 121200
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 3.201300164893176e-05,
+      "loss": 1.6976,
+      "step": 121300
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 3.199817312597127e-05,
+      "loss": 1.6896,
+      "step": 121400
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 3.198334460301078e-05,
+      "loss": 1.6818,
+      "step": 121500
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 3.19685160800503e-05,
+      "loss": 1.6793,
+      "step": 121600
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 3.195368755708982e-05,
+      "loss": 1.6818,
+      "step": 121700
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 3.193885903412933e-05,
+      "loss": 1.6772,
+      "step": 121800
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 3.192403051116884e-05,
+      "loss": 1.6714,
+      "step": 121900
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 3.190920198820836e-05,
+      "loss": 1.6694,
+      "step": 122000
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 3.189437346524787e-05,
+      "loss": 1.6712,
+      "step": 122100
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 3.187954494228739e-05,
+      "loss": 1.6667,
+      "step": 122200
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 3.1864716419326903e-05,
+      "loss": 1.6695,
+      "step": 122300
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 3.184988789636642e-05,
+      "loss": 1.6665,
+      "step": 122400
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 3.1835059373405934e-05,
+      "loss": 1.6647,
+      "step": 122500
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 3.1820230850445446e-05,
+      "loss": 1.6574,
+      "step": 122600
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 3.1805402327484965e-05,
+      "loss": 1.6542,
+      "step": 122700
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 3.179057380452448e-05,
+      "loss": 1.6567,
+      "step": 122800
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 3.1775745281563995e-05,
+      "loss": 1.6537,
+      "step": 122900
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 3.176091675860351e-05,
+      "loss": 1.6494,
+      "step": 123000
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 3.1746088235643026e-05,
+      "loss": 1.6515,
+      "step": 123100
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 3.1731259712682544e-05,
+      "loss": 1.646,
+      "step": 123200
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 3.1716431189722056e-05,
+      "loss": 1.6456,
+      "step": 123300
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 3.170160266676157e-05,
+      "loss": 1.6433,
+      "step": 123400
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 3.168677414380109e-05,
+      "loss": 1.6382,
+      "step": 123500
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 3.16719456208406e-05,
+      "loss": 1.6406,
+      "step": 123600
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 3.165711709788012e-05,
+      "loss": 1.6316,
+      "step": 123700
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 3.164228857491963e-05,
+      "loss": 1.637,
+      "step": 123800
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 3.162746005195915e-05,
+      "loss": 1.6322,
+      "step": 123900
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 3.161263152899866e-05,
+      "loss": 1.6321,
+      "step": 124000
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 3.159780300603817e-05,
+      "loss": 1.6289,
+      "step": 124100
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 3.158297448307769e-05,
+      "loss": 1.6241,
+      "step": 124200
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 3.156814596011721e-05,
+      "loss": 1.6247,
+      "step": 124300
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 3.155331743715672e-05,
+      "loss": 1.6229,
+      "step": 124400
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 3.153848891419623e-05,
+      "loss": 1.6192,
+      "step": 124500
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 3.152366039123575e-05,
+      "loss": 1.619,
+      "step": 124600
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 3.150883186827527e-05,
+      "loss": 1.6203,
+      "step": 124700
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 3.149400334531478e-05,
+      "loss": 1.6139,
+      "step": 124800
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 3.1479174822354294e-05,
+      "loss": 1.6154,
+      "step": 124900
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 3.146434629939381e-05,
+      "loss": 1.6159,
+      "step": 125000
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 3.144951777643333e-05,
+      "loss": 1.6124,
+      "step": 125100
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 3.143468925347284e-05,
+      "loss": 1.6126,
+      "step": 125200
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 3.1419860730512355e-05,
+      "loss": 1.6095,
+      "step": 125300
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 3.1405032207551874e-05,
+      "loss": 1.6051,
+      "step": 125400
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 3.1390203684591386e-05,
+      "loss": 1.6035,
+      "step": 125500
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 3.1375375161630904e-05,
+      "loss": 1.6025,
+      "step": 125600
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 3.1360546638670416e-05,
+      "loss": 1.6039,
+      "step": 125700
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 3.1345718115709935e-05,
+      "loss": 1.6036,
+      "step": 125800
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 3.133088959274945e-05,
+      "loss": 1.5949,
+      "step": 125900
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 3.131606106978896e-05,
+      "loss": 1.5965,
+      "step": 126000
+    },
+    {
+      "epoch": 4.48,
+      "eval_loss": 1.4935563802719116,
+      "eval_runtime": 5265.7828,
+      "eval_samples_per_second": 607.13,
+      "eval_steps_per_second": 0.395,
+      "step": 126000
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 3.130123254682848e-05,
+      "loss": 1.5982,
+      "step": 126100
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 3.1286404023867996e-05,
+      "loss": 1.5976,
+      "step": 126200
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 3.127157550090751e-05,
+      "loss": 1.5872,
+      "step": 126300
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 3.125674697794702e-05,
+      "loss": 1.5846,
+      "step": 126400
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 3.124191845498654e-05,
+      "loss": 1.5899,
+      "step": 126500
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 3.122708993202606e-05,
+      "loss": 1.5856,
+      "step": 126600
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 3.121226140906557e-05,
+      "loss": 1.5869,
+      "step": 126700
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 3.119743288610508e-05,
+      "loss": 1.586,
+      "step": 126800
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 3.11826043631446e-05,
+      "loss": 1.5838,
+      "step": 126900
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 3.116777584018411e-05,
+      "loss": 1.5824,
+      "step": 127000
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 3.115294731722363e-05,
+      "loss": 1.5809,
+      "step": 127100
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 3.113811879426314e-05,
+      "loss": 1.5782,
+      "step": 127200
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 3.112329027130266e-05,
+      "loss": 1.5721,
+      "step": 127300
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 3.110846174834217e-05,
+      "loss": 1.5722,
+      "step": 127400
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 3.1093633225381685e-05,
+      "loss": 1.5741,
+      "step": 127500
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 3.10788047024212e-05,
+      "loss": 1.5802,
+      "step": 127600
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 3.106397617946072e-05,
+      "loss": 1.5717,
+      "step": 127700
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 3.1049147656500234e-05,
+      "loss": 1.5707,
+      "step": 127800
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 3.1034319133539746e-05,
+      "loss": 1.5683,
+      "step": 127900
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 3.101949061057926e-05,
+      "loss": 1.5671,
+      "step": 128000
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 3.100466208761878e-05,
+      "loss": 1.5686,
+      "step": 128100
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 3.0989833564658295e-05,
+      "loss": 1.5651,
+      "step": 128200
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 3.097500504169781e-05,
+      "loss": 1.5657,
+      "step": 128300
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 3.096017651873732e-05,
+      "loss": 1.5648,
+      "step": 128400
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 3.094534799577684e-05,
+      "loss": 1.564,
+      "step": 128500
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 3.0930519472816356e-05,
+      "loss": 1.5625,
+      "step": 128600
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 3.091569094985587e-05,
+      "loss": 1.5579,
+      "step": 128700
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 3.090086242689538e-05,
+      "loss": 1.5563,
+      "step": 128800
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 3.08860339039349e-05,
+      "loss": 1.5579,
+      "step": 128900
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 3.087120538097441e-05,
+      "loss": 1.5584,
+      "step": 129000
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 3.085637685801393e-05,
+      "loss": 1.5535,
+      "step": 129100
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 3.084154833505344e-05,
+      "loss": 1.5497,
+      "step": 129200
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 3.082671981209296e-05,
+      "loss": 1.5538,
+      "step": 129300
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 3.081189128913247e-05,
+      "loss": 1.5526,
+      "step": 129400
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 3.079706276617199e-05,
+      "loss": 1.5479,
+      "step": 129500
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 3.07822342432115e-05,
+      "loss": 1.5457,
+      "step": 129600
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.076740572025102e-05,
+      "loss": 1.548,
+      "step": 129700
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.075257719729053e-05,
+      "loss": 1.5494,
+      "step": 129800
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.0737748674330045e-05,
+      "loss": 1.5455,
+      "step": 129900
+    },
+    {
+      "epoch": 4.63,
+      "learning_rate": 3.072292015136956e-05,
+      "loss": 1.5455,
+      "step": 130000
+    },
+    {
+      "epoch": 4.63,
+      "learning_rate": 3.070809162840908e-05,
+      "loss": 1.5384,
+      "step": 130100
+    },
+    {
+      "epoch": 4.63,
+      "learning_rate": 3.0693263105448594e-05,
+      "loss": 1.5358,
+      "step": 130200
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 3.0678434582488106e-05,
+      "loss": 1.5392,
+      "step": 130300
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 3.0663606059527624e-05,
+      "loss": 1.5343,
+      "step": 130400
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 3.064877753656714e-05,
+      "loss": 1.5395,
+      "step": 130500
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.0633949013606655e-05,
+      "loss": 1.5365,
+      "step": 130600
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.061912049064617e-05,
+      "loss": 1.5328,
+      "step": 130700
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.0604291967685685e-05,
+      "loss": 1.535,
+      "step": 130800
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 3.05894634447252e-05,
+      "loss": 1.5325,
+      "step": 130900
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 3.0574634921764716e-05,
+      "loss": 1.5394,
+      "step": 131000
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.055980639880423e-05,
+      "loss": 1.5318,
+      "step": 131100
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.0544977875843747e-05,
+      "loss": 1.5266,
+      "step": 131200
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.053014935288326e-05,
+      "loss": 1.5321,
+      "step": 131300
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 3.0515320829922774e-05,
+      "loss": 1.5228,
+      "step": 131400
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 3.0500492306962292e-05,
+      "loss": 1.5286,
+      "step": 131500
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 3.0485663784001804e-05,
+      "loss": 1.5237,
+      "step": 131600
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.047083526104132e-05,
+      "loss": 1.5213,
+      "step": 131700
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.0456006738080835e-05,
+      "loss": 1.5261,
+      "step": 131800
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.0441178215120347e-05,
+      "loss": 1.5268,
+      "step": 131900
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 3.0426349692159865e-05,
+      "loss": 1.523,
+      "step": 132000
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 3.041152116919938e-05,
+      "loss": 1.5209,
+      "step": 132100
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 3.0396692646238896e-05,
+      "loss": 1.5209,
+      "step": 132200
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 3.0381864123278408e-05,
+      "loss": 1.5145,
+      "step": 132300
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 3.0367035600317923e-05,
+      "loss": 1.5202,
+      "step": 132400
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 3.0352207077357442e-05,
+      "loss": 1.5181,
+      "step": 132500
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 3.0337378554396957e-05,
+      "loss": 1.5158,
+      "step": 132600
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 3.032255003143647e-05,
+      "loss": 1.512,
+      "step": 132700
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 3.0307721508475984e-05,
+      "loss": 1.5124,
+      "step": 132800
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 3.0292892985515496e-05,
+      "loss": 1.5116,
+      "step": 132900
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 3.0278064462555018e-05,
+      "loss": 1.5115,
+      "step": 133000
+    },
+    {
+      "epoch": 4.73,
+      "eval_loss": 1.406484842300415,
+      "eval_runtime": 5267.2899,
+      "eval_samples_per_second": 606.956,
+      "eval_steps_per_second": 0.395,
+      "step": 133000
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 3.026323593959453e-05,
+      "loss": 1.5082,
+      "step": 133100
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 3.0248407416634045e-05,
+      "loss": 1.5074,
+      "step": 133200
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 3.0233578893673557e-05,
+      "loss": 1.5071,
+      "step": 133300
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 3.021875037071308e-05,
+      "loss": 1.5138,
+      "step": 133400
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 3.020392184775259e-05,
+      "loss": 1.5083,
+      "step": 133500
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 3.0189093324792107e-05,
+      "loss": 1.5055,
+      "step": 133600
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 3.017426480183162e-05,
+      "loss": 1.5071,
+      "step": 133700
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 3.0159436278871134e-05,
+      "loss": 1.504,
+      "step": 133800
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 3.0144607755910652e-05,
+      "loss": 1.5002,
+      "step": 133900
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 3.0129779232950168e-05,
+      "loss": 1.5044,
+      "step": 134000
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 3.011495070998968e-05,
+      "loss": 1.5019,
+      "step": 134100
+    },
+    {
+      "epoch": 4.78,
+      "learning_rate": 3.0100122187029195e-05,
+      "loss": 1.4975,
+      "step": 134200
+    },
+    {
+      "epoch": 4.78,
+      "learning_rate": 3.008529366406871e-05,
+      "loss": 1.5001,
+      "step": 134300
+    },
+    {
+      "epoch": 4.78,
+      "learning_rate": 3.007046514110823e-05,
+      "loss": 1.4989,
+      "step": 134400
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 3.005563661814774e-05,
+      "loss": 1.4973,
+      "step": 134500
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 3.0040808095187256e-05,
+      "loss": 1.5,
+      "step": 134600
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 3.002597957222677e-05,
+      "loss": 1.4921,
+      "step": 134700
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 3.0011151049266283e-05,
+      "loss": 1.4924,
+      "step": 134800
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 2.9996322526305802e-05,
+      "loss": 1.4897,
+      "step": 134900
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 2.9981494003345317e-05,
+      "loss": 1.4939,
+      "step": 135000
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 2.9966665480384832e-05,
+      "loss": 1.4914,
+      "step": 135100
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 2.9951836957424344e-05,
+      "loss": 1.4865,
+      "step": 135200
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 2.993700843446386e-05,
+      "loss": 1.4868,
+      "step": 135300
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 2.9922179911503378e-05,
+      "loss": 1.4868,
+      "step": 135400
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 2.9907351388542893e-05,
+      "loss": 1.49,
+      "step": 135500
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 2.9892522865582405e-05,
+      "loss": 1.4842,
+      "step": 135600
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 2.987769434262192e-05,
+      "loss": 1.4873,
+      "step": 135700
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 2.9862865819661433e-05,
+      "loss": 1.4853,
+      "step": 135800
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 2.9848037296700955e-05,
+      "loss": 1.4922,
+      "step": 135900
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 2.9833208773740467e-05,
+      "loss": 1.4846,
+      "step": 136000
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 2.9818380250779982e-05,
+      "loss": 1.4864,
+      "step": 136100
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 2.9803551727819494e-05,
+      "loss": 1.4808,
+      "step": 136200
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 2.978872320485901e-05,
+      "loss": 1.4811,
+      "step": 136300
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 2.9773894681898528e-05,
+      "loss": 1.4791,
+      "step": 136400
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 2.9759066158938043e-05,
+      "loss": 1.4797,
+      "step": 136500
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 2.9744237635977558e-05,
+      "loss": 1.4807,
+      "step": 136600
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 2.972940911301707e-05,
+      "loss": 1.479,
+      "step": 136700
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 2.9714580590056585e-05,
+      "loss": 1.4794,
+      "step": 136800
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 2.9699752067096104e-05,
+      "loss": 1.4774,
+      "step": 136900
+    },
+    {
+      "epoch": 4.88,
+      "learning_rate": 2.968492354413562e-05,
+      "loss": 1.4758,
+      "step": 137000
+    },
+    {
+      "epoch": 4.88,
+      "learning_rate": 2.967009502117513e-05,
+      "loss": 1.4745,
+      "step": 137100
+    },
+    {
+      "epoch": 4.88,
+      "learning_rate": 2.9655266498214646e-05,
+      "loss": 1.476,
+      "step": 137200
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 2.9640437975254165e-05,
+      "loss": 1.4696,
+      "step": 137300
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 2.962560945229368e-05,
+      "loss": 1.47,
+      "step": 137400
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 2.9610780929333192e-05,
+      "loss": 1.4709,
+      "step": 137500
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 2.9595952406372708e-05,
+      "loss": 1.4681,
+      "step": 137600
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 2.958112388341222e-05,
+      "loss": 1.4672,
+      "step": 137700
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 2.956629536045174e-05,
+      "loss": 1.4698,
+      "step": 137800
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 2.9551466837491253e-05,
+      "loss": 1.4667,
+      "step": 137900
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 2.953663831453077e-05,
+      "loss": 1.4699,
+      "step": 138000
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 2.952180979157028e-05,
+      "loss": 1.4624,
+      "step": 138100
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 2.9506981268609796e-05,
+      "loss": 1.4707,
+      "step": 138200
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 2.9492152745649315e-05,
+      "loss": 1.4635,
+      "step": 138300
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 2.947732422268883e-05,
+      "loss": 1.467,
+      "step": 138400
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 2.9462495699728342e-05,
+      "loss": 1.4611,
+      "step": 138500
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 2.9447667176767857e-05,
+      "loss": 1.4624,
+      "step": 138600
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 2.9432838653807372e-05,
+      "loss": 1.465,
+      "step": 138700
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 2.941801013084689e-05,
+      "loss": 1.4667,
+      "step": 138800
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 2.9403181607886403e-05,
+      "loss": 1.4608,
+      "step": 138900
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 2.9388353084925918e-05,
+      "loss": 1.46,
+      "step": 139000
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 2.9373524561965433e-05,
+      "loss": 1.4618,
+      "step": 139100
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 2.9358696039004945e-05,
+      "loss": 1.4591,
+      "step": 139200
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 2.9343867516044464e-05,
+      "loss": 1.4572,
+      "step": 139300
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 2.932903899308398e-05,
+      "loss": 1.4575,
+      "step": 139400
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 2.9314210470123495e-05,
+      "loss": 1.4535,
+      "step": 139500
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 2.9299381947163006e-05,
+      "loss": 1.4577,
+      "step": 139600
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 2.9284553424202522e-05,
+      "loss": 1.4532,
+      "step": 139700
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 2.926972490124204e-05,
+      "loss": 1.456,
+      "step": 139800
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 2.9254896378281556e-05,
+      "loss": 1.4543,
+      "step": 139900
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 2.9240067855321068e-05,
+      "loss": 1.4536,
+      "step": 140000
+    },
+    {
+      "epoch": 4.98,
+      "eval_loss": 1.347744345664978,
+      "eval_runtime": 5266.9309,
+      "eval_samples_per_second": 606.998,
+      "eval_steps_per_second": 0.395,
+      "step": 140000
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 2.9225239332360583e-05,
+      "loss": 1.4541,
+      "step": 140100
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 2.9210410809400095e-05,
+      "loss": 1.4511,
+      "step": 140200
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 2.9195582286439617e-05,
+      "loss": 1.4538,
+      "step": 140300
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.918075376347913e-05,
+      "loss": 1.4509,
+      "step": 140400
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.9165925240518644e-05,
+      "loss": 1.446,
+      "step": 140500
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 2.9151096717558156e-05,
+      "loss": 1.447,
+      "step": 140600
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 2.913626819459767e-05,
+      "loss": 1.4463,
+      "step": 140700
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 2.912143967163719e-05,
+      "loss": 1.4456,
+      "step": 140800
+    },
+    {
+      "epoch": 5.01,
+      "learning_rate": 2.9106611148676705e-05,
+      "loss": 1.4439,
+      "step": 140900
+    },
+    {
+      "epoch": 5.02,
+      "learning_rate": 2.9091782625716217e-05,
+      "loss": 1.4504,
+      "step": 141000
+    },
+    {
+      "epoch": 5.02,
+      "learning_rate": 2.9076954102755732e-05,
+      "loss": 1.4403,
+      "step": 141100
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 2.906212557979525e-05,
+      "loss": 1.4411,
+      "step": 141200
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 2.9047297056834766e-05,
+      "loss": 1.4432,
+      "step": 141300
+    },
+    {
+      "epoch": 5.03,
+      "learning_rate": 2.9032468533874278e-05,
+      "loss": 1.4474,
+      "step": 141400
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 2.9017640010913793e-05,
+      "loss": 1.4458,
+      "step": 141500
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 2.900281148795331e-05,
+      "loss": 1.4414,
+      "step": 141600
+    },
+    {
+      "epoch": 5.04,
+      "learning_rate": 2.8987982964992827e-05,
+      "loss": 1.4397,
+      "step": 141700
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 2.897315444203234e-05,
+      "loss": 1.4379,
+      "step": 141800
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 2.8958325919071855e-05,
+      "loss": 1.4366,
+      "step": 141900
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 2.894349739611137e-05,
+      "loss": 1.4409,
+      "step": 142000
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 2.8928668873150882e-05,
+      "loss": 1.4328,
+      "step": 142100
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 2.89138403501904e-05,
+      "loss": 1.4375,
+      "step": 142200
+    },
+    {
+      "epoch": 5.06,
+      "learning_rate": 2.8899011827229916e-05,
+      "loss": 1.4345,
+      "step": 142300
+    },
+    {
+      "epoch": 5.07,
+      "learning_rate": 2.888418330426943e-05,
+      "loss": 1.4348,
+      "step": 142400
+    },
+    {
+      "epoch": 5.07,
+      "learning_rate": 2.8869354781308943e-05,
+      "loss": 1.4358,
+      "step": 142500
+    },
+    {
+      "epoch": 5.07,
+      "learning_rate": 2.8854526258348458e-05,
+      "loss": 1.4339,
+      "step": 142600
+    },
+    {
+      "epoch": 5.08,
+      "learning_rate": 2.8839697735387977e-05,
+      "loss": 1.4373,
+      "step": 142700
+    },
+    {
+      "epoch": 5.08,
+      "learning_rate": 2.8824869212427492e-05,
+      "loss": 1.4298,
+      "step": 142800
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 2.8810040689467004e-05,
+      "loss": 1.4299,
+      "step": 142900
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 2.879521216650652e-05,
+      "loss": 1.4268,
+      "step": 143000
+    },
+    {
+      "epoch": 5.09,
+      "learning_rate": 2.878038364354603e-05,
+      "loss": 1.4307,
+      "step": 143100
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 2.8765555120585553e-05,
+      "loss": 1.432,
+      "step": 143200
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 2.8750726597625065e-05,
+      "loss": 1.4299,
+      "step": 143300
+    },
+    {
+      "epoch": 5.1,
+      "learning_rate": 2.873589807466458e-05,
+      "loss": 1.4255,
+      "step": 143400
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 2.8721069551704096e-05,
+      "loss": 1.4291,
+      "step": 143500
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 2.8706241028743608e-05,
+      "loss": 1.4341,
+      "step": 143600
+    },
+    {
+      "epoch": 5.11,
+      "learning_rate": 2.8691412505783126e-05,
+      "loss": 1.4292,
+      "step": 143700
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 2.867658398282264e-05,
+      "loss": 1.4275,
+      "step": 143800
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 2.8661755459862157e-05,
+      "loss": 1.4284,
+      "step": 143900
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 2.864692693690167e-05,
+      "loss": 1.4261,
+      "step": 144000
+    },
+    {
+      "epoch": 5.13,
+      "learning_rate": 2.8632098413941184e-05,
+      "loss": 1.4226,
+      "step": 144100
+    },
+    {
+      "epoch": 5.13,
+      "learning_rate": 2.8617269890980703e-05,
+      "loss": 1.4253,
+      "step": 144200
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 2.8602441368020218e-05,
+      "loss": 1.4203,
+      "step": 144300
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 2.858761284505973e-05,
+      "loss": 1.4256,
+      "step": 144400
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 2.8572784322099245e-05,
+      "loss": 1.421,
+      "step": 144500
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 2.8557955799138757e-05,
+      "loss": 1.4241,
+      "step": 144600
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 2.854312727617828e-05,
+      "loss": 1.4224,
+      "step": 144700
+    },
+    {
+      "epoch": 5.15,
+      "learning_rate": 2.852829875321779e-05,
+      "loss": 1.4218,
+      "step": 144800
+    },
+    {
+      "epoch": 5.16,
+      "learning_rate": 2.8513470230257306e-05,
+      "loss": 1.421,
+      "step": 144900
+    },
+    {
+      "epoch": 5.16,
+      "learning_rate": 2.8498641707296818e-05,
+      "loss": 1.4184,
+      "step": 145000
+    },
+    {
+      "epoch": 5.16,
+      "learning_rate": 2.848381318433634e-05,
+      "loss": 1.4194,
+      "step": 145100
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 2.8468984661375852e-05,
+      "loss": 1.4187,
+      "step": 145200
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 2.8454156138415367e-05,
+      "loss": 1.4169,
+      "step": 145300
+    },
+    {
+      "epoch": 5.17,
+      "learning_rate": 2.843932761545488e-05,
+      "loss": 1.4151,
+      "step": 145400
+    },
+    {
+      "epoch": 5.18,
+      "learning_rate": 2.8424499092494395e-05,
+      "loss": 1.4164,
+      "step": 145500
+    },
+    {
+      "epoch": 5.18,
+      "learning_rate": 2.8409670569533913e-05,
+      "loss": 1.4192,
+      "step": 145600
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 2.839484204657343e-05,
+      "loss": 1.4171,
+      "step": 145700
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 2.838001352361294e-05,
+      "loss": 1.4178,
+      "step": 145800
+    },
+    {
+      "epoch": 5.19,
+      "learning_rate": 2.8365185000652456e-05,
+      "loss": 1.4169,
+      "step": 145900
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 2.835035647769197e-05,
+      "loss": 1.4152,
+      "step": 146000
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 2.833552795473149e-05,
+      "loss": 1.4179,
+      "step": 146100
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 2.8320699431771e-05,
+      "loss": 1.4128,
+      "step": 146200
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 2.8305870908810517e-05,
+      "loss": 1.4114,
+      "step": 146300
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 2.8291042385850032e-05,
+      "loss": 1.4137,
+      "step": 146400
+    },
+    {
+      "epoch": 5.21,
+      "learning_rate": 2.8276213862889544e-05,
+      "loss": 1.4109,
+      "step": 146500
+    },
+    {
+      "epoch": 5.22,
+      "learning_rate": 2.8261385339929063e-05,
+      "loss": 1.4152,
+      "step": 146600
+    },
+    {
+      "epoch": 5.22,
+      "learning_rate": 2.8246556816968578e-05,
+      "loss": 1.4132,
+      "step": 146700
+    },
+    {
+      "epoch": 5.22,
+      "learning_rate": 2.8231728294008093e-05,
+      "loss": 1.4139,
+      "step": 146800
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 2.8216899771047605e-05,
+      "loss": 1.4097,
+      "step": 146900
+    },
+    {
+      "epoch": 5.23,
+      "learning_rate": 2.820207124808712e-05,
+      "loss": 1.4096,
+      "step": 147000
+    },
+    {
+      "epoch": 5.23,
+      "eval_loss": 1.3063678741455078,
+      "eval_runtime": 5266.6946,
+      "eval_samples_per_second": 607.025,
+      "eval_steps_per_second": 0.395,
+      "step": 147000
+    },
+    {
+      "epoch": 5.24,
+      "learning_rate": 2.818724272512664e-05,
+      "loss": 1.4057,
+      "step": 147100
+    },
+    {
+      "epoch": 5.24,
+      "learning_rate": 2.8172414202166154e-05,
+      "loss": 1.4041,
+      "step": 147200
+    },
+    {
+      "epoch": 5.24,
+      "learning_rate": 2.8157585679205666e-05,
+      "loss": 1.4117,
+      "step": 147300
+    },
+    {
+      "epoch": 5.25,
+      "learning_rate": 2.814275715624518e-05,
+      "loss": 1.4044,
+      "step": 147400
+    },
+    {
+      "epoch": 5.25,
+      "learning_rate": 2.8127928633284693e-05,
+      "loss": 1.4054,
+      "step": 147500
+    },
+    {
+      "epoch": 5.25,
+      "learning_rate": 2.8113100110324215e-05,
+      "loss": 1.4058,
+      "step": 147600
+    },
+    {
+      "epoch": 5.26,
+      "learning_rate": 2.8098271587363727e-05,
+      "loss": 1.4034,
+      "step": 147700
+    },
+    {
+      "epoch": 5.26,
+      "learning_rate": 2.8083443064403243e-05,
+      "loss": 1.4057,
+      "step": 147800
+    },
+    {
+      "epoch": 5.26,
+      "learning_rate": 2.8068614541442754e-05,
+      "loss": 1.4041,
+      "step": 147900
+    },
+    {
+      "epoch": 5.27,
+      "learning_rate": 2.805378601848227e-05,
+      "loss": 1.4012,
+      "step": 148000
+    },
+    {
+      "epoch": 5.27,
+      "learning_rate": 2.803895749552179e-05,
+      "loss": 1.4003,
+      "step": 148100
+    },
+    {
+      "epoch": 5.27,
+      "learning_rate": 2.8024128972561304e-05,
+      "loss": 1.4017,
+      "step": 148200
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 2.8009300449600816e-05,
+      "loss": 1.4006,
+      "step": 148300
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 2.799447192664033e-05,
+      "loss": 1.4011,
+      "step": 148400
+    },
+    {
+      "epoch": 5.28,
+      "learning_rate": 2.7979643403679846e-05,
+      "loss": 1.3981,
+      "step": 148500
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 2.7964814880719365e-05,
+      "loss": 1.3981,
+      "step": 148600
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 2.7949986357758877e-05,
+      "loss": 1.4017,
+      "step": 148700
+    },
+    {
+      "epoch": 5.3,
+      "learning_rate": 2.7935157834798392e-05,
+      "loss": 1.3974,
+      "step": 148800
+    },
+    {
+      "epoch": 5.3,
+      "learning_rate": 2.7920329311837907e-05,
+      "loss": 1.399,
+      "step": 148900
+    },
+    {
+      "epoch": 5.3,
+      "learning_rate": 2.7905500788877426e-05,
+      "loss": 1.3965,
+      "step": 149000
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 2.7890672265916938e-05,
+      "loss": 1.3995,
+      "step": 149100
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 2.7875843742956453e-05,
+      "loss": 1.3926,
+      "step": 149200
+    },
+    {
+      "epoch": 5.31,
+      "learning_rate": 2.786101521999597e-05,
+      "loss": 1.3932,
+      "step": 149300
+    },
+    {
+      "epoch": 5.32,
+      "learning_rate": 2.784618669703548e-05,
+      "loss": 1.3966,
+      "step": 149400
+    },
+    {
+      "epoch": 5.32,
+      "learning_rate": 2.7831358174075e-05,
+      "loss": 1.3986,
+      "step": 149500
+    },
+    {
+      "epoch": 5.32,
+      "learning_rate": 2.7816529651114514e-05,
+      "loss": 1.3959,
+      "step": 149600
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 2.780170112815403e-05,
+      "loss": 1.3959,
+      "step": 149700
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 2.778687260519354e-05,
+      "loss": 1.3982,
+      "step": 149800
+    },
+    {
+      "epoch": 5.33,
+      "learning_rate": 2.7772044082233057e-05,
+      "loss": 1.3925,
+      "step": 149900
+    },
+    {
+      "epoch": 5.34,
+      "learning_rate": 2.7757215559272575e-05,
+      "loss": 1.3927,
+      "step": 150000
+    },
+    {
+      "epoch": 5.34,
+      "learning_rate": 2.774238703631209e-05,
+      "loss": 1.3906,
+      "step": 150100
+    },
+    {
+      "epoch": 5.35,
+      "learning_rate": 2.7727558513351603e-05,
+      "loss": 1.3885,
+      "step": 150200
+    },
+    {
+      "epoch": 5.35,
+      "learning_rate": 2.7712729990391118e-05,
+      "loss": 1.3915,
+      "step": 150300
+    },
+    {
+      "epoch": 5.35,
+      "learning_rate": 2.769790146743063e-05,
+      "loss": 1.3892,
+      "step": 150400
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 2.7683072944470152e-05,
+      "loss": 1.3881,
+      "step": 150500
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 2.7668244421509664e-05,
+      "loss": 1.389,
+      "step": 150600
+    },
+    {
+      "epoch": 5.36,
+      "learning_rate": 2.765341589854918e-05,
+      "loss": 1.3808,
+      "step": 150700
+    },
+    {
+      "epoch": 5.37,
+      "learning_rate": 2.7638587375588694e-05,
+      "loss": 1.3872,
+      "step": 150800
+    },
+    {
+      "epoch": 5.37,
+      "learning_rate": 2.7623758852628206e-05,
+      "loss": 1.3853,
+      "step": 150900
+    },
+    {
+      "epoch": 5.37,
+      "learning_rate": 2.7608930329667725e-05,
+      "loss": 1.3855,
+      "step": 151000
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 2.759410180670724e-05,
+      "loss": 1.389,
+      "step": 151100
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 2.7579273283746755e-05,
+      "loss": 1.3837,
+      "step": 151200
+    },
+    {
+      "epoch": 5.38,
+      "learning_rate": 2.7564444760786267e-05,
+      "loss": 1.3848,
+      "step": 151300
+    },
+    {
+      "epoch": 5.39,
+      "learning_rate": 2.7549616237825783e-05,
+      "loss": 1.3872,
+      "step": 151400
+    },
+    {
+      "epoch": 5.39,
+      "learning_rate": 2.75347877148653e-05,
+      "loss": 1.3812,
+      "step": 151500
+    },
+    {
+      "epoch": 5.4,
+      "learning_rate": 2.7519959191904817e-05,
+      "loss": 1.3859,
+      "step": 151600
+    },
+    {
+      "epoch": 5.4,
+      "learning_rate": 2.750513066894433e-05,
+      "loss": 1.3824,
+      "step": 151700
+    },
+    {
+      "epoch": 5.4,
+      "learning_rate": 2.7490302145983844e-05,
+      "loss": 1.3857,
+      "step": 151800
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 2.7475473623023356e-05,
+      "loss": 1.3829,
+      "step": 151900
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 2.7460645100062878e-05,
+      "loss": 1.384,
+      "step": 152000
+    },
+    {
+      "epoch": 5.41,
+      "learning_rate": 2.744581657710239e-05,
+      "loss": 1.3846,
+      "step": 152100
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 2.7430988054141905e-05,
+      "loss": 1.3823,
+      "step": 152200
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 2.7416159531181417e-05,
+      "loss": 1.3836,
+      "step": 152300
+    },
+    {
+      "epoch": 5.42,
+      "learning_rate": 2.7401331008220932e-05,
+      "loss": 1.3819,
+      "step": 152400
+    },
+    {
+      "epoch": 5.43,
+      "learning_rate": 2.738650248526045e-05,
+      "loss": 1.3802,
+      "step": 152500
+    },
+    {
+      "epoch": 5.43,
+      "learning_rate": 2.7371673962299966e-05,
+      "loss": 1.3825,
+      "step": 152600
+    },
+    {
+      "epoch": 5.43,
+      "learning_rate": 2.7356845439339478e-05,
+      "loss": 1.385,
+      "step": 152700
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 2.7342016916378993e-05,
+      "loss": 1.3845,
+      "step": 152800
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 2.7327188393418512e-05,
+      "loss": 1.378,
+      "step": 152900
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 2.7312359870458027e-05,
+      "loss": 1.3807,
+      "step": 153000
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 2.729753134749754e-05,
+      "loss": 1.3773,
+      "step": 153100
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 2.7282702824537054e-05,
+      "loss": 1.3797,
+      "step": 153200
+    },
+    {
+      "epoch": 5.46,
+      "learning_rate": 2.726787430157657e-05,
+      "loss": 1.3765,
+      "step": 153300
+    },
+    {
+      "epoch": 5.46,
+      "learning_rate": 2.7253045778616088e-05,
+      "loss": 1.3761,
+      "step": 153400
+    },
+    {
+      "epoch": 5.46,
+      "learning_rate": 2.72382172556556e-05,
+      "loss": 1.376,
+      "step": 153500
+    },
+    {
+      "epoch": 5.47,
+      "learning_rate": 2.7223388732695115e-05,
+      "loss": 1.3766,
+      "step": 153600
+    },
+    {
+      "epoch": 5.47,
+      "learning_rate": 2.720856020973463e-05,
+      "loss": 1.377,
+      "step": 153700
+    },
+    {
+      "epoch": 5.47,
+      "learning_rate": 2.7193731686774143e-05,
+      "loss": 1.3772,
+      "step": 153800
+    },
+    {
+      "epoch": 5.48,
+      "learning_rate": 2.717890316381366e-05,
+      "loss": 1.3705,
+      "step": 153900
+    },
+    {
+      "epoch": 5.48,
+      "learning_rate": 2.7164074640853176e-05,
+      "loss": 1.3737,
+      "step": 154000
+    },
+    {
+      "epoch": 5.48,
+      "eval_loss": 1.2714881896972656,
+      "eval_runtime": 5266.973,
+      "eval_samples_per_second": 606.993,
+      "eval_steps_per_second": 0.395,
+      "step": 154000
+    },
+    {
+      "epoch": 5.48,
+      "learning_rate": 2.7149246117892692e-05,
+      "loss": 1.3707,
+      "step": 154100
+    },
+    {
+      "epoch": 5.49,
+      "learning_rate": 2.7134417594932204e-05,
+      "loss": 1.3735,
+      "step": 154200
+    },
+    {
+      "epoch": 5.49,
+      "learning_rate": 2.711958907197172e-05,
+      "loss": 1.37,
+      "step": 154300
+    },
+    {
+      "epoch": 5.49,
+      "learning_rate": 2.7104760549011238e-05,
+      "loss": 1.3709,
+      "step": 154400
+    },
+    {
+      "epoch": 5.5,
+      "learning_rate": 2.7089932026050753e-05,
+      "loss": 1.3713,
+      "step": 154500
+    },
+    {
+      "epoch": 5.5,
+      "learning_rate": 2.7075103503090265e-05,
+      "loss": 1.3695,
+      "step": 154600
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 2.706027498012978e-05,
+      "loss": 1.3724,
+      "step": 154700
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 2.7045446457169292e-05,
+      "loss": 1.3718,
+      "step": 154800
+    },
+    {
+      "epoch": 5.51,
+      "learning_rate": 2.7030617934208814e-05,
+      "loss": 1.3705,
+      "step": 154900
+    },
+    {
+      "epoch": 5.52,
+      "learning_rate": 2.7015789411248326e-05,
+      "loss": 1.3695,
+      "step": 155000
+    },
+    {
+      "epoch": 5.52,
+      "learning_rate": 2.700096088828784e-05,
+      "loss": 1.3681,
+      "step": 155100
+    },
+    {
+      "epoch": 5.52,
+      "learning_rate": 2.6986132365327353e-05,
+      "loss": 1.3727,
+      "step": 155200
+    },
+    {
+      "epoch": 5.53,
+      "learning_rate": 2.697130384236687e-05,
+      "loss": 1.3673,
+      "step": 155300
+    },
+    {
+      "epoch": 5.53,
+      "learning_rate": 2.6956475319406387e-05,
+      "loss": 1.3698,
+      "step": 155400
+    },
+    {
+      "epoch": 5.53,
+      "learning_rate": 2.6941646796445902e-05,
+      "loss": 1.3658,
+      "step": 155500
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 2.6926818273485414e-05,
+      "loss": 1.3688,
+      "step": 155600
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 2.691198975052493e-05,
+      "loss": 1.365,
+      "step": 155700
+    },
+    {
+      "epoch": 5.54,
+      "learning_rate": 2.6897161227564445e-05,
+      "loss": 1.3676,
+      "step": 155800
+    },
+    {
+      "epoch": 5.55,
+      "learning_rate": 2.6882332704603963e-05,
+      "loss": 1.3694,
+      "step": 155900
+    },
+    {
+      "epoch": 5.55,
+      "learning_rate": 2.6867504181643475e-05,
+      "loss": 1.3679,
+      "step": 156000
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 2.685267565868299e-05,
+      "loss": 1.3646,
+      "step": 156100
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 2.6837847135722506e-05,
+      "loss": 1.3682,
+      "step": 156200
+    },
+    {
+      "epoch": 5.56,
+      "learning_rate": 2.6823018612762018e-05,
+      "loss": 1.367,
+      "step": 156300
+    },
+    {
+      "epoch": 5.57,
+      "learning_rate": 2.6808190089801536e-05,
+      "loss": 1.3673,
+      "step": 156400
+    },
+    {
+      "epoch": 5.57,
+      "learning_rate": 2.6793361566841052e-05,
+      "loss": 1.3674,
+      "step": 156500
+    },
+    {
+      "epoch": 5.57,
+      "learning_rate": 2.6778533043880567e-05,
+      "loss": 1.3638,
+      "step": 156600
+    },
+    {
+      "epoch": 5.58,
+      "learning_rate": 2.676370452092008e-05,
+      "loss": 1.3681,
+      "step": 156700
+    },
+    {
+      "epoch": 5.58,
+      "learning_rate": 2.6748875997959594e-05,
+      "loss": 1.3589,
+      "step": 156800
+    },
+    {
+      "epoch": 5.58,
+      "learning_rate": 2.6734047474999113e-05,
+      "loss": 1.3589,
+      "step": 156900
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 2.6719218952038628e-05,
+      "loss": 1.361,
+      "step": 157000
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 2.670439042907814e-05,
+      "loss": 1.3622,
+      "step": 157100
+    },
+    {
+      "epoch": 5.59,
+      "learning_rate": 2.6689561906117655e-05,
+      "loss": 1.359,
+      "step": 157200
+    },
+    {
+      "epoch": 5.6,
+      "learning_rate": 2.6674733383157174e-05,
+      "loss": 1.3586,
+      "step": 157300
+    },
+    {
+      "epoch": 5.6,
+      "learning_rate": 2.665990486019669e-05,
+      "loss": 1.3584,
+      "step": 157400
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 2.66450763372362e-05,
+      "loss": 1.3601,
+      "step": 157500
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 2.6630247814275716e-05,
+      "loss": 1.3536,
+      "step": 157600
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 2.6615419291315232e-05,
+      "loss": 1.3556,
+      "step": 157700
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 2.660059076835475e-05,
+      "loss": 1.3578,
+      "step": 157800
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 2.6585762245394262e-05,
+      "loss": 1.3576,
+      "step": 157900
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 2.6570933722433778e-05,
+      "loss": 1.3573,
+      "step": 158000
+    },
+    {
+      "epoch": 5.63,
+      "learning_rate": 2.6556105199473293e-05,
+      "loss": 1.3575,
+      "step": 158100
+    },
+    {
+      "epoch": 5.63,
+      "learning_rate": 2.6541276676512805e-05,
+      "loss": 1.3539,
+      "step": 158200
+    },
+    {
+      "epoch": 5.63,
+      "learning_rate": 2.6526448153552323e-05,
+      "loss": 1.3535,
+      "step": 158300
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 2.651161963059184e-05,
+      "loss": 1.3574,
+      "step": 158400
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 2.6496791107631354e-05,
+      "loss": 1.3544,
+      "step": 158500
+    },
+    {
+      "epoch": 5.64,
+      "learning_rate": 2.6481962584670866e-05,
+      "loss": 1.3585,
+      "step": 158600
+    },
+    {
+      "epoch": 5.65,
+      "learning_rate": 2.646713406171038e-05,
+      "loss": 1.3563,
+      "step": 158700
+    },
+    {
+      "epoch": 5.65,
+      "learning_rate": 2.64523055387499e-05,
+      "loss": 1.3498,
+      "step": 158800
+    },
+    {
+      "epoch": 5.66,
+      "learning_rate": 2.6437477015789415e-05,
+      "loss": 1.3512,
+      "step": 158900
+    },
+    {
+      "epoch": 5.66,
+      "learning_rate": 2.6422648492828927e-05,
+      "loss": 1.3525,
+      "step": 159000
+    },
+    {
+      "epoch": 5.66,
+      "learning_rate": 2.6407819969868442e-05,
+      "loss": 1.3546,
+      "step": 159100
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 2.6392991446907954e-05,
+      "loss": 1.351,
+      "step": 159200
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 2.6378162923947476e-05,
+      "loss": 1.3531,
+      "step": 159300
+    },
+    {
+      "epoch": 5.67,
+      "learning_rate": 2.6363334400986988e-05,
+      "loss": 1.3488,
+      "step": 159400
+    },
+    {
+      "epoch": 5.68,
+      "learning_rate": 2.6348505878026503e-05,
+      "loss": 1.349,
+      "step": 159500
+    },
+    {
+      "epoch": 5.68,
+      "learning_rate": 2.6333677355066015e-05,
+      "loss": 1.3524,
+      "step": 159600
+    },
+    {
+      "epoch": 5.68,
+      "learning_rate": 2.631884883210553e-05,
+      "loss": 1.3487,
+      "step": 159700
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 2.630402030914505e-05,
+      "loss": 1.3497,
+      "step": 159800
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 2.6289191786184565e-05,
+      "loss": 1.3498,
+      "step": 159900
+    },
+    {
+      "epoch": 5.69,
+      "learning_rate": 2.6274363263224076e-05,
+      "loss": 1.3533,
+      "step": 160000
+    },
+    {
+      "epoch": 5.7,
+      "learning_rate": 2.6259534740263592e-05,
+      "loss": 1.3474,
+      "step": 160100
+    },
+    {
+      "epoch": 5.7,
+      "learning_rate": 2.6244706217303107e-05,
+      "loss": 1.3488,
+      "step": 160200
+    },
+    {
+      "epoch": 5.7,
+      "learning_rate": 2.6229877694342626e-05,
+      "loss": 1.3496,
+      "step": 160300
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 2.6215049171382138e-05,
+      "loss": 1.352,
+      "step": 160400
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 2.6200220648421653e-05,
+      "loss": 1.3465,
+      "step": 160500
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 2.6185392125461168e-05,
+      "loss": 1.3448,
+      "step": 160600
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 2.617056360250068e-05,
+      "loss": 1.3499,
+      "step": 160700
+    },
+    {
+      "epoch": 5.72,
+      "learning_rate": 2.61557350795402e-05,
+      "loss": 1.3477,
+      "step": 160800
+    },
+    {
+      "epoch": 5.73,
+      "learning_rate": 2.6140906556579714e-05,
+      "loss": 1.3424,
+      "step": 160900
+    },
+    {
+      "epoch": 5.73,
+      "learning_rate": 2.612607803361923e-05,
+      "loss": 1.3457,
+      "step": 161000
+    },
+    {
+      "epoch": 5.73,
+      "eval_loss": 1.243394374847412,
+      "eval_runtime": 5266.0707,
+      "eval_samples_per_second": 607.097,
+      "eval_steps_per_second": 0.395,
+      "step": 161000
+    },
+    {
+      "epoch": 5.73,
+      "learning_rate": 2.611124951065874e-05,
+      "loss": 1.3487,
+      "step": 161100
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 2.609642098769826e-05,
+      "loss": 1.3396,
+      "step": 161200
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 2.6081592464737775e-05,
+      "loss": 1.3445,
+      "step": 161300
+    },
+    {
+      "epoch": 5.74,
+      "learning_rate": 2.606676394177729e-05,
+      "loss": 1.3422,
+      "step": 161400
+    },
+    {
+      "epoch": 5.75,
+      "learning_rate": 2.6051935418816802e-05,
+      "loss": 1.3394,
+      "step": 161500
+    },
+    {
+      "epoch": 5.75,
+      "learning_rate": 2.6037106895856318e-05,
+      "loss": 1.3459,
+      "step": 161600
+    },
+    {
+      "epoch": 5.75,
+      "learning_rate": 2.6022278372895836e-05,
+      "loss": 1.338,
+      "step": 161700
+    },
+    {
+      "epoch": 5.76,
+      "learning_rate": 2.600744984993535e-05,
+      "loss": 1.3411,
+      "step": 161800
+    },
+    {
+      "epoch": 5.76,
+      "learning_rate": 2.5992621326974863e-05,
+      "loss": 1.3402,
+      "step": 161900
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 2.597779280401438e-05,
+      "loss": 1.344,
+      "step": 162000
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 2.596296428105389e-05,
+      "loss": 1.3379,
+      "step": 162100
+    },
+    {
+      "epoch": 5.77,
+      "learning_rate": 2.5948135758093413e-05,
+      "loss": 1.3384,
+      "step": 162200
+    },
+    {
+      "epoch": 5.78,
+      "learning_rate": 2.5933307235132925e-05,
+      "loss": 1.3445,
+      "step": 162300
+    },
+    {
+      "epoch": 5.78,
+      "learning_rate": 2.591847871217244e-05,
+      "loss": 1.338,
+      "step": 162400
+    },
+    {
+      "epoch": 5.78,
+      "learning_rate": 2.590365018921195e-05,
+      "loss": 1.334,
+      "step": 162500
+    },
+    {
+      "epoch": 5.79,
+      "learning_rate": 2.5888821666251467e-05,
+      "loss": 1.3393,
+      "step": 162600
+    },
+    {
+      "epoch": 5.79,
+      "learning_rate": 2.5873993143290986e-05,
+      "loss": 1.3389,
+      "step": 162700
+    },
+    {
+      "epoch": 5.79,
+      "learning_rate": 2.58591646203305e-05,
+      "loss": 1.3422,
+      "step": 162800
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 2.5844336097370013e-05,
+      "loss": 1.3348,
+      "step": 162900
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 2.5829507574409528e-05,
+      "loss": 1.3366,
+      "step": 163000
+    },
+    {
+      "epoch": 5.8,
+      "learning_rate": 2.5814679051449043e-05,
+      "loss": 1.3369,
+      "step": 163100
+    },
+    {
+      "epoch": 5.81,
+      "learning_rate": 2.5799850528488562e-05,
+      "loss": 1.3359,
+      "step": 163200
+    },
+    {
+      "epoch": 5.81,
+      "learning_rate": 2.5785022005528074e-05,
+      "loss": 1.3395,
+      "step": 163300
+    },
+    {
+      "epoch": 5.82,
+      "learning_rate": 2.577019348256759e-05,
+      "loss": 1.3401,
+      "step": 163400
+    },
+    {
+      "epoch": 5.82,
+      "learning_rate": 2.5755364959607104e-05,
+      "loss": 1.3355,
+      "step": 163500
+    },
+    {
+      "epoch": 5.82,
+      "learning_rate": 2.5740536436646616e-05,
+      "loss": 1.3383,
+      "step": 163600
+    },
+    {
+      "epoch": 5.83,
+      "learning_rate": 2.572570791368614e-05,
+      "loss": 1.3343,
+      "step": 163700
+    },
+    {
+      "epoch": 5.83,
+      "learning_rate": 2.571087939072565e-05,
+      "loss": 1.3327,
+      "step": 163800
+    },
+    {
+      "epoch": 5.83,
+      "learning_rate": 2.5696050867765166e-05,
+      "loss": 1.3324,
+      "step": 163900
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 2.5681222344804678e-05,
+      "loss": 1.335,
+      "step": 164000
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 2.5666393821844193e-05,
+      "loss": 1.3318,
+      "step": 164100
+    },
+    {
+      "epoch": 5.84,
+      "learning_rate": 2.565156529888371e-05,
+      "loss": 1.3382,
+      "step": 164200
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 2.5636736775923227e-05,
+      "loss": 1.3324,
+      "step": 164300
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 2.562190825296274e-05,
+      "loss": 1.3308,
+      "step": 164400
+    },
+    {
+      "epoch": 5.85,
+      "learning_rate": 2.5607079730002254e-05,
+      "loss": 1.3345,
+      "step": 164500
+    },
+    {
+      "epoch": 5.86,
+      "learning_rate": 2.5592251207041766e-05,
+      "loss": 1.3252,
+      "step": 164600
+    },
+    {
+      "epoch": 5.86,
+      "learning_rate": 2.5577422684081288e-05,
+      "loss": 1.3304,
+      "step": 164700
+    },
+    {
+      "epoch": 5.86,
+      "learning_rate": 2.55625941611208e-05,
+      "loss": 1.3318,
+      "step": 164800
+    },
+    {
+      "epoch": 5.87,
+      "learning_rate": 2.5547765638160315e-05,
+      "loss": 1.3314,
+      "step": 164900
+    },
+    {
+      "epoch": 5.87,
+      "learning_rate": 2.553293711519983e-05,
+      "loss": 1.3309,
+      "step": 165000
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 2.551810859223935e-05,
+      "loss": 1.3303,
+      "step": 165100
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 2.550328006927886e-05,
+      "loss": 1.3324,
+      "step": 165200
+    },
+    {
+      "epoch": 5.88,
+      "learning_rate": 2.5488451546318376e-05,
+      "loss": 1.3338,
+      "step": 165300
+    },
+    {
+      "epoch": 5.89,
+      "learning_rate": 2.547362302335789e-05,
+      "loss": 1.3247,
+      "step": 165400
+    },
+    {
+      "epoch": 5.89,
+      "learning_rate": 2.5458794500397403e-05,
+      "loss": 1.3286,
+      "step": 165500
+    },
+    {
+      "epoch": 5.89,
+      "learning_rate": 2.5443965977436922e-05,
+      "loss": 1.3303,
+      "step": 165600
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 2.5429137454476437e-05,
+      "loss": 1.3325,
+      "step": 165700
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 2.5414308931515953e-05,
+      "loss": 1.3298,
+      "step": 165800
+    },
+    {
+      "epoch": 5.9,
+      "learning_rate": 2.5399480408555464e-05,
+      "loss": 1.3264,
+      "step": 165900
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 2.538465188559498e-05,
+      "loss": 1.3209,
+      "step": 166000
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 2.53698233626345e-05,
+      "loss": 1.3323,
+      "step": 166100
+    },
+    {
+      "epoch": 5.91,
+      "learning_rate": 2.5354994839674014e-05,
+      "loss": 1.325,
+      "step": 166200
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 2.5340166316713526e-05,
+      "loss": 1.3292,
+      "step": 166300
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 2.532533779375304e-05,
+      "loss": 1.325,
+      "step": 166400
+    },
+    {
+      "epoch": 5.93,
+      "learning_rate": 2.5310509270792553e-05,
+      "loss": 1.3289,
+      "step": 166500
+    },
+    {
+      "epoch": 5.93,
+      "learning_rate": 2.5295680747832075e-05,
+      "loss": 1.3288,
+      "step": 166600
+    },
+    {
+      "epoch": 5.93,
+      "learning_rate": 2.5280852224871587e-05,
+      "loss": 1.3311,
+      "step": 166700
+    },
+    {
+      "epoch": 5.94,
+      "learning_rate": 2.5266023701911102e-05,
+      "loss": 1.3238,
+      "step": 166800
+    },
+    {
+      "epoch": 5.94,
+      "learning_rate": 2.5251195178950614e-05,
+      "loss": 1.3244,
+      "step": 166900
+    },
+    {
+      "epoch": 5.94,
+      "learning_rate": 2.523636665599013e-05,
+      "loss": 1.3227,
+      "step": 167000
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 2.5221538133029648e-05,
+      "loss": 1.3237,
+      "step": 167100
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 2.5206709610069163e-05,
+      "loss": 1.3265,
+      "step": 167200
+    },
+    {
+      "epoch": 5.95,
+      "learning_rate": 2.5191881087108675e-05,
+      "loss": 1.3273,
+      "step": 167300
+    },
+    {
+      "epoch": 5.96,
+      "learning_rate": 2.517705256414819e-05,
+      "loss": 1.3249,
+      "step": 167400
+    },
+    {
+      "epoch": 5.96,
+      "learning_rate": 2.5162224041187706e-05,
+      "loss": 1.3199,
+      "step": 167500
+    },
+    {
+      "epoch": 5.96,
+      "learning_rate": 2.5147395518227224e-05,
+      "loss": 1.3194,
+      "step": 167600
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 2.5132566995266736e-05,
+      "loss": 1.325,
+      "step": 167700
+    },
+    {
+      "epoch": 5.97,
+      "learning_rate": 2.511773847230625e-05,
+      "loss": 1.3222,
+      "step": 167800
+    },
+    {
+      "epoch": 5.98,
+      "learning_rate": 2.5102909949345767e-05,
+      "loss": 1.315,
+      "step": 167900
+    },
+    {
+      "epoch": 5.98,
+      "learning_rate": 2.508808142638528e-05,
+      "loss": 1.3229,
+      "step": 168000
+    },
+    {
+      "epoch": 5.98,
+      "eval_loss": 1.2230604887008667,
+      "eval_runtime": 5266.4527,
+      "eval_samples_per_second": 607.053,
+      "eval_steps_per_second": 0.395,
+      "step": 168000
+    },
+    {
+      "epoch": 5.98,
+      "learning_rate": 2.5073252903424797e-05,
+      "loss": 1.3212,
+      "step": 168100
+    },
+    {
+      "epoch": 5.99,
+      "learning_rate": 2.5058424380464313e-05,
+      "loss": 1.3205,
+      "step": 168200
+    },
+    {
+      "epoch": 5.99,
+      "learning_rate": 2.5043595857503828e-05,
+      "loss": 1.3239,
+      "step": 168300
+    },
+    {
+      "epoch": 5.99,
+      "learning_rate": 2.502876733454334e-05,
+      "loss": 1.3216,
+      "step": 168400
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.5013938811582855e-05,
+      "loss": 1.3163,
+      "step": 168500
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.499911028862237e-05,
+      "loss": 1.3194,
+      "step": 168600
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 2.498428176566189e-05,
+      "loss": 1.3145,
+      "step": 168700
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 2.49694532427014e-05,
+      "loss": 1.3186,
+      "step": 168800
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 2.495462471974092e-05,
+      "loss": 1.314,
+      "step": 168900
+    },
+    {
+      "epoch": 6.01,
+      "learning_rate": 2.493979619678043e-05,
+      "loss": 1.3172,
+      "step": 169000
+    },
+    {
+      "epoch": 6.02,
+      "learning_rate": 2.4924967673819947e-05,
+      "loss": 1.3172,
+      "step": 169100
+    },
+    {
+      "epoch": 6.02,
+      "learning_rate": 2.4910139150859462e-05,
+      "loss": 1.318,
+      "step": 169200
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 2.4895310627898977e-05,
+      "loss": 1.3157,
+      "step": 169300
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 2.4880482104938493e-05,
+      "loss": 1.3163,
+      "step": 169400
+    },
+    {
+      "epoch": 6.03,
+      "learning_rate": 2.4865653581978008e-05,
+      "loss": 1.3201,
+      "step": 169500
+    },
+    {
+      "epoch": 6.04,
+      "learning_rate": 2.485082505901752e-05,
+      "loss": 1.3154,
+      "step": 169600
+    },
+    {
+      "epoch": 6.04,
+      "learning_rate": 2.483599653605704e-05,
+      "loss": 1.3131,
+      "step": 169700
+    },
+    {
+      "epoch": 6.04,
+      "learning_rate": 2.482116801309655e-05,
+      "loss": 1.3172,
+      "step": 169800
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 2.480633949013607e-05,
+      "loss": 1.3122,
+      "step": 169900
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 2.479151096717558e-05,
+      "loss": 1.3191,
+      "step": 170000
+    },
+    {
+      "epoch": 6.05,
+      "learning_rate": 2.47766824442151e-05,
+      "loss": 1.3149,
+      "step": 170100
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 2.476185392125461e-05,
+      "loss": 1.3182,
+      "step": 170200
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 2.4747025398294127e-05,
+      "loss": 1.3152,
+      "step": 170300
+    },
+    {
+      "epoch": 6.06,
+      "learning_rate": 2.4732196875333642e-05,
+      "loss": 1.3085,
+      "step": 170400
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 2.4717368352373157e-05,
+      "loss": 1.3178,
+      "step": 170500
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 2.4702539829412673e-05,
+      "loss": 1.3162,
+      "step": 170600
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 2.4687711306452188e-05,
+      "loss": 1.3108,
+      "step": 170700
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 2.4672882783491703e-05,
+      "loss": 1.3183,
+      "step": 170800
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 2.465805426053122e-05,
+      "loss": 1.3152,
+      "step": 170900
+    },
+    {
+      "epoch": 6.09,
+      "learning_rate": 2.4643225737570734e-05,
+      "loss": 1.3154,
+      "step": 171000
+    },
+    {
+      "epoch": 6.09,
+      "learning_rate": 2.462839721461025e-05,
+      "loss": 1.3075,
+      "step": 171100
+    },
+    {
+      "epoch": 6.09,
+      "learning_rate": 2.4613568691649764e-05,
+      "loss": 1.3091,
+      "step": 171200
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 2.4598740168689276e-05,
+      "loss": 1.315,
+      "step": 171300
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 2.4583911645728795e-05,
+      "loss": 1.3132,
+      "step": 171400
+    },
+    {
+      "epoch": 6.1,
+      "learning_rate": 2.4569083122768307e-05,
+      "loss": 1.3146,
+      "step": 171500
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 2.4554254599807825e-05,
+      "loss": 1.3087,
+      "step": 171600
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 2.4539426076847337e-05,
+      "loss": 1.309,
+      "step": 171700
+    },
+    {
+      "epoch": 6.11,
+      "learning_rate": 2.4524597553886856e-05,
+      "loss": 1.3085,
+      "step": 171800
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 2.4509769030926368e-05,
+      "loss": 1.3101,
+      "step": 171900
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 2.4494940507965883e-05,
+      "loss": 1.313,
+      "step": 172000
+    },
+    {
+      "epoch": 6.12,
+      "learning_rate": 2.44801119850054e-05,
+      "loss": 1.3124,
+      "step": 172100
+    },
+    {
+      "epoch": 6.13,
+      "learning_rate": 2.4465283462044914e-05,
+      "loss": 1.3066,
+      "step": 172200
+    },
+    {
+      "epoch": 6.13,
+      "learning_rate": 2.445045493908443e-05,
+      "loss": 1.3072,
+      "step": 172300
+    },
+    {
+      "epoch": 6.14,
+      "learning_rate": 2.4435626416123944e-05,
+      "loss": 1.3081,
+      "step": 172400
+    },
+    {
+      "epoch": 6.14,
+      "learning_rate": 2.442079789316346e-05,
+      "loss": 1.3062,
+      "step": 172500
+    },
+    {
+      "epoch": 6.14,
+      "learning_rate": 2.4405969370202975e-05,
+      "loss": 1.3059,
+      "step": 172600
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 2.439114084724249e-05,
+      "loss": 1.3053,
+      "step": 172700
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 2.4376312324282005e-05,
+      "loss": 1.3128,
+      "step": 172800
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 2.436148380132152e-05,
+      "loss": 1.3061,
+      "step": 172900
+    },
+    {
+      "epoch": 6.16,
+      "learning_rate": 2.4346655278361032e-05,
+      "loss": 1.3082,
+      "step": 173000
+    },
+    {
+      "epoch": 6.16,
+      "learning_rate": 2.433182675540055e-05,
+      "loss": 1.3055,
+      "step": 173100
+    },
+    {
+      "epoch": 6.16,
+      "learning_rate": 2.4316998232440063e-05,
+      "loss": 1.3003,
+      "step": 173200
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 2.4302169709479582e-05,
+      "loss": 1.3058,
+      "step": 173300
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 2.4287341186519094e-05,
+      "loss": 1.3028,
+      "step": 173400
+    },
+    {
+      "epoch": 6.17,
+      "learning_rate": 2.427251266355861e-05,
+      "loss": 1.3054,
+      "step": 173500
+    },
+    {
+      "epoch": 6.18,
+      "learning_rate": 2.4257684140598124e-05,
+      "loss": 1.3067,
+      "step": 173600
+    },
+    {
+      "epoch": 6.18,
+      "learning_rate": 2.424285561763764e-05,
+      "loss": 1.3015,
+      "step": 173700
+    },
+    {
+      "epoch": 6.19,
+      "learning_rate": 2.4228027094677155e-05,
+      "loss": 1.3067,
+      "step": 173800
+    },
+    {
+      "epoch": 6.19,
+      "learning_rate": 2.421319857171667e-05,
+      "loss": 1.3042,
+      "step": 173900
+    },
+    {
+      "epoch": 6.19,
+      "learning_rate": 2.4198370048756185e-05,
+      "loss": 1.305,
+      "step": 174000
+    },
+    {
+      "epoch": 6.2,
+      "learning_rate": 2.41835415257957e-05,
+      "loss": 1.3004,
+      "step": 174100
+    },
+    {
+      "epoch": 6.2,
+      "learning_rate": 2.4168713002835212e-05,
+      "loss": 1.3073,
+      "step": 174200
+    },
+    {
+      "epoch": 6.2,
+      "learning_rate": 2.415388447987473e-05,
+      "loss": 1.3023,
+      "step": 174300
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 2.4139055956914243e-05,
+      "loss": 1.3045,
+      "step": 174400
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 2.4124227433953762e-05,
+      "loss": 1.3022,
+      "step": 174500
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 2.4109398910993274e-05,
+      "loss": 1.3098,
+      "step": 174600
+    },
+    {
+      "epoch": 6.22,
+      "learning_rate": 2.409457038803279e-05,
+      "loss": 1.2989,
+      "step": 174700
+    },
+    {
+      "epoch": 6.22,
+      "learning_rate": 2.4079741865072304e-05,
+      "loss": 1.2992,
+      "step": 174800
+    },
+    {
+      "epoch": 6.22,
+      "learning_rate": 2.406491334211182e-05,
+      "loss": 1.3,
+      "step": 174900
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 2.4050084819151335e-05,
+      "loss": 1.3005,
+      "step": 175000
+    },
+    {
+      "epoch": 6.23,
+      "eval_loss": 1.2012759447097778,
+      "eval_runtime": 5266.8158,
+      "eval_samples_per_second": 607.011,
+      "eval_steps_per_second": 0.395,
+      "step": 175000
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 2.403525629619085e-05,
+      "loss": 1.3033,
+      "step": 175100
+    },
+    {
+      "epoch": 6.24,
+      "learning_rate": 2.4020427773230365e-05,
+      "loss": 1.2997,
+      "step": 175200
+    },
+    {
+      "epoch": 6.24,
+      "learning_rate": 2.400559925026988e-05,
+      "loss": 1.3008,
+      "step": 175300
+    },
+    {
+      "epoch": 6.24,
+      "learning_rate": 2.3990770727309396e-05,
+      "loss": 1.2992,
+      "step": 175400
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 2.397594220434891e-05,
+      "loss": 1.3016,
+      "step": 175500
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 2.3961113681388426e-05,
+      "loss": 1.3031,
+      "step": 175600
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 2.394628515842794e-05,
+      "loss": 1.3023,
+      "step": 175700
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 2.3931456635467457e-05,
+      "loss": 1.2968,
+      "step": 175800
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 2.391662811250697e-05,
+      "loss": 1.2974,
+      "step": 175900
+    },
+    {
+      "epoch": 6.26,
+      "learning_rate": 2.3901799589546488e-05,
+      "loss": 1.2937,
+      "step": 176000
+    },
+    {
+      "epoch": 6.27,
+      "learning_rate": 2.3886971066586e-05,
+      "loss": 1.2998,
+      "step": 176100
+    },
+    {
+      "epoch": 6.27,
+      "learning_rate": 2.3872142543625518e-05,
+      "loss": 1.2975,
+      "step": 176200
+    },
+    {
+      "epoch": 6.27,
+      "learning_rate": 2.385731402066503e-05,
+      "loss": 1.2968,
+      "step": 176300
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 2.3842485497704545e-05,
+      "loss": 1.2966,
+      "step": 176400
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 2.382765697474406e-05,
+      "loss": 1.2927,
+      "step": 176500
+    },
+    {
+      "epoch": 6.28,
+      "learning_rate": 2.3812828451783576e-05,
+      "loss": 1.2953,
+      "step": 176600
+    },
+    {
+      "epoch": 6.29,
+      "learning_rate": 2.379799992882309e-05,
+      "loss": 1.2927,
+      "step": 176700
+    },
+    {
+      "epoch": 6.29,
+      "learning_rate": 2.3783171405862606e-05,
+      "loss": 1.2981,
+      "step": 176800
+    },
+    {
+      "epoch": 6.3,
+      "learning_rate": 2.3768342882902118e-05,
+      "loss": 1.2937,
+      "step": 176900
+    },
+    {
+      "epoch": 6.3,
+      "learning_rate": 2.3753514359941637e-05,
+      "loss": 1.2939,
+      "step": 177000
+    },
+    {
+      "epoch": 6.3,
+      "learning_rate": 2.373868583698115e-05,
+      "loss": 1.2901,
+      "step": 177100
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 2.3723857314020668e-05,
+      "loss": 1.2962,
+      "step": 177200
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 2.370902879106018e-05,
+      "loss": 1.2964,
+      "step": 177300
+    },
+    {
+      "epoch": 6.31,
+      "learning_rate": 2.3694200268099695e-05,
+      "loss": 1.2917,
+      "step": 177400
+    },
+    {
+      "epoch": 6.32,
+      "learning_rate": 2.367937174513921e-05,
+      "loss": 1.2927,
+      "step": 177500
+    },
+    {
+      "epoch": 6.32,
+      "learning_rate": 2.3664543222178725e-05,
+      "loss": 1.2932,
+      "step": 177600
+    },
+    {
+      "epoch": 6.32,
+      "learning_rate": 2.364971469921824e-05,
+      "loss": 1.2884,
+      "step": 177700
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 2.3634886176257756e-05,
+      "loss": 1.2955,
+      "step": 177800
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 2.3620057653297275e-05,
+      "loss": 1.2967,
+      "step": 177900
+    },
+    {
+      "epoch": 6.33,
+      "learning_rate": 2.3605229130336786e-05,
+      "loss": 1.2925,
+      "step": 178000
+    },
+    {
+      "epoch": 6.34,
+      "learning_rate": 2.35904006073763e-05,
+      "loss": 1.2931,
+      "step": 178100
+    },
+    {
+      "epoch": 6.34,
+      "learning_rate": 2.3575572084415817e-05,
+      "loss": 1.2892,
+      "step": 178200
+    },
+    {
+      "epoch": 6.35,
+      "learning_rate": 2.3560743561455332e-05,
+      "loss": 1.2877,
+      "step": 178300
+    },
+    {
+      "epoch": 6.35,
+      "learning_rate": 2.3545915038494848e-05,
+      "loss": 1.2891,
+      "step": 178400
+    },
+    {
+      "epoch": 6.35,
+      "learning_rate": 2.3531086515534363e-05,
+      "loss": 1.2887,
+      "step": 178500
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 2.3516257992573875e-05,
+      "loss": 1.2845,
+      "step": 178600
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 2.3501429469613393e-05,
+      "loss": 1.2909,
+      "step": 178700
+    },
+    {
+      "epoch": 6.36,
+      "learning_rate": 2.3486600946652905e-05,
+      "loss": 1.288,
+      "step": 178800
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 2.3471772423692424e-05,
+      "loss": 1.293,
+      "step": 178900
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 2.3456943900731936e-05,
+      "loss": 1.2886,
+      "step": 179000
+    },
+    {
+      "epoch": 6.37,
+      "learning_rate": 2.344211537777145e-05,
+      "loss": 1.2937,
+      "step": 179100
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 2.3427286854810966e-05,
+      "loss": 1.292,
+      "step": 179200
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 2.341245833185048e-05,
+      "loss": 1.2916,
+      "step": 179300
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 2.3397629808889997e-05,
+      "loss": 1.2883,
+      "step": 179400
+    },
+    {
+      "epoch": 6.39,
+      "learning_rate": 2.3382801285929512e-05,
+      "loss": 1.2859,
+      "step": 179500
+    },
+    {
+      "epoch": 6.39,
+      "learning_rate": 2.3367972762969028e-05,
+      "loss": 1.2873,
+      "step": 179600
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 2.3353144240008543e-05,
+      "loss": 1.2869,
+      "step": 179700
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 2.3338315717048058e-05,
+      "loss": 1.2862,
+      "step": 179800
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 2.3323487194087573e-05,
+      "loss": 1.2856,
+      "step": 179900
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 2.330865867112709e-05,
+      "loss": 1.2867,
+      "step": 180000
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 2.3293830148166604e-05,
+      "loss": 1.2824,
+      "step": 180100
+    },
+    {
+      "epoch": 6.41,
+      "learning_rate": 2.327900162520612e-05,
+      "loss": 1.2836,
+      "step": 180200
+    },
+    {
+      "epoch": 6.42,
+      "learning_rate": 2.326417310224563e-05,
+      "loss": 1.2834,
+      "step": 180300
+    },
+    {
+      "epoch": 6.42,
+      "learning_rate": 2.324934457928515e-05,
+      "loss": 1.2855,
+      "step": 180400
+    },
+    {
+      "epoch": 6.42,
+      "learning_rate": 2.323451605632466e-05,
+      "loss": 1.2856,
+      "step": 180500
+    },
+    {
+      "epoch": 6.43,
+      "learning_rate": 2.321968753336418e-05,
+      "loss": 1.2844,
+      "step": 180600
+    },
+    {
+      "epoch": 6.43,
+      "learning_rate": 2.3204859010403692e-05,
+      "loss": 1.2845,
+      "step": 180700
+    },
+    {
+      "epoch": 6.43,
+      "learning_rate": 2.3190030487443208e-05,
+      "loss": 1.2849,
+      "step": 180800
+    },
+    {
+      "epoch": 6.44,
+      "learning_rate": 2.3175201964482723e-05,
+      "loss": 1.2853,
+      "step": 180900
+    },
+    {
+      "epoch": 6.44,
+      "learning_rate": 2.3160373441522238e-05,
+      "loss": 1.2836,
+      "step": 181000
+    },
+    {
+      "epoch": 6.45,
+      "learning_rate": 2.3145544918561753e-05,
+      "loss": 1.2805,
+      "step": 181100
+    },
+    {
+      "epoch": 6.45,
+      "learning_rate": 2.313071639560127e-05,
+      "loss": 1.2853,
+      "step": 181200
+    },
+    {
+      "epoch": 6.45,
+      "learning_rate": 2.311588787264078e-05,
+      "loss": 1.2846,
+      "step": 181300
+    },
+    {
+      "epoch": 6.46,
+      "learning_rate": 2.31010593496803e-05,
+      "loss": 1.2868,
+      "step": 181400
+    },
+    {
+      "epoch": 6.46,
+      "learning_rate": 2.308623082671981e-05,
+      "loss": 1.2813,
+      "step": 181500
+    },
+    {
+      "epoch": 6.46,
+      "learning_rate": 2.307140230375933e-05,
+      "loss": 1.2833,
+      "step": 181600
+    },
+    {
+      "epoch": 6.47,
+      "learning_rate": 2.305657378079884e-05,
+      "loss": 1.2839,
+      "step": 181700
+    },
+    {
+      "epoch": 6.47,
+      "learning_rate": 2.304174525783836e-05,
+      "loss": 1.2851,
+      "step": 181800
+    },
+    {
+      "epoch": 6.47,
+      "learning_rate": 2.3026916734877872e-05,
+      "loss": 1.2804,
+      "step": 181900
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 2.3012088211917387e-05,
+      "loss": 1.2813,
+      "step": 182000
+    },
+    {
+      "epoch": 6.48,
+      "eval_loss": 1.1860889196395874,
+      "eval_runtime": 5267.0817,
+      "eval_samples_per_second": 606.98,
+      "eval_steps_per_second": 0.395,
+      "step": 182000
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 2.2997259688956903e-05,
+      "loss": 1.2841,
+      "step": 182100
+    },
+    {
+      "epoch": 6.48,
+      "learning_rate": 2.2982431165996418e-05,
+      "loss": 1.2878,
+      "step": 182200
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 2.2967602643035933e-05,
+      "loss": 1.2791,
+      "step": 182300
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 2.295277412007545e-05,
+      "loss": 1.2807,
+      "step": 182400
+    },
+    {
+      "epoch": 6.49,
+      "learning_rate": 2.2937945597114964e-05,
+      "loss": 1.2833,
+      "step": 182500
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 2.292311707415448e-05,
+      "loss": 1.2803,
+      "step": 182600
+    },
+    {
+      "epoch": 6.5,
+      "learning_rate": 2.2908288551193994e-05,
+      "loss": 1.2761,
+      "step": 182700
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 2.289346002823351e-05,
+      "loss": 1.2797,
+      "step": 182800
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 2.2878631505273025e-05,
+      "loss": 1.2803,
+      "step": 182900
+    },
+    {
+      "epoch": 6.51,
+      "learning_rate": 2.2863802982312537e-05,
+      "loss": 1.2855,
+      "step": 183000
+    },
+    {
+      "epoch": 6.52,
+      "learning_rate": 2.2848974459352056e-05,
+      "loss": 1.2768,
+      "step": 183100
+    },
+    {
+      "epoch": 6.52,
+      "learning_rate": 2.2834145936391567e-05,
+      "loss": 1.278,
+      "step": 183200
+    },
+    {
+      "epoch": 6.52,
+      "learning_rate": 2.2819317413431086e-05,
+      "loss": 1.2747,
+      "step": 183300
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 2.2804488890470598e-05,
+      "loss": 1.2807,
+      "step": 183400
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 2.2789660367510113e-05,
+      "loss": 1.2807,
+      "step": 183500
+    },
+    {
+      "epoch": 6.53,
+      "learning_rate": 2.277483184454963e-05,
+      "loss": 1.2776,
+      "step": 183600
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 2.2760003321589144e-05,
+      "loss": 1.2819,
+      "step": 183700
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 2.274517479862866e-05,
+      "loss": 1.2819,
+      "step": 183800
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 2.2730346275668174e-05,
+      "loss": 1.2827,
+      "step": 183900
+    },
+    {
+      "epoch": 6.55,
+      "learning_rate": 2.271551775270769e-05,
+      "loss": 1.2778,
+      "step": 184000
+    },
+    {
+      "epoch": 6.55,
+      "learning_rate": 2.2700689229747205e-05,
+      "loss": 1.2801,
+      "step": 184100
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 2.2685860706786717e-05,
+      "loss": 1.2756,
+      "step": 184200
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 2.2671032183826236e-05,
+      "loss": 1.2737,
+      "step": 184300
+    },
+    {
+      "epoch": 6.56,
+      "learning_rate": 2.2656203660865747e-05,
+      "loss": 1.276,
+      "step": 184400
+    },
+    {
+      "epoch": 6.57,
+      "learning_rate": 2.2641375137905266e-05,
+      "loss": 1.2793,
+      "step": 184500
+    },
+    {
+      "epoch": 6.57,
+      "learning_rate": 2.2626546614944778e-05,
+      "loss": 1.2804,
+      "step": 184600
+    },
+    {
+      "epoch": 6.57,
+      "learning_rate": 2.2611718091984293e-05,
+      "loss": 1.2723,
+      "step": 184700
+    },
+    {
+      "epoch": 6.58,
+      "learning_rate": 2.259688956902381e-05,
+      "loss": 1.2749,
+      "step": 184800
+    },
+    {
+      "epoch": 6.58,
+      "learning_rate": 2.2582061046063324e-05,
+      "loss": 1.274,
+      "step": 184900
+    },
+    {
+      "epoch": 6.58,
+      "learning_rate": 2.2567232523102843e-05,
+      "loss": 1.2763,
+      "step": 185000
+    },
+    {
+      "epoch": 6.59,
+      "learning_rate": 2.2552404000142354e-05,
+      "loss": 1.2756,
+      "step": 185100
+    },
+    {
+      "epoch": 6.59,
+      "learning_rate": 2.253757547718187e-05,
+      "loss": 1.2792,
+      "step": 185200
+    },
+    {
+      "epoch": 6.59,
+      "learning_rate": 2.2522746954221385e-05,
+      "loss": 1.2746,
+      "step": 185300
+    },
+    {
+      "epoch": 6.6,
+      "learning_rate": 2.25079184312609e-05,
+      "loss": 1.274,
+      "step": 185400
+    },
+    {
+      "epoch": 6.6,
+      "learning_rate": 2.2493089908300416e-05,
+      "loss": 1.2751,
+      "step": 185500
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 2.247826138533993e-05,
+      "loss": 1.2739,
+      "step": 185600
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 2.2463432862379446e-05,
+      "loss": 1.2769,
+      "step": 185700
+    },
+    {
+      "epoch": 6.61,
+      "learning_rate": 2.244860433941896e-05,
+      "loss": 1.2763,
+      "step": 185800
+    },
+    {
+      "epoch": 6.62,
+      "learning_rate": 2.2433775816458473e-05,
+      "loss": 1.2711,
+      "step": 185900
+    },
+    {
+      "epoch": 6.62,
+      "learning_rate": 2.2418947293497992e-05,
+      "loss": 1.2722,
+      "step": 186000
+    },
+    {
+      "epoch": 6.62,
+      "learning_rate": 2.2404118770537504e-05,
+      "loss": 1.2724,
+      "step": 186100
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 2.2389290247577023e-05,
+      "loss": 1.2692,
+      "step": 186200
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 2.2374461724616534e-05,
+      "loss": 1.2714,
+      "step": 186300
+    },
+    {
+      "epoch": 6.63,
+      "learning_rate": 2.235963320165605e-05,
+      "loss": 1.2675,
+      "step": 186400
+    },
+    {
+      "epoch": 6.64,
+      "learning_rate": 2.2344804678695565e-05,
+      "loss": 1.2726,
+      "step": 186500
+    },
+    {
+      "epoch": 6.64,
+      "learning_rate": 2.232997615573508e-05,
+      "loss": 1.2715,
+      "step": 186600
+    },
+    {
+      "epoch": 6.64,
+      "learning_rate": 2.2315147632774596e-05,
+      "loss": 1.2685,
+      "step": 186700
+    },
+    {
+      "epoch": 6.65,
+      "learning_rate": 2.230031910981411e-05,
+      "loss": 1.2692,
+      "step": 186800
+    },
+    {
+      "epoch": 6.65,
+      "learning_rate": 2.2285490586853626e-05,
+      "loss": 1.2688,
+      "step": 186900
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 2.227066206389314e-05,
+      "loss": 1.2696,
+      "step": 187000
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 2.2255833540932657e-05,
+      "loss": 1.2709,
+      "step": 187100
+    },
+    {
+      "epoch": 6.66,
+      "learning_rate": 2.2241005017972172e-05,
+      "loss": 1.2692,
+      "step": 187200
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 2.2226176495011687e-05,
+      "loss": 1.2714,
+      "step": 187300
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 2.22113479720512e-05,
+      "loss": 1.2685,
+      "step": 187400
+    },
+    {
+      "epoch": 6.67,
+      "learning_rate": 2.2196519449090718e-05,
+      "loss": 1.2756,
+      "step": 187500
+    },
+    {
+      "epoch": 6.68,
+      "learning_rate": 2.218169092613023e-05,
+      "loss": 1.2669,
+      "step": 187600
+    },
+    {
+      "epoch": 6.68,
+      "learning_rate": 2.216686240316975e-05,
+      "loss": 1.2698,
+      "step": 187700
+    },
+    {
+      "epoch": 6.68,
+      "learning_rate": 2.215203388020926e-05,
+      "loss": 1.2722,
+      "step": 187800
+    },
+    {
+      "epoch": 6.69,
+      "learning_rate": 2.213720535724878e-05,
+      "loss": 1.2699,
+      "step": 187900
+    },
+    {
+      "epoch": 6.69,
+      "learning_rate": 2.212237683428829e-05,
+      "loss": 1.2684,
+      "step": 188000
+    },
+    {
+      "epoch": 6.69,
+      "learning_rate": 2.2107548311327806e-05,
+      "loss": 1.2669,
+      "step": 188100
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 2.209271978836732e-05,
+      "loss": 1.2695,
+      "step": 188200
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 2.2077891265406837e-05,
+      "loss": 1.2637,
+      "step": 188300
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 2.2063062742446352e-05,
+      "loss": 1.2707,
+      "step": 188400
+    },
+    {
+      "epoch": 6.71,
+      "learning_rate": 2.2048234219485867e-05,
+      "loss": 1.2682,
+      "step": 188500
+    },
+    {
+      "epoch": 6.71,
+      "learning_rate": 2.203340569652538e-05,
+      "loss": 1.2688,
+      "step": 188600
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 2.2018577173564898e-05,
+      "loss": 1.272,
+      "step": 188700
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 2.200374865060441e-05,
+      "loss": 1.2665,
+      "step": 188800
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 2.198892012764393e-05,
+      "loss": 1.2698,
+      "step": 188900
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 2.197409160468344e-05,
+      "loss": 1.2679,
+      "step": 189000
+    },
+    {
+      "epoch": 6.73,
+      "eval_loss": 1.170892357826233,
+      "eval_runtime": 5266.3993,
+      "eval_samples_per_second": 607.059,
+      "eval_steps_per_second": 0.395,
+      "step": 189000
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 2.1959263081722956e-05,
+      "loss": 1.2677,
+      "step": 189100
+    },
+    {
+      "epoch": 6.73,
+      "learning_rate": 2.194443455876247e-05,
+      "loss": 1.2689,
+      "step": 189200
+    },
+    {
+      "epoch": 6.74,
+      "learning_rate": 2.1929606035801986e-05,
+      "loss": 1.265,
+      "step": 189300
+    },
+    {
+      "epoch": 6.74,
+      "learning_rate": 2.19147775128415e-05,
+      "loss": 1.2661,
+      "step": 189400
+    },
+    {
+      "epoch": 6.74,
+      "learning_rate": 2.1899948989881017e-05,
+      "loss": 1.2663,
+      "step": 189500
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 2.1885120466920532e-05,
+      "loss": 1.2661,
+      "step": 189600
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 2.1870291943960047e-05,
+      "loss": 1.2632,
+      "step": 189700
+    },
+    {
+      "epoch": 6.75,
+      "learning_rate": 2.1855463420999562e-05,
+      "loss": 1.2658,
+      "step": 189800
+    },
+    {
+      "epoch": 6.76,
+      "learning_rate": 2.1840634898039078e-05,
+      "loss": 1.2675,
+      "step": 189900
+    },
+    {
+      "epoch": 6.76,
+      "learning_rate": 2.1825806375078593e-05,
+      "loss": 1.2598,
+      "step": 190000
+    },
+    {
+      "epoch": 6.77,
+      "learning_rate": 2.181097785211811e-05,
+      "loss": 1.267,
+      "step": 190100
+    },
+    {
+      "epoch": 6.77,
+      "learning_rate": 2.1796149329157624e-05,
+      "loss": 1.2648,
+      "step": 190200
+    },
+    {
+      "epoch": 6.77,
+      "learning_rate": 2.1781320806197136e-05,
+      "loss": 1.2633,
+      "step": 190300
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 2.1766492283236654e-05,
+      "loss": 1.2623,
+      "step": 190400
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 2.1751663760276166e-05,
+      "loss": 1.2598,
+      "step": 190500
+    },
+    {
+      "epoch": 6.78,
+      "learning_rate": 2.1736835237315685e-05,
+      "loss": 1.2626,
+      "step": 190600
+    },
+    {
+      "epoch": 6.79,
+      "learning_rate": 2.1722006714355197e-05,
+      "loss": 1.262,
+      "step": 190700
+    },
+    {
+      "epoch": 6.79,
+      "learning_rate": 2.1707178191394712e-05,
+      "loss": 1.2679,
+      "step": 190800
+    },
+    {
+      "epoch": 6.79,
+      "learning_rate": 2.1692349668434227e-05,
+      "loss": 1.2631,
+      "step": 190900
+    },
+    {
+      "epoch": 6.8,
+      "learning_rate": 2.1677521145473742e-05,
+      "loss": 1.2628,
+      "step": 191000
+    },
+    {
+      "epoch": 6.8,
+      "learning_rate": 2.1662692622513258e-05,
+      "loss": 1.2652,
+      "step": 191100
+    },
+    {
+      "epoch": 6.8,
+      "learning_rate": 2.1647864099552773e-05,
+      "loss": 1.2665,
+      "step": 191200
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 2.1633035576592285e-05,
+      "loss": 1.2591,
+      "step": 191300
+    },
+    {
+      "epoch": 6.81,
+      "learning_rate": 2.1618207053631804e-05,
+      "loss": 1.2633,
+      "step": 191400
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 2.1603378530671315e-05,
+      "loss": 1.2594,
+      "step": 191500
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 2.1588550007710834e-05,
+      "loss": 1.2628,
+      "step": 191600
+    },
+    {
+      "epoch": 6.82,
+      "learning_rate": 2.1573721484750346e-05,
+      "loss": 1.2587,
+      "step": 191700
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 2.1558892961789865e-05,
+      "loss": 1.2597,
+      "step": 191800
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 2.1544064438829377e-05,
+      "loss": 1.2604,
+      "step": 191900
+    },
+    {
+      "epoch": 6.83,
+      "learning_rate": 2.1529235915868892e-05,
+      "loss": 1.2609,
+      "step": 192000
+    },
+    {
+      "epoch": 6.84,
+      "learning_rate": 2.151440739290841e-05,
+      "loss": 1.2617,
+      "step": 192100
+    },
+    {
+      "epoch": 6.84,
+      "learning_rate": 2.1499578869947922e-05,
+      "loss": 1.2587,
+      "step": 192200
+    },
+    {
+      "epoch": 6.84,
+      "learning_rate": 2.148475034698744e-05,
+      "loss": 1.2614,
+      "step": 192300
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 2.1469921824026953e-05,
+      "loss": 1.2599,
+      "step": 192400
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 2.145509330106647e-05,
+      "loss": 1.2644,
+      "step": 192500
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 2.1440264778105984e-05,
+      "loss": 1.2592,
+      "step": 192600
+    },
+    {
+      "epoch": 6.86,
+      "learning_rate": 2.14254362551455e-05,
+      "loss": 1.2598,
+      "step": 192700
+    },
+    {
+      "epoch": 6.86,
+      "learning_rate": 2.1410607732185014e-05,
+      "loss": 1.2573,
+      "step": 192800
+    },
+    {
+      "epoch": 6.87,
+      "learning_rate": 2.139577920922453e-05,
+      "loss": 1.2585,
+      "step": 192900
+    },
+    {
+      "epoch": 6.87,
+      "learning_rate": 2.138095068626404e-05,
+      "loss": 1.2591,
+      "step": 193000
+    },
+    {
+      "epoch": 6.87,
+      "learning_rate": 2.136612216330356e-05,
+      "loss": 1.2566,
+      "step": 193100
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 2.1351293640343072e-05,
+      "loss": 1.2579,
+      "step": 193200
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 2.133646511738259e-05,
+      "loss": 1.2511,
+      "step": 193300
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 2.1321636594422102e-05,
+      "loss": 1.2584,
+      "step": 193400
+    },
+    {
+      "epoch": 6.89,
+      "learning_rate": 2.1306808071461618e-05,
+      "loss": 1.2591,
+      "step": 193500
+    },
+    {
+      "epoch": 6.89,
+      "learning_rate": 2.1291979548501133e-05,
+      "loss": 1.2577,
+      "step": 193600
+    },
+    {
+      "epoch": 6.89,
+      "learning_rate": 2.1277151025540648e-05,
+      "loss": 1.2577,
+      "step": 193700
+    },
+    {
+      "epoch": 6.9,
+      "learning_rate": 2.1262322502580164e-05,
+      "loss": 1.2569,
+      "step": 193800
+    },
+    {
+      "epoch": 6.9,
+      "learning_rate": 2.124749397961968e-05,
+      "loss": 1.2588,
+      "step": 193900
+    },
+    {
+      "epoch": 6.9,
+      "learning_rate": 2.1232665456659194e-05,
+      "loss": 1.255,
+      "step": 194000
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 2.121783693369871e-05,
+      "loss": 1.2521,
+      "step": 194100
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 2.1203008410738225e-05,
+      "loss": 1.2516,
+      "step": 194200
+    },
+    {
+      "epoch": 6.91,
+      "learning_rate": 2.118817988777774e-05,
+      "loss": 1.2533,
+      "step": 194300
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 2.1173351364817255e-05,
+      "loss": 1.253,
+      "step": 194400
+    },
+    {
+      "epoch": 6.92,
+      "learning_rate": 2.115852284185677e-05,
+      "loss": 1.2543,
+      "step": 194500
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 2.1143694318896286e-05,
+      "loss": 1.2535,
+      "step": 194600
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 2.1128865795935798e-05,
+      "loss": 1.2546,
+      "step": 194700
+    },
+    {
+      "epoch": 6.93,
+      "learning_rate": 2.1114037272975316e-05,
+      "loss": 1.2518,
+      "step": 194800
+    },
+    {
+      "epoch": 6.94,
+      "learning_rate": 2.1099208750014828e-05,
+      "loss": 1.256,
+      "step": 194900
+    },
+    {
+      "epoch": 6.94,
+      "learning_rate": 2.1084380227054347e-05,
+      "loss": 1.2514,
+      "step": 195000
+    },
+    {
+      "epoch": 6.94,
+      "learning_rate": 2.106955170409386e-05,
+      "loss": 1.2542,
+      "step": 195100
+    },
+    {
+      "epoch": 6.95,
+      "learning_rate": 2.1054723181133374e-05,
+      "loss": 1.2504,
+      "step": 195200
+    },
+    {
+      "epoch": 6.95,
+      "learning_rate": 2.103989465817289e-05,
+      "loss": 1.2593,
+      "step": 195300
+    },
+    {
+      "epoch": 6.95,
+      "learning_rate": 2.1025066135212405e-05,
+      "loss": 1.2587,
+      "step": 195400
+    },
+    {
+      "epoch": 6.96,
+      "learning_rate": 2.101023761225192e-05,
+      "loss": 1.2552,
+      "step": 195500
+    },
+    {
+      "epoch": 6.96,
+      "learning_rate": 2.0995409089291435e-05,
+      "loss": 1.2563,
+      "step": 195600
+    },
+    {
+      "epoch": 6.96,
+      "learning_rate": 2.098058056633095e-05,
+      "loss": 1.256,
+      "step": 195700
+    },
+    {
+      "epoch": 6.97,
+      "learning_rate": 2.0965752043370466e-05,
+      "loss": 1.2545,
+      "step": 195800
+    },
+    {
+      "epoch": 6.97,
+      "learning_rate": 2.0950923520409978e-05,
+      "loss": 1.2573,
+      "step": 195900
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 2.0936094997449496e-05,
+      "loss": 1.2489,
+      "step": 196000
+    },
+    {
+      "epoch": 6.98,
+      "eval_loss": 1.157165765762329,
+      "eval_runtime": 5266.6739,
+      "eval_samples_per_second": 607.027,
+      "eval_steps_per_second": 0.395,
+      "step": 196000
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 2.0921266474489008e-05,
+      "loss": 1.2478,
+      "step": 196100
+    },
+    {
+      "epoch": 6.98,
+      "learning_rate": 2.0906437951528527e-05,
+      "loss": 1.2517,
+      "step": 196200
+    },
+    {
+      "epoch": 6.99,
+      "learning_rate": 2.089160942856804e-05,
+      "loss": 1.2536,
+      "step": 196300
+    },
+    {
+      "epoch": 6.99,
+      "learning_rate": 2.0876780905607554e-05,
+      "loss": 1.253,
+      "step": 196400
+    },
+    {
+      "epoch": 6.99,
+      "learning_rate": 2.086195238264707e-05,
+      "loss": 1.2536,
+      "step": 196500
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.0847123859686585e-05,
+      "loss": 1.2563,
+      "step": 196600
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.08322953367261e-05,
+      "loss": 1.2538,
+      "step": 196700
+    },
+    {
+      "epoch": 7.0,
+      "learning_rate": 2.0817466813765615e-05,
+      "loss": 1.2497,
+      "step": 196800
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 2.080263829080513e-05,
+      "loss": 1.2493,
+      "step": 196900
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 2.0787809767844646e-05,
+      "loss": 1.247,
+      "step": 197000
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 2.077298124488416e-05,
+      "loss": 1.2537,
+      "step": 197100
+    },
+    {
+      "epoch": 7.02,
+      "learning_rate": 2.0758152721923676e-05,
+      "loss": 1.2545,
+      "step": 197200
+    },
+    {
+      "epoch": 7.02,
+      "learning_rate": 2.074332419896319e-05,
+      "loss": 1.2509,
+      "step": 197300
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 2.0728495676002704e-05,
+      "loss": 1.2503,
+      "step": 197400
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 2.0713667153042222e-05,
+      "loss": 1.2514,
+      "step": 197500
+    },
+    {
+      "epoch": 7.03,
+      "learning_rate": 2.0698838630081734e-05,
+      "loss": 1.244,
+      "step": 197600
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 2.0684010107121253e-05,
+      "loss": 1.2541,
+      "step": 197700
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 2.0669181584160765e-05,
+      "loss": 1.2527,
+      "step": 197800
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 2.0654353061200283e-05,
+      "loss": 1.2455,
+      "step": 197900
+    },
+    {
+      "epoch": 7.05,
+      "learning_rate": 2.0639524538239795e-05,
+      "loss": 1.249,
+      "step": 198000
+    },
+    {
+      "epoch": 7.05,
+      "learning_rate": 2.062469601527931e-05,
+      "loss": 1.2475,
+      "step": 198100
+    },
+    {
+      "epoch": 7.05,
+      "learning_rate": 2.0609867492318826e-05,
+      "loss": 1.245,
+      "step": 198200
+    },
+    {
+      "epoch": 7.06,
+      "learning_rate": 2.059503896935834e-05,
+      "loss": 1.2526,
+      "step": 198300
+    },
+    {
+      "epoch": 7.06,
+      "learning_rate": 2.0580210446397856e-05,
+      "loss": 1.2445,
+      "step": 198400
+    },
+    {
+      "epoch": 7.06,
+      "learning_rate": 2.056538192343737e-05,
+      "loss": 1.2496,
+      "step": 198500
+    },
+    {
+      "epoch": 7.07,
+      "learning_rate": 2.0550553400476884e-05,
+      "loss": 1.2455,
+      "step": 198600
+    },
+    {
+      "epoch": 7.07,
+      "learning_rate": 2.0535724877516402e-05,
+      "loss": 1.2488,
+      "step": 198700
+    },
+    {
+      "epoch": 7.07,
+      "learning_rate": 2.0520896354555914e-05,
+      "loss": 1.2534,
+      "step": 198800
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 2.0506067831595433e-05,
+      "loss": 1.2483,
+      "step": 198900
+    },
+    {
+      "epoch": 7.08,
+      "learning_rate": 2.0491239308634945e-05,
+      "loss": 1.2498,
+      "step": 199000
+    },
+    {
+      "epoch": 7.09,
+      "learning_rate": 2.047641078567446e-05,
+      "loss": 1.2453,
+      "step": 199100
+    },
+    {
+      "epoch": 7.09,
+      "learning_rate": 2.046158226271398e-05,
+      "loss": 1.2446,
+      "step": 199200
+    },
+    {
+      "epoch": 7.09,
+      "learning_rate": 2.044675373975349e-05,
+      "loss": 1.2473,
+      "step": 199300
+    },
+    {
+      "epoch": 7.1,
+      "learning_rate": 2.043192521679301e-05,
+      "loss": 1.2422,
+      "step": 199400
+    },
+    {
+      "epoch": 7.1,
+      "learning_rate": 2.041709669383252e-05,
+      "loss": 1.2472,
+      "step": 199500
+    },
+    {
+      "epoch": 7.1,
+      "learning_rate": 2.040226817087204e-05,
+      "loss": 1.2426,
+      "step": 199600
+    },
+    {
+      "epoch": 7.11,
+      "learning_rate": 2.038743964791155e-05,
+      "loss": 1.2448,
+      "step": 199700
+    },
+    {
+      "epoch": 7.11,
+      "learning_rate": 2.0372611124951067e-05,
+      "loss": 1.2432,
+      "step": 199800
+    },
+    {
+      "epoch": 7.11,
+      "learning_rate": 2.0357782601990582e-05,
+      "loss": 1.243,
+      "step": 199900
+    },
+    {
+      "epoch": 7.12,
+      "learning_rate": 2.0342954079030097e-05,
+      "loss": 1.2458,
+      "step": 200000
+    },
+    {
+      "epoch": 7.12,
+      "learning_rate": 2.0328125556069613e-05,
+      "loss": 1.2455,
+      "step": 200100
+    },
+    {
+      "epoch": 7.12,
+      "learning_rate": 2.0313297033109128e-05,
+      "loss": 1.2436,
+      "step": 200200
+    },
+    {
+      "epoch": 7.13,
+      "learning_rate": 2.029846851014864e-05,
+      "loss": 1.2458,
+      "step": 200300
+    },
+    {
+      "epoch": 7.13,
+      "learning_rate": 2.028363998718816e-05,
+      "loss": 1.2452,
+      "step": 200400
+    },
+    {
+      "epoch": 7.14,
+      "learning_rate": 2.026881146422767e-05,
+      "loss": 1.2451,
+      "step": 200500
+    },
+    {
+      "epoch": 7.14,
+      "learning_rate": 2.025398294126719e-05,
+      "loss": 1.2452,
+      "step": 200600
+    },
+    {
+      "epoch": 7.14,
+      "learning_rate": 2.02391544183067e-05,
+      "loss": 1.247,
+      "step": 200700
+    },
+    {
+      "epoch": 7.15,
+      "learning_rate": 2.0224325895346216e-05,
+      "loss": 1.2424,
+      "step": 200800
+    },
+    {
+      "epoch": 7.15,
+      "learning_rate": 2.020949737238573e-05,
+      "loss": 1.2414,
+      "step": 200900
+    },
+    {
+      "epoch": 7.15,
+      "learning_rate": 2.0194668849425247e-05,
+      "loss": 1.2474,
+      "step": 201000
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 2.0179840326464762e-05,
+      "loss": 1.2442,
+      "step": 201100
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 2.0165011803504277e-05,
+      "loss": 1.2418,
+      "step": 201200
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 2.0150183280543793e-05,
+      "loss": 1.2459,
+      "step": 201300
+    },
+    {
+      "epoch": 7.17,
+      "learning_rate": 2.0135354757583308e-05,
+      "loss": 1.2449,
+      "step": 201400
+    },
+    {
+      "epoch": 7.17,
+      "learning_rate": 2.0120526234622823e-05,
+      "loss": 1.2432,
+      "step": 201500
+    },
+    {
+      "epoch": 7.17,
+      "learning_rate": 2.010569771166234e-05,
+      "loss": 1.2367,
+      "step": 201600
+    },
+    {
+      "epoch": 7.18,
+      "learning_rate": 2.0090869188701854e-05,
+      "loss": 1.2444,
+      "step": 201700
+    },
+    {
+      "epoch": 7.18,
+      "learning_rate": 2.007604066574137e-05,
+      "loss": 1.2402,
+      "step": 201800
+    },
+    {
+      "epoch": 7.19,
+      "learning_rate": 2.0061212142780884e-05,
+      "loss": 1.2437,
+      "step": 201900
+    },
+    {
+      "epoch": 7.19,
+      "learning_rate": 2.0046383619820396e-05,
+      "loss": 1.2435,
+      "step": 202000
+    },
+    {
+      "epoch": 7.19,
+      "learning_rate": 2.0031555096859915e-05,
+      "loss": 1.2444,
+      "step": 202100
+    },
+    {
+      "epoch": 7.2,
+      "learning_rate": 2.0016726573899427e-05,
+      "loss": 1.2407,
+      "step": 202200
+    },
+    {
+      "epoch": 7.2,
+      "learning_rate": 2.0001898050938946e-05,
+      "loss": 1.2428,
+      "step": 202300
+    },
+    {
+      "epoch": 7.2,
+      "learning_rate": 1.9987069527978457e-05,
+      "loss": 1.2359,
+      "step": 202400
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 1.9972241005017973e-05,
+      "loss": 1.2366,
+      "step": 202500
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 1.9957412482057488e-05,
+      "loss": 1.2422,
+      "step": 202600
+    },
+    {
+      "epoch": 7.21,
+      "learning_rate": 1.9942583959097003e-05,
+      "loss": 1.2437,
+      "step": 202700
+    },
+    {
+      "epoch": 7.22,
+      "learning_rate": 1.992775543613652e-05,
+      "loss": 1.2457,
+      "step": 202800
+    },
+    {
+      "epoch": 7.22,
+      "learning_rate": 1.9912926913176034e-05,
+      "loss": 1.2426,
+      "step": 202900
+    },
+    {
+      "epoch": 7.22,
+      "learning_rate": 1.9898098390215546e-05,
+      "loss": 1.2363,
+      "step": 203000
+    },
+    {
+      "epoch": 7.22,
+      "eval_loss": 1.1457105875015259,
+      "eval_runtime": 5266.5683,
+      "eval_samples_per_second": 607.04,
+      "eval_steps_per_second": 0.395,
+      "step": 203000
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 1.9883269867255064e-05,
+      "loss": 1.2382,
+      "step": 203100
+    },
+    {
+      "epoch": 7.23,
+      "learning_rate": 1.9868441344294576e-05,
+      "loss": 1.2411,
+      "step": 203200
+    },
+    {
+      "epoch": 7.24,
+      "learning_rate": 1.9853612821334095e-05,
+      "loss": 1.2397,
+      "step": 203300
+    },
+    {
+      "epoch": 7.24,
+      "learning_rate": 1.9838784298373607e-05,
+      "loss": 1.2392,
+      "step": 203400
+    },
+    {
+      "epoch": 7.24,
+      "learning_rate": 1.9823955775413126e-05,
+      "loss": 1.2401,
+      "step": 203500
+    },
+    {
+      "epoch": 7.25,
+      "learning_rate": 1.9809127252452637e-05,
+      "loss": 1.2336,
+      "step": 203600
+    },
+    {
+      "epoch": 7.25,
+      "learning_rate": 1.9794298729492153e-05,
+      "loss": 1.2402,
+      "step": 203700
+    },
+    {
+      "epoch": 7.25,
+      "learning_rate": 1.9779470206531668e-05,
+      "loss": 1.2372,
+      "step": 203800
+    },
+    {
+      "epoch": 7.26,
+      "learning_rate": 1.9764641683571183e-05,
+      "loss": 1.2382,
+      "step": 203900
+    },
+    {
+      "epoch": 7.26,
+      "learning_rate": 1.97498131606107e-05,
+      "loss": 1.2372,
+      "step": 204000
+    },
+    {
+      "epoch": 7.26,
+      "learning_rate": 1.9734984637650214e-05,
+      "loss": 1.2382,
+      "step": 204100
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 1.972015611468973e-05,
+      "loss": 1.2377,
+      "step": 204200
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 1.9705327591729244e-05,
+      "loss": 1.2402,
+      "step": 204300
+    },
+    {
+      "epoch": 7.27,
+      "learning_rate": 1.969049906876876e-05,
+      "loss": 1.2396,
+      "step": 204400
+    },
+    {
+      "epoch": 7.28,
+      "learning_rate": 1.9675670545808275e-05,
+      "loss": 1.241,
+      "step": 204500
+    },
+    {
+      "epoch": 7.28,
+      "learning_rate": 1.966084202284779e-05,
+      "loss": 1.2342,
+      "step": 204600
+    },
+    {
+      "epoch": 7.28,
+      "learning_rate": 1.9646013499887302e-05,
+      "loss": 1.24,
+      "step": 204700
+    },
+    {
+      "epoch": 7.29,
+      "learning_rate": 1.963118497692682e-05,
+      "loss": 1.2401,
+      "step": 204800
+    },
+    {
+      "epoch": 7.29,
+      "learning_rate": 1.9616356453966333e-05,
+      "loss": 1.2378,
+      "step": 204900
+    },
+    {
+      "epoch": 7.3,
+      "learning_rate": 1.960152793100585e-05,
+      "loss": 1.2361,
+      "step": 205000
+    },
+    {
+      "epoch": 7.3,
+      "learning_rate": 1.9586699408045363e-05,
+      "loss": 1.2343,
+      "step": 205100
+    },
+    {
+      "epoch": 7.3,
+      "learning_rate": 1.957187088508488e-05,
+      "loss": 1.2387,
+      "step": 205200
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 1.9557042362124394e-05,
+      "loss": 1.2374,
+      "step": 205300
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 1.954221383916391e-05,
+      "loss": 1.2341,
+      "step": 205400
+    },
+    {
+      "epoch": 7.31,
+      "learning_rate": 1.9527385316203424e-05,
+      "loss": 1.2394,
+      "step": 205500
+    },
+    {
+      "epoch": 7.32,
+      "learning_rate": 1.951255679324294e-05,
+      "loss": 1.2364,
+      "step": 205600
+    },
+    {
+      "epoch": 7.32,
+      "learning_rate": 1.9497728270282455e-05,
+      "loss": 1.2341,
+      "step": 205700
+    },
+    {
+      "epoch": 7.32,
+      "learning_rate": 1.948289974732197e-05,
+      "loss": 1.2405,
+      "step": 205800
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 1.9468071224361482e-05,
+      "loss": 1.2359,
+      "step": 205900
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 1.9453242701401e-05,
+      "loss": 1.2335,
+      "step": 206000
+    },
+    {
+      "epoch": 7.33,
+      "learning_rate": 1.9438414178440513e-05,
+      "loss": 1.2394,
+      "step": 206100
+    },
+    {
+      "epoch": 7.34,
+      "learning_rate": 1.942358565548003e-05,
+      "loss": 1.2299,
+      "step": 206200
+    },
+    {
+      "epoch": 7.34,
+      "learning_rate": 1.9408757132519547e-05,
+      "loss": 1.2391,
+      "step": 206300
+    },
+    {
+      "epoch": 7.35,
+      "learning_rate": 1.939392860955906e-05,
+      "loss": 1.2398,
+      "step": 206400
+    },
+    {
+      "epoch": 7.35,
+      "learning_rate": 1.9379100086598577e-05,
+      "loss": 1.2369,
+      "step": 206500
+    },
+    {
+      "epoch": 7.35,
+      "learning_rate": 1.936427156363809e-05,
+      "loss": 1.2337,
+      "step": 206600
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 1.9349443040677608e-05,
+      "loss": 1.2322,
+      "step": 206700
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 1.933461451771712e-05,
+      "loss": 1.233,
+      "step": 206800
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 1.9319785994756635e-05,
+      "loss": 1.2543,
+      "step": 206900
+    },
+    {
+      "epoch": 7.37,
+      "learning_rate": 1.930495747179615e-05,
+      "loss": 1.2308,
+      "step": 207000
+    },
+    {
+      "epoch": 7.37,
+      "learning_rate": 1.9290128948835666e-05,
+      "loss": 1.2373,
+      "step": 207100
+    },
+    {
+      "epoch": 7.37,
+      "learning_rate": 1.927530042587518e-05,
+      "loss": 1.2342,
+      "step": 207200
+    },
+    {
+      "epoch": 7.38,
+      "learning_rate": 1.9260471902914696e-05,
+      "loss": 1.2328,
+      "step": 207300
+    },
+    {
+      "epoch": 7.38,
+      "learning_rate": 1.9245643379954208e-05,
+      "loss": 1.2339,
+      "step": 207400
+    },
+    {
+      "epoch": 7.38,
+      "learning_rate": 1.9230814856993727e-05,
+      "loss": 1.231,
+      "step": 207500
+    },
+    {
+      "epoch": 7.39,
+      "learning_rate": 1.921598633403324e-05,
+      "loss": 1.2304,
+      "step": 207600
+    },
+    {
+      "epoch": 7.39,
+      "learning_rate": 1.9201157811072757e-05,
+      "loss": 1.2385,
+      "step": 207700
+    },
+    {
+      "epoch": 7.4,
+      "learning_rate": 1.918632928811227e-05,
+      "loss": 1.2326,
+      "step": 207800
+    },
+    {
+      "epoch": 7.4,
+      "learning_rate": 1.9171500765151788e-05,
+      "loss": 1.232,
+      "step": 207900
+    },
+    {
+      "epoch": 7.4,
+      "learning_rate": 1.91566722421913e-05,
+      "loss": 1.2331,
+      "step": 208000
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 1.9141843719230815e-05,
+      "loss": 1.2306,
+      "step": 208100
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 1.912701519627033e-05,
+      "loss": 1.233,
+      "step": 208200
+    },
+    {
+      "epoch": 7.41,
+      "learning_rate": 1.9112186673309845e-05,
+      "loss": 1.2325,
+      "step": 208300
+    },
+    {
+      "epoch": 7.42,
+      "learning_rate": 1.909735815034936e-05,
+      "loss": 1.2358,
+      "step": 208400
+    },
+    {
+      "epoch": 7.42,
+      "learning_rate": 1.9082529627388876e-05,
+      "loss": 1.2335,
+      "step": 208500
+    },
+    {
+      "epoch": 7.42,
+      "learning_rate": 1.906770110442839e-05,
+      "loss": 1.2319,
+      "step": 208600
+    },
+    {
+      "epoch": 7.43,
+      "learning_rate": 1.9052872581467907e-05,
+      "loss": 1.2331,
+      "step": 208700
+    },
+    {
+      "epoch": 7.43,
+      "learning_rate": 1.9038044058507422e-05,
+      "loss": 1.2252,
+      "step": 208800
+    },
+    {
+      "epoch": 7.43,
+      "learning_rate": 1.9023215535546937e-05,
+      "loss": 1.226,
+      "step": 208900
+    },
+    {
+      "epoch": 7.44,
+      "learning_rate": 1.9008387012586452e-05,
+      "loss": 1.231,
+      "step": 209000
+    },
+    {
+      "epoch": 7.44,
+      "learning_rate": 1.8993558489625964e-05,
+      "loss": 1.2307,
+      "step": 209100
+    },
+    {
+      "epoch": 7.45,
+      "learning_rate": 1.8978729966665483e-05,
+      "loss": 1.2359,
+      "step": 209200
+    },
+    {
+      "epoch": 7.45,
+      "learning_rate": 1.8963901443704995e-05,
+      "loss": 1.2324,
+      "step": 209300
+    },
+    {
+      "epoch": 7.45,
+      "learning_rate": 1.8949072920744514e-05,
+      "loss": 1.2297,
+      "step": 209400
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 1.8934244397784025e-05,
+      "loss": 1.2305,
+      "step": 209500
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 1.8919415874823544e-05,
+      "loss": 1.2325,
+      "step": 209600
+    },
+    {
+      "epoch": 7.46,
+      "learning_rate": 1.8904587351863056e-05,
+      "loss": 1.2364,
+      "step": 209700
+    },
+    {
+      "epoch": 7.47,
+      "learning_rate": 1.888975882890257e-05,
+      "loss": 1.2298,
+      "step": 209800
+    },
+    {
+      "epoch": 7.47,
+      "learning_rate": 1.8874930305942087e-05,
+      "loss": 1.2278,
+      "step": 209900
+    },
+    {
+      "epoch": 7.47,
+      "learning_rate": 1.8860101782981602e-05,
+      "loss": 1.2278,
+      "step": 210000
+    },
+    {
+      "epoch": 7.47,
+      "eval_loss": 1.1346683502197266,
+      "eval_runtime": 5266.1777,
+      "eval_samples_per_second": 607.085,
+      "eval_steps_per_second": 0.395,
+      "step": 210000
+    }
+  ],
+  "max_steps": 337188,
+  "num_train_epochs": 12,
+  "total_flos": 5.933940663465429e+20,
+  "trial_name": null,
+  "trial_params": null
+}