diff --git "a/trainer_state.json" "b/trainer_state.json"
new file mode 100644--- /dev/null
+++ "b/trainer_state.json"
@@ -0,0 +1,4742 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.0,
+  "eval_steps": 200,
+  "global_step": 668,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0029940119760479044,
+      "grad_norm": 3.259950093725545,
+      "learning_rate": 9.999944704978835e-06,
+      "loss": 0.6837,
+      "step": 1
+    },
+    {
+      "epoch": 0.005988023952095809,
+      "grad_norm": 2.1193780216166265,
+      "learning_rate": 9.999778821138357e-06,
+      "loss": 0.6327,
+      "step": 2
+    },
+    {
+      "epoch": 0.008982035928143712,
+      "grad_norm": 1.7844829129809667,
+      "learning_rate": 9.999502352147583e-06,
+      "loss": 0.6227,
+      "step": 3
+    },
+    {
+      "epoch": 0.011976047904191617,
+      "grad_norm": 1.9352449833407315,
+      "learning_rate": 9.999115304121459e-06,
+      "loss": 0.6527,
+      "step": 4
+    },
+    {
+      "epoch": 0.014970059880239521,
+      "grad_norm": 1.6819173347650456,
+      "learning_rate": 9.998617685620715e-06,
+      "loss": 0.5031,
+      "step": 5
+    },
+    {
+      "epoch": 0.017964071856287425,
+      "grad_norm": 1.7485891282277255,
+      "learning_rate": 9.998009507651683e-06,
+      "loss": 0.5746,
+      "step": 6
+    },
+    {
+      "epoch": 0.020958083832335328,
+      "grad_norm": 1.4703048546330955,
+      "learning_rate": 9.997290783666048e-06,
+      "loss": 0.5183,
+      "step": 7
+    },
+    {
+      "epoch": 0.023952095808383235,
+      "grad_norm": 1.4631906509909376,
+      "learning_rate": 9.996461529560553e-06,
+      "loss": 0.4926,
+      "step": 8
+    },
+    {
+      "epoch": 0.02694610778443114,
+      "grad_norm": 1.6711536923141481,
+      "learning_rate": 9.995521763676645e-06,
+      "loss": 0.5591,
+      "step": 9
+    },
+    {
+      "epoch": 0.029940119760479042,
+      "grad_norm": 1.5673028169158034,
+      "learning_rate": 9.994471506800078e-06,
+      "loss": 0.513,
+      "step": 10
+    },
+    {
+      "epoch": 0.03293413173652695,
+      "grad_norm": 1.394756660121799,
+      "learning_rate": 9.993310782160439e-06,
+      "loss": 0.5497,
+      "step": 11
+    },
+    {
+      "epoch": 0.03592814371257485,
+      "grad_norm": 1.3694721696029546,
+      "learning_rate": 9.992039615430648e-06,
+      "loss": 0.5117,
+      "step": 12
+    },
+    {
+      "epoch": 0.038922155688622756,
+      "grad_norm": 1.2715344506940187,
+      "learning_rate": 9.99065803472638e-06,
+      "loss": 0.4864,
+      "step": 13
+    },
+    {
+      "epoch": 0.041916167664670656,
+      "grad_norm": 1.2769046364437955,
+      "learning_rate": 9.989166070605447e-06,
+      "loss": 0.4715,
+      "step": 14
+    },
+    {
+      "epoch": 0.04491017964071856,
+      "grad_norm": 1.1844958764027513,
+      "learning_rate": 9.98756375606713e-06,
+      "loss": 0.4543,
+      "step": 15
+    },
+    {
+      "epoch": 0.04790419161676647,
+      "grad_norm": 1.3159385727684432,
+      "learning_rate": 9.985851126551428e-06,
+      "loss": 0.494,
+      "step": 16
+    },
+    {
+      "epoch": 0.05089820359281437,
+      "grad_norm": 1.0707726214815312,
+      "learning_rate": 9.9840282199383e-06,
+      "loss": 0.4121,
+      "step": 17
+    },
+    {
+      "epoch": 0.05389221556886228,
+      "grad_norm": 1.0567438546615648,
+      "learning_rate": 9.982095076546806e-06,
+      "loss": 0.4648,
+      "step": 18
+    },
+    {
+      "epoch": 0.05688622754491018,
+      "grad_norm": 1.207486961184982,
+      "learning_rate": 9.980051739134235e-06,
+      "loss": 0.476,
+      "step": 19
+    },
+    {
+      "epoch": 0.059880239520958084,
+      "grad_norm": 1.2091498427752334,
+      "learning_rate": 9.977898252895133e-06,
+      "loss": 0.4831,
+      "step": 20
+    },
+    {
+      "epoch": 0.06287425149700598,
+      "grad_norm": 1.0968158716026501,
+      "learning_rate": 9.975634665460333e-06,
+      "loss": 0.4557,
+      "step": 21
+    },
+    {
+      "epoch": 0.0658682634730539,
+      "grad_norm": 1.1054683684689441,
+      "learning_rate": 9.973261026895878e-06,
+      "loss": 0.4188,
+      "step": 22
+    },
+    {
+      "epoch": 0.0688622754491018,
+      "grad_norm": 1.1174450618308993,
+      "learning_rate": 9.970777389701927e-06,
+      "loss": 0.5169,
+      "step": 23
+    },
+    {
+      "epoch": 0.0718562874251497,
+      "grad_norm": 1.1159228384303175,
+      "learning_rate": 9.968183808811586e-06,
+      "loss": 0.4098,
+      "step": 24
+    },
+    {
+      "epoch": 0.0748502994011976,
+      "grad_norm": 1.2481387970310578,
+      "learning_rate": 9.965480341589702e-06,
+      "loss": 0.4818,
+      "step": 25
+    },
+    {
+      "epoch": 0.07784431137724551,
+      "grad_norm": 1.1115867999551607,
+      "learning_rate": 9.962667047831585e-06,
+      "loss": 0.4839,
+      "step": 26
+    },
+    {
+      "epoch": 0.08083832335329341,
+      "grad_norm": 1.1093981787257718,
+      "learning_rate": 9.95974398976169e-06,
+      "loss": 0.4173,
+      "step": 27
+    },
+    {
+      "epoch": 0.08383233532934131,
+      "grad_norm": 1.1564097343154072,
+      "learning_rate": 9.95671123203224e-06,
+      "loss": 0.4705,
+      "step": 28
+    },
+    {
+      "epoch": 0.08682634730538923,
+      "grad_norm": 1.2506421659125888,
+      "learning_rate": 9.953568841721796e-06,
+      "loss": 0.4981,
+      "step": 29
+    },
+    {
+      "epoch": 0.08982035928143713,
+      "grad_norm": 0.9546754723137117,
+      "learning_rate": 9.950316888333775e-06,
+      "loss": 0.4062,
+      "step": 30
+    },
+    {
+      "epoch": 0.09281437125748503,
+      "grad_norm": 1.0659792816470735,
+      "learning_rate": 9.946955443794908e-06,
+      "loss": 0.4911,
+      "step": 31
+    },
+    {
+      "epoch": 0.09580838323353294,
+      "grad_norm": 1.257533089253244,
+      "learning_rate": 9.943484582453653e-06,
+      "loss": 0.4617,
+      "step": 32
+    },
+    {
+      "epoch": 0.09880239520958084,
+      "grad_norm": 1.007258124748217,
+      "learning_rate": 9.939904381078553e-06,
+      "loss": 0.4096,
+      "step": 33
+    },
+    {
+      "epoch": 0.10179640718562874,
+      "grad_norm": 0.9710165177273038,
+      "learning_rate": 9.93621491885653e-06,
+      "loss": 0.3812,
+      "step": 34
+    },
+    {
+      "epoch": 0.10479041916167664,
+      "grad_norm": 1.150014297638306,
+      "learning_rate": 9.932416277391144e-06,
+      "loss": 0.4371,
+      "step": 35
+    },
+    {
+      "epoch": 0.10778443113772455,
+      "grad_norm": 1.170029440581812,
+      "learning_rate": 9.928508540700775e-06,
+      "loss": 0.45,
+      "step": 36
+    },
+    {
+      "epoch": 0.11077844311377245,
+      "grad_norm": 1.1671295508620236,
+      "learning_rate": 9.924491795216777e-06,
+      "loss": 0.4929,
+      "step": 37
+    },
+    {
+      "epoch": 0.11377245508982035,
+      "grad_norm": 1.1458069242764883,
+      "learning_rate": 9.920366129781564e-06,
+      "loss": 0.4552,
+      "step": 38
+    },
+    {
+      "epoch": 0.11676646706586827,
+      "grad_norm": 0.9675082028492001,
+      "learning_rate": 9.916131635646635e-06,
+      "loss": 0.4063,
+      "step": 39
+    },
+    {
+      "epoch": 0.11976047904191617,
+      "grad_norm": 1.1584269214634861,
+      "learning_rate": 9.91178840647057e-06,
+      "loss": 0.4878,
+      "step": 40
+    },
+    {
+      "epoch": 0.12275449101796407,
+      "grad_norm": 1.134707014243856,
+      "learning_rate": 9.907336538316946e-06,
+      "loss": 0.4251,
+      "step": 41
+    },
+    {
+      "epoch": 0.12574850299401197,
+      "grad_norm": 1.040949641670139,
+      "learning_rate": 9.902776129652223e-06,
+      "loss": 0.3753,
+      "step": 42
+    },
+    {
+      "epoch": 0.12874251497005987,
+      "grad_norm": 1.0767995131714176,
+      "learning_rate": 9.898107281343557e-06,
+      "loss": 0.3864,
+      "step": 43
+    },
+    {
+      "epoch": 0.1317365269461078,
+      "grad_norm": 1.033911280652776,
+      "learning_rate": 9.893330096656576e-06,
+      "loss": 0.3601,
+      "step": 44
+    },
+    {
+      "epoch": 0.1347305389221557,
+      "grad_norm": 1.180765403788847,
+      "learning_rate": 9.888444681253087e-06,
+      "loss": 0.4495,
+      "step": 45
+    },
+    {
+      "epoch": 0.1377245508982036,
+      "grad_norm": 1.1916801308659744,
+      "learning_rate": 9.883451143188753e-06,
+      "loss": 0.4436,
+      "step": 46
+    },
+    {
+      "epoch": 0.1407185628742515,
+      "grad_norm": 1.0368790892305528,
+      "learning_rate": 9.878349592910694e-06,
+      "loss": 0.3669,
+      "step": 47
+    },
+    {
+      "epoch": 0.1437125748502994,
+      "grad_norm": 1.1944608101710639,
+      "learning_rate": 9.873140143255035e-06,
+      "loss": 0.4716,
+      "step": 48
+    },
+    {
+      "epoch": 0.1467065868263473,
+      "grad_norm": 1.10748255044691,
+      "learning_rate": 9.867822909444435e-06,
+      "loss": 0.4625,
+      "step": 49
+    },
+    {
+      "epoch": 0.1497005988023952,
+      "grad_norm": 1.121703122304384,
+      "learning_rate": 9.862398009085511e-06,
+      "loss": 0.4283,
+      "step": 50
+    },
+    {
+      "epoch": 0.15269461077844312,
+      "grad_norm": 1.0990878171571523,
+      "learning_rate": 9.856865562166256e-06,
+      "loss": 0.4111,
+      "step": 51
+    },
+    {
+      "epoch": 0.15568862275449102,
+      "grad_norm": 1.0133021102506585,
+      "learning_rate": 9.851225691053382e-06,
+      "loss": 0.3811,
+      "step": 52
+    },
+    {
+      "epoch": 0.15868263473053892,
+      "grad_norm": 1.1296451828299812,
+      "learning_rate": 9.8454785204896e-06,
+      "loss": 0.4555,
+      "step": 53
+    },
+    {
+      "epoch": 0.16167664670658682,
+      "grad_norm": 0.9753075044500134,
+      "learning_rate": 9.83962417759088e-06,
+      "loss": 0.3884,
+      "step": 54
+    },
+    {
+      "epoch": 0.16467065868263472,
+      "grad_norm": 1.1544755998823522,
+      "learning_rate": 9.833662791843628e-06,
+      "loss": 0.3811,
+      "step": 55
+    },
+    {
+      "epoch": 0.16766467065868262,
+      "grad_norm": 1.0990987577144584,
+      "learning_rate": 9.827594495101824e-06,
+      "loss": 0.4135,
+      "step": 56
+    },
+    {
+      "epoch": 0.17065868263473055,
+      "grad_norm": 1.0956201587644776,
+      "learning_rate": 9.821419421584108e-06,
+      "loss": 0.4202,
+      "step": 57
+    },
+    {
+      "epoch": 0.17365269461077845,
+      "grad_norm": 1.0924915445545975,
+      "learning_rate": 9.815137707870806e-06,
+      "loss": 0.4598,
+      "step": 58
+    },
+    {
+      "epoch": 0.17664670658682635,
+      "grad_norm": 1.1881374676276943,
+      "learning_rate": 9.808749492900917e-06,
+      "loss": 0.4095,
+      "step": 59
+    },
+    {
+      "epoch": 0.17964071856287425,
+      "grad_norm": 1.0572103069418428,
+      "learning_rate": 9.802254917969033e-06,
+      "loss": 0.4331,
+      "step": 60
+    },
+    {
+      "epoch": 0.18263473053892215,
+      "grad_norm": 1.1834700167751402,
+      "learning_rate": 9.795654126722218e-06,
+      "loss": 0.4239,
+      "step": 61
+    },
+    {
+      "epoch": 0.18562874251497005,
+      "grad_norm": 1.07892600736689,
+      "learning_rate": 9.788947265156828e-06,
+      "loss": 0.4052,
+      "step": 62
+    },
+    {
+      "epoch": 0.18862275449101795,
+      "grad_norm": 1.1349211532599017,
+      "learning_rate": 9.782134481615282e-06,
+      "loss": 0.3767,
+      "step": 63
+    },
+    {
+      "epoch": 0.19161676646706588,
+      "grad_norm": 1.0945327464824868,
+      "learning_rate": 9.775215926782788e-06,
+      "loss": 0.4458,
+      "step": 64
+    },
+    {
+      "epoch": 0.19461077844311378,
+      "grad_norm": 1.1057913503723604,
+      "learning_rate": 9.768191753683997e-06,
+      "loss": 0.3984,
+      "step": 65
+    },
+    {
+      "epoch": 0.19760479041916168,
+      "grad_norm": 1.1065303995428613,
+      "learning_rate": 9.761062117679632e-06,
+      "loss": 0.3965,
+      "step": 66
+    },
+    {
+      "epoch": 0.20059880239520958,
+      "grad_norm": 1.0445587902291722,
+      "learning_rate": 9.75382717646304e-06,
+      "loss": 0.3729,
+      "step": 67
+    },
+    {
+      "epoch": 0.20359281437125748,
+      "grad_norm": 1.120239829262365,
+      "learning_rate": 9.746487090056712e-06,
+      "loss": 0.4688,
+      "step": 68
+    },
+    {
+      "epoch": 0.20658682634730538,
+      "grad_norm": 0.9257004678886442,
+      "learning_rate": 9.739042020808746e-06,
+      "loss": 0.3645,
+      "step": 69
+    },
+    {
+      "epoch": 0.20958083832335328,
+      "grad_norm": 1.103169557253983,
+      "learning_rate": 9.73149213338924e-06,
+      "loss": 0.4514,
+      "step": 70
+    },
+    {
+      "epoch": 0.2125748502994012,
+      "grad_norm": 1.0844893436135883,
+      "learning_rate": 9.72383759478667e-06,
+      "loss": 0.4497,
+      "step": 71
+    },
+    {
+      "epoch": 0.2155688622754491,
+      "grad_norm": 1.1395474166795472,
+      "learning_rate": 9.71607857430419e-06,
+      "loss": 0.3537,
+      "step": 72
+    },
+    {
+      "epoch": 0.218562874251497,
+      "grad_norm": 1.1874299280487752,
+      "learning_rate": 9.708215243555875e-06,
+      "loss": 0.4452,
+      "step": 73
+    },
+    {
+      "epoch": 0.2215568862275449,
+      "grad_norm": 1.2696354704333974,
+      "learning_rate": 9.700247776462944e-06,
+      "loss": 0.4223,
+      "step": 74
+    },
+    {
+      "epoch": 0.2245508982035928,
+      "grad_norm": 1.1635085870073603,
+      "learning_rate": 9.6921763492499e-06,
+      "loss": 0.4083,
+      "step": 75
+    },
+    {
+      "epoch": 0.2275449101796407,
+      "grad_norm": 1.028053724760608,
+      "learning_rate": 9.68400114044064e-06,
+      "loss": 0.3323,
+      "step": 76
+    },
+    {
+      "epoch": 0.23053892215568864,
+      "grad_norm": 1.156924602473931,
+      "learning_rate": 9.6757223308545e-06,
+      "loss": 0.4198,
+      "step": 77
+    },
+    {
+      "epoch": 0.23353293413173654,
+      "grad_norm": 1.073712750409912,
+      "learning_rate": 9.667340103602263e-06,
+      "loss": 0.4638,
+      "step": 78
+    },
+    {
+      "epoch": 0.23652694610778444,
+      "grad_norm": 1.0364495880614395,
+      "learning_rate": 9.658854644082099e-06,
+      "loss": 0.3719,
+      "step": 79
+    },
+    {
+      "epoch": 0.23952095808383234,
+      "grad_norm": 1.34892906991759,
+      "learning_rate": 9.650266139975474e-06,
+      "loss": 0.451,
+      "step": 80
+    },
+    {
+      "epoch": 0.24251497005988024,
+      "grad_norm": 1.206344655320952,
+      "learning_rate": 9.641574781242999e-06,
+      "loss": 0.4029,
+      "step": 81
+    },
+    {
+      "epoch": 0.24550898203592814,
+      "grad_norm": 1.135722708996098,
+      "learning_rate": 9.632780760120217e-06,
+      "loss": 0.4125,
+      "step": 82
+    },
+    {
+      "epoch": 0.24850299401197604,
+      "grad_norm": 1.1305769258182499,
+      "learning_rate": 9.62388427111336e-06,
+      "loss": 0.3835,
+      "step": 83
+    },
+    {
+      "epoch": 0.25149700598802394,
+      "grad_norm": 1.03562720561894,
+      "learning_rate": 9.614885510995047e-06,
+      "loss": 0.3457,
+      "step": 84
+    },
+    {
+      "epoch": 0.25449101796407186,
+      "grad_norm": 1.173918313099111,
+      "learning_rate": 9.605784678799934e-06,
+      "loss": 0.3942,
+      "step": 85
+    },
+    {
+      "epoch": 0.25748502994011974,
+      "grad_norm": 1.1048394098762548,
+      "learning_rate": 9.596581975820304e-06,
+      "loss": 0.4048,
+      "step": 86
+    },
+    {
+      "epoch": 0.26047904191616766,
+      "grad_norm": 1.1737197943681497,
+      "learning_rate": 9.587277605601617e-06,
+      "loss": 0.4326,
+      "step": 87
+    },
+    {
+      "epoch": 0.2634730538922156,
+      "grad_norm": 1.1899908765811393,
+      "learning_rate": 9.577871773938013e-06,
+      "loss": 0.4483,
+      "step": 88
+    },
+    {
+      "epoch": 0.26646706586826346,
+      "grad_norm": 1.1925008303368345,
+      "learning_rate": 9.568364688867757e-06,
+      "loss": 0.505,
+      "step": 89
+    },
+    {
+      "epoch": 0.2694610778443114,
+      "grad_norm": 1.1263003488717263,
+      "learning_rate": 9.558756560668637e-06,
+      "loss": 0.3827,
+      "step": 90
+    },
+    {
+      "epoch": 0.27245508982035926,
+      "grad_norm": 1.215099365818016,
+      "learning_rate": 9.549047601853313e-06,
+      "loss": 0.397,
+      "step": 91
+    },
+    {
+      "epoch": 0.2754491017964072,
+      "grad_norm": 1.2355257019028731,
+      "learning_rate": 9.539238027164618e-06,
+      "loss": 0.4289,
+      "step": 92
+    },
+    {
+      "epoch": 0.27844311377245506,
+      "grad_norm": 1.1381564442046626,
+      "learning_rate": 9.52932805357081e-06,
+      "loss": 0.4106,
+      "step": 93
+    },
+    {
+      "epoch": 0.281437125748503,
+      "grad_norm": 1.194984646418279,
+      "learning_rate": 9.519317900260769e-06,
+      "loss": 0.4379,
+      "step": 94
+    },
+    {
+      "epoch": 0.2844311377245509,
+      "grad_norm": 1.1266187084293326,
+      "learning_rate": 9.509207788639148e-06,
+      "loss": 0.4311,
+      "step": 95
+    },
+    {
+      "epoch": 0.2874251497005988,
+      "grad_norm": 1.048077100842723,
+      "learning_rate": 9.498997942321484e-06,
+      "loss": 0.4322,
+      "step": 96
+    },
+    {
+      "epoch": 0.2904191616766467,
+      "grad_norm": 1.0178471047646642,
+      "learning_rate": 9.488688587129243e-06,
+      "loss": 0.4437,
+      "step": 97
+    },
+    {
+      "epoch": 0.2934131736526946,
+      "grad_norm": 1.2549035741975805,
+      "learning_rate": 9.47827995108483e-06,
+      "loss": 0.3885,
+      "step": 98
+    },
+    {
+      "epoch": 0.2964071856287425,
+      "grad_norm": 1.3211241919185797,
+      "learning_rate": 9.467772264406545e-06,
+      "loss": 0.408,
+      "step": 99
+    },
+    {
+      "epoch": 0.2994011976047904,
+      "grad_norm": 0.9818987942004095,
+      "learning_rate": 9.457165759503492e-06,
+      "loss": 0.373,
+      "step": 100
+    },
+    {
+      "epoch": 0.3023952095808383,
+      "grad_norm": 1.1075327027029354,
+      "learning_rate": 9.446460670970436e-06,
+      "loss": 0.3786,
+      "step": 101
+    },
+    {
+      "epoch": 0.30538922155688625,
+      "grad_norm": 1.1276656589404976,
+      "learning_rate": 9.435657235582616e-06,
+      "loss": 0.3274,
+      "step": 102
+    },
+    {
+      "epoch": 0.3083832335329341,
+      "grad_norm": 1.2205331141402274,
+      "learning_rate": 9.424755692290507e-06,
+      "loss": 0.4101,
+      "step": 103
+    },
+    {
+      "epoch": 0.31137724550898205,
+      "grad_norm": 1.2618230806531148,
+      "learning_rate": 9.413756282214538e-06,
+      "loss": 0.4449,
+      "step": 104
+    },
+    {
+      "epoch": 0.3143712574850299,
+      "grad_norm": 1.0520106562946805,
+      "learning_rate": 9.402659248639749e-06,
+      "loss": 0.3933,
+      "step": 105
+    },
+    {
+      "epoch": 0.31736526946107785,
+      "grad_norm": 1.126593413200309,
+      "learning_rate": 9.391464837010428e-06,
+      "loss": 0.3925,
+      "step": 106
+    },
+    {
+      "epoch": 0.3203592814371258,
+      "grad_norm": 1.1840151749474808,
+      "learning_rate": 9.380173294924661e-06,
+      "loss": 0.4483,
+      "step": 107
+    },
+    {
+      "epoch": 0.32335329341317365,
+      "grad_norm": 1.1401220955142177,
+      "learning_rate": 9.368784872128877e-06,
+      "loss": 0.3974,
+      "step": 108
+    },
+    {
+      "epoch": 0.3263473053892216,
+      "grad_norm": 1.1049699516554348,
+      "learning_rate": 9.357299820512305e-06,
+      "loss": 0.3905,
+      "step": 109
+    },
+    {
+      "epoch": 0.32934131736526945,
+      "grad_norm": 1.225559720675576,
+      "learning_rate": 9.345718394101412e-06,
+      "loss": 0.3896,
+      "step": 110
+    },
+    {
+      "epoch": 0.3323353293413174,
+      "grad_norm": 1.1039681914080792,
+      "learning_rate": 9.334040849054288e-06,
+      "loss": 0.4224,
+      "step": 111
+    },
+    {
+      "epoch": 0.33532934131736525,
+      "grad_norm": 1.1554425258899756,
+      "learning_rate": 9.322267443654974e-06,
+      "loss": 0.4006,
+      "step": 112
+    },
+    {
+      "epoch": 0.3383233532934132,
+      "grad_norm": 1.1507807802109296,
+      "learning_rate": 9.310398438307747e-06,
+      "loss": 0.4471,
+      "step": 113
+    },
+    {
+      "epoch": 0.3413173652694611,
+      "grad_norm": 1.0507958470661567,
+      "learning_rate": 9.29843409553137e-06,
+      "loss": 0.3723,
+      "step": 114
+    },
+    {
+      "epoch": 0.344311377245509,
+      "grad_norm": 1.0671404578750951,
+      "learning_rate": 9.286374679953278e-06,
+      "loss": 0.3782,
+      "step": 115
+    },
+    {
+      "epoch": 0.3473053892215569,
+      "grad_norm": 1.1835063144117342,
+      "learning_rate": 9.274220458303727e-06,
+      "loss": 0.3592,
+      "step": 116
+    },
+    {
+      "epoch": 0.3502994011976048,
+      "grad_norm": 1.1145664210443766,
+      "learning_rate": 9.261971699409893e-06,
+      "loss": 0.421,
+      "step": 117
+    },
+    {
+      "epoch": 0.3532934131736527,
+      "grad_norm": 1.135842783324298,
+      "learning_rate": 9.249628674189928e-06,
+      "loss": 0.3906,
+      "step": 118
+    },
+    {
+      "epoch": 0.3562874251497006,
+      "grad_norm": 1.1590128112553568,
+      "learning_rate": 9.237191655646972e-06,
+      "loss": 0.4299,
+      "step": 119
+    },
+    {
+      "epoch": 0.3592814371257485,
+      "grad_norm": 1.2045411874650218,
+      "learning_rate": 9.224660918863104e-06,
+      "loss": 0.464,
+      "step": 120
+    },
+    {
+      "epoch": 0.36227544910179643,
+      "grad_norm": 1.1880907121946593,
+      "learning_rate": 9.212036740993265e-06,
+      "loss": 0.4024,
+      "step": 121
+    },
+    {
+      "epoch": 0.3652694610778443,
+      "grad_norm": 1.1701863478922876,
+      "learning_rate": 9.199319401259132e-06,
+      "loss": 0.4057,
+      "step": 122
+    },
+    {
+      "epoch": 0.36826347305389223,
+      "grad_norm": 1.082825207105692,
+      "learning_rate": 9.186509180942928e-06,
+      "loss": 0.3792,
+      "step": 123
+    },
+    {
+      "epoch": 0.3712574850299401,
+      "grad_norm": 1.0885680784831713,
+      "learning_rate": 9.173606363381218e-06,
+      "loss": 0.4497,
+      "step": 124
+    },
+    {
+      "epoch": 0.37425149700598803,
+      "grad_norm": 1.1505079040028416,
+      "learning_rate": 9.16061123395863e-06,
+      "loss": 0.3994,
+      "step": 125
+    },
+    {
+      "epoch": 0.3772455089820359,
+      "grad_norm": 1.1835704868079155,
+      "learning_rate": 9.147524080101543e-06,
+      "loss": 0.3975,
+      "step": 126
+    },
+    {
+      "epoch": 0.38023952095808383,
+      "grad_norm": 1.2354339997293993,
+      "learning_rate": 9.134345191271742e-06,
+      "loss": 0.4629,
+      "step": 127
+    },
+    {
+      "epoch": 0.38323353293413176,
+      "grad_norm": 1.1432081506581395,
+      "learning_rate": 9.121074858959997e-06,
+      "loss": 0.4332,
+      "step": 128
+    },
+    {
+      "epoch": 0.38622754491017963,
+      "grad_norm": 1.0795243542728965,
+      "learning_rate": 9.107713376679634e-06,
+      "loss": 0.4078,
+      "step": 129
+    },
+    {
+      "epoch": 0.38922155688622756,
+      "grad_norm": 1.2840882370761184,
+      "learning_rate": 9.094261039960028e-06,
+      "loss": 0.4147,
+      "step": 130
+    },
+    {
+      "epoch": 0.39221556886227543,
+      "grad_norm": 1.2931824190048564,
+      "learning_rate": 9.08071814634008e-06,
+      "loss": 0.4684,
+      "step": 131
+    },
+    {
+      "epoch": 0.39520958083832336,
+      "grad_norm": 1.2746240706900507,
+      "learning_rate": 9.067084995361623e-06,
+      "loss": 0.4157,
+      "step": 132
+    },
+    {
+      "epoch": 0.39820359281437123,
+      "grad_norm": 1.163872903602833,
+      "learning_rate": 9.053361888562807e-06,
+      "loss": 0.4516,
+      "step": 133
+    },
+    {
+      "epoch": 0.40119760479041916,
+      "grad_norm": 1.1928684684875797,
+      "learning_rate": 9.039549129471423e-06,
+      "loss": 0.3875,
+      "step": 134
+    },
+    {
+      "epoch": 0.4041916167664671,
+      "grad_norm": 1.0081150850935165,
+      "learning_rate": 9.025647023598196e-06,
+      "loss": 0.3676,
+      "step": 135
+    },
+    {
+      "epoch": 0.40718562874251496,
+      "grad_norm": 1.105084183608223,
+      "learning_rate": 9.011655878430018e-06,
+      "loss": 0.3701,
+      "step": 136
+    },
+    {
+      "epoch": 0.4101796407185629,
+      "grad_norm": 1.2546267562705486,
+      "learning_rate": 8.99757600342316e-06,
+      "loss": 0.4444,
+      "step": 137
+    },
+    {
+      "epoch": 0.41317365269461076,
+      "grad_norm": 1.2107710311478224,
+      "learning_rate": 8.983407709996415e-06,
+      "loss": 0.4204,
+      "step": 138
+    },
+    {
+      "epoch": 0.4161676646706587,
+      "grad_norm": 1.2054823473820004,
+      "learning_rate": 8.969151311524215e-06,
+      "loss": 0.3943,
+      "step": 139
+    },
+    {
+      "epoch": 0.41916167664670656,
+      "grad_norm": 1.2948158699767098,
+      "learning_rate": 8.954807123329703e-06,
+      "loss": 0.4264,
+      "step": 140
+    },
+    {
+      "epoch": 0.4221556886227545,
+      "grad_norm": 1.0313331260539953,
+      "learning_rate": 8.940375462677758e-06,
+      "loss": 0.3468,
+      "step": 141
+    },
+    {
+      "epoch": 0.4251497005988024,
+      "grad_norm": 1.1686876294846669,
+      "learning_rate": 8.92585664876797e-06,
+      "loss": 0.3931,
+      "step": 142
+    },
+    {
+      "epoch": 0.4281437125748503,
+      "grad_norm": 1.0452432248470267,
+      "learning_rate": 8.911251002727588e-06,
+      "loss": 0.3844,
+      "step": 143
+    },
+    {
+      "epoch": 0.4311377245508982,
+      "grad_norm": 1.0505109002932773,
+      "learning_rate": 8.896558847604414e-06,
+      "loss": 0.3659,
+      "step": 144
+    },
+    {
+      "epoch": 0.4341317365269461,
+      "grad_norm": 1.129331665998729,
+      "learning_rate": 8.881780508359661e-06,
+      "loss": 0.4801,
+      "step": 145
+    },
+    {
+      "epoch": 0.437125748502994,
+      "grad_norm": 1.393398520346271,
+      "learning_rate": 8.86691631186076e-06,
+      "loss": 0.4036,
+      "step": 146
+    },
+    {
+      "epoch": 0.44011976047904194,
+      "grad_norm": 0.9863676723538085,
+      "learning_rate": 8.851966586874138e-06,
+      "loss": 0.3535,
+      "step": 147
+    },
+    {
+      "epoch": 0.4431137724550898,
+      "grad_norm": 1.1480635087391566,
+      "learning_rate": 8.836931664057935e-06,
+      "loss": 0.3786,
+      "step": 148
+    },
+    {
+      "epoch": 0.44610778443113774,
+      "grad_norm": 1.2900219805732152,
+      "learning_rate": 8.821811875954705e-06,
+      "loss": 0.4084,
+      "step": 149
+    },
+    {
+      "epoch": 0.4491017964071856,
+      "grad_norm": 1.2434850983890071,
+      "learning_rate": 8.806607556984045e-06,
+      "loss": 0.3516,
+      "step": 150
+    },
+    {
+      "epoch": 0.45209580838323354,
+      "grad_norm": 1.1103675301393234,
+      "learning_rate": 8.791319043435213e-06,
+      "loss": 0.3922,
+      "step": 151
+    },
+    {
+      "epoch": 0.4550898203592814,
+      "grad_norm": 1.0724977713112234,
+      "learning_rate": 8.775946673459682e-06,
+      "loss": 0.3981,
+      "step": 152
+    },
+    {
+      "epoch": 0.45808383233532934,
+      "grad_norm": 1.1356579458756828,
+      "learning_rate": 8.76049078706366e-06,
+      "loss": 0.3702,
+      "step": 153
+    },
+    {
+      "epoch": 0.46107784431137727,
+      "grad_norm": 1.1684890846448495,
+      "learning_rate": 8.744951726100572e-06,
+      "loss": 0.3481,
+      "step": 154
+    },
+    {
+      "epoch": 0.46407185628742514,
+      "grad_norm": 1.3333674870196746,
+      "learning_rate": 8.729329834263503e-06,
+      "loss": 0.3872,
+      "step": 155
+    },
+    {
+      "epoch": 0.46706586826347307,
+      "grad_norm": 1.1814505291404438,
+      "learning_rate": 8.713625457077585e-06,
+      "loss": 0.4156,
+      "step": 156
+    },
+    {
+      "epoch": 0.47005988023952094,
+      "grad_norm": 1.1901307927041114,
+      "learning_rate": 8.697838941892371e-06,
+      "loss": 0.3688,
+      "step": 157
+    },
+    {
+      "epoch": 0.47305389221556887,
+      "grad_norm": 1.1331732167485076,
+      "learning_rate": 8.681970637874131e-06,
+      "loss": 0.4304,
+      "step": 158
+    },
+    {
+      "epoch": 0.47604790419161674,
+      "grad_norm": 1.1721302728278837,
+      "learning_rate": 8.666020895998154e-06,
+      "loss": 0.4005,
+      "step": 159
+    },
+    {
+      "epoch": 0.47904191616766467,
+      "grad_norm": 1.2229824490802395,
+      "learning_rate": 8.64999006904096e-06,
+      "loss": 0.4113,
+      "step": 160
+    },
+    {
+      "epoch": 0.4820359281437126,
+      "grad_norm": 1.2111824624744718,
+      "learning_rate": 8.63387851157252e-06,
+      "loss": 0.41,
+      "step": 161
+    },
+    {
+      "epoch": 0.48502994011976047,
+      "grad_norm": 1.0889742162795037,
+      "learning_rate": 8.617686579948396e-06,
+      "loss": 0.3524,
+      "step": 162
+    },
+    {
+      "epoch": 0.4880239520958084,
+      "grad_norm": 1.1930303197173053,
+      "learning_rate": 8.60141463230187e-06,
+      "loss": 0.3692,
+      "step": 163
+    },
+    {
+      "epoch": 0.49101796407185627,
+      "grad_norm": 1.2140600758712097,
+      "learning_rate": 8.585063028536015e-06,
+      "loss": 0.4604,
+      "step": 164
+    },
+    {
+      "epoch": 0.4940119760479042,
+      "grad_norm": 1.1421676297750811,
+      "learning_rate": 8.568632130315747e-06,
+      "loss": 0.3981,
+      "step": 165
+    },
+    {
+      "epoch": 0.49700598802395207,
+      "grad_norm": 1.145417740720218,
+      "learning_rate": 8.552122301059807e-06,
+      "loss": 0.3911,
+      "step": 166
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 1.1476294066364943,
+      "learning_rate": 8.535533905932739e-06,
+      "loss": 0.3682,
+      "step": 167
+    },
+    {
+      "epoch": 0.5029940119760479,
+      "grad_norm": 1.1768749197812503,
+      "learning_rate": 8.518867311836808e-06,
+      "loss": 0.4223,
+      "step": 168
+    },
+    {
+      "epoch": 0.5059880239520959,
+      "grad_norm": 1.1985092770477974,
+      "learning_rate": 8.502122887403882e-06,
+      "loss": 0.4269,
+      "step": 169
+    },
+    {
+      "epoch": 0.5089820359281437,
+      "grad_norm": 1.212921332884743,
+      "learning_rate": 8.485301002987285e-06,
+      "loss": 0.3513,
+      "step": 170
+    },
+    {
+      "epoch": 0.5119760479041916,
+      "grad_norm": 1.1219898239500206,
+      "learning_rate": 8.468402030653598e-06,
+      "loss": 0.4447,
+      "step": 171
+    },
+    {
+      "epoch": 0.5149700598802395,
+      "grad_norm": 1.0761180417261034,
+      "learning_rate": 8.451426344174433e-06,
+      "loss": 0.3786,
+      "step": 172
+    },
+    {
+      "epoch": 0.5179640718562875,
+      "grad_norm": 1.2553935405791927,
+      "learning_rate": 8.434374319018165e-06,
+      "loss": 0.456,
+      "step": 173
+    },
+    {
+      "epoch": 0.5209580838323353,
+      "grad_norm": 1.0722401547866776,
+      "learning_rate": 8.417246332341638e-06,
+      "loss": 0.3743,
+      "step": 174
+    },
+    {
+      "epoch": 0.5239520958083832,
+      "grad_norm": 1.0707404580910937,
+      "learning_rate": 8.4000427629818e-06,
+      "loss": 0.3872,
+      "step": 175
+    },
+    {
+      "epoch": 0.5269461077844312,
+      "grad_norm": 1.1259713806131728,
+      "learning_rate": 8.382763991447344e-06,
+      "loss": 0.4198,
+      "step": 176
+    },
+    {
+      "epoch": 0.5299401197604791,
+      "grad_norm": 1.210912459346478,
+      "learning_rate": 8.365410399910287e-06,
+      "loss": 0.4018,
+      "step": 177
+    },
+    {
+      "epoch": 0.5329341317365269,
+      "grad_norm": 1.2935878353750303,
+      "learning_rate": 8.347982372197515e-06,
+      "loss": 0.3901,
+      "step": 178
+    },
+    {
+      "epoch": 0.5359281437125748,
+      "grad_norm": 1.1200566168420634,
+      "learning_rate": 8.33048029378229e-06,
+      "loss": 0.4114,
+      "step": 179
+    },
+    {
+      "epoch": 0.5389221556886228,
+      "grad_norm": 1.0556979580147778,
+      "learning_rate": 8.312904551775731e-06,
+      "loss": 0.3538,
+      "step": 180
+    },
+    {
+      "epoch": 0.5419161676646707,
+      "grad_norm": 1.0967090952160596,
+      "learning_rate": 8.295255534918249e-06,
+      "loss": 0.3965,
+      "step": 181
+    },
+    {
+      "epoch": 0.5449101796407185,
+      "grad_norm": 1.3097373213391845,
+      "learning_rate": 8.277533633570948e-06,
+      "loss": 0.4531,
+      "step": 182
+    },
+    {
+      "epoch": 0.5479041916167665,
+      "grad_norm": 1.1226582918594663,
+      "learning_rate": 8.25973923970699e-06,
+      "loss": 0.4183,
+      "step": 183
+    },
+    {
+      "epoch": 0.5508982035928144,
+      "grad_norm": 1.039219167343767,
+      "learning_rate": 8.241872746902934e-06,
+      "loss": 0.4657,
+      "step": 184
+    },
+    {
+      "epoch": 0.5538922155688623,
+      "grad_norm": 1.1275158184140772,
+      "learning_rate": 8.223934550330015e-06,
+      "loss": 0.3734,
+      "step": 185
+    },
+    {
+      "epoch": 0.5568862275449101,
+      "grad_norm": 1.1477458063871067,
+      "learning_rate": 8.20592504674542e-06,
+      "loss": 0.4347,
+      "step": 186
+    },
+    {
+      "epoch": 0.5598802395209581,
+      "grad_norm": 1.021684936128821,
+      "learning_rate": 8.187844634483495e-06,
+      "loss": 0.4344,
+      "step": 187
+    },
+    {
+      "epoch": 0.562874251497006,
+      "grad_norm": 1.2168769543525833,
+      "learning_rate": 8.16969371344696e-06,
+      "loss": 0.4936,
+      "step": 188
+    },
+    {
+      "epoch": 0.5658682634730539,
+      "grad_norm": 1.0110104450513255,
+      "learning_rate": 8.151472685098037e-06,
+      "loss": 0.395,
+      "step": 189
+    },
+    {
+      "epoch": 0.5688622754491018,
+      "grad_norm": 1.063716255568497,
+      "learning_rate": 8.13318195244958e-06,
+      "loss": 0.3752,
+      "step": 190
+    },
+    {
+      "epoch": 0.5718562874251497,
+      "grad_norm": 1.062381590001417,
+      "learning_rate": 8.114821920056177e-06,
+      "loss": 0.3963,
+      "step": 191
+    },
+    {
+      "epoch": 0.5748502994011976,
+      "grad_norm": 1.0793181592301355,
+      "learning_rate": 8.096392994005177e-06,
+      "loss": 0.4227,
+      "step": 192
+    },
+    {
+      "epoch": 0.5778443113772455,
+      "grad_norm": 0.950772622388193,
+      "learning_rate": 8.077895581907719e-06,
+      "loss": 0.3883,
+      "step": 193
+    },
+    {
+      "epoch": 0.5808383233532934,
+      "grad_norm": 0.9855401248100542,
+      "learning_rate": 8.059330092889724e-06,
+      "loss": 0.3485,
+      "step": 194
+    },
+    {
+      "epoch": 0.5838323353293413,
+      "grad_norm": 1.0787923858980941,
+      "learning_rate": 8.040696937582833e-06,
+      "loss": 0.3958,
+      "step": 195
+    },
+    {
+      "epoch": 0.5868263473053892,
+      "grad_norm": 1.1469559132322573,
+      "learning_rate": 8.021996528115335e-06,
+      "loss": 0.3682,
+      "step": 196
+    },
+    {
+      "epoch": 0.5898203592814372,
+      "grad_norm": 1.1597858690820446,
+      "learning_rate": 8.003229278103044e-06,
+      "loss": 0.371,
+      "step": 197
+    },
+    {
+      "epoch": 0.592814371257485,
+      "grad_norm": 1.1123736948863556,
+      "learning_rate": 7.984395602640153e-06,
+      "loss": 0.376,
+      "step": 198
+    },
+    {
+      "epoch": 0.5958083832335329,
+      "grad_norm": 1.2373592482761206,
+      "learning_rate": 7.96549591829006e-06,
+      "loss": 0.3909,
+      "step": 199
+    },
+    {
+      "epoch": 0.5988023952095808,
+      "grad_norm": 1.0110935828683547,
+      "learning_rate": 7.946530643076138e-06,
+      "loss": 0.342,
+      "step": 200
+    },
+    {
+      "epoch": 0.5988023952095808,
+      "eval_loss": 0.38591626286506653,
+      "eval_runtime": 2.289,
+      "eval_samples_per_second": 23.591,
+      "eval_steps_per_second": 6.116,
+      "step": 200
+    },
+    {
+      "epoch": 0.6017964071856288,
+      "grad_norm": 1.0729832153985162,
+      "learning_rate": 7.927500196472506e-06,
+      "loss": 0.3912,
+      "step": 201
+    },
+    {
+      "epoch": 0.6047904191616766,
+      "grad_norm": 1.18932385649841,
+      "learning_rate": 7.908404999394747e-06,
+      "loss": 0.3427,
+      "step": 202
+    },
+    {
+      "epoch": 0.6077844311377245,
+      "grad_norm": 1.1044327171319122,
+      "learning_rate": 7.889245474190588e-06,
+      "loss": 0.3868,
+      "step": 203
+    },
+    {
+      "epoch": 0.6107784431137725,
+      "grad_norm": 1.2691585011376387,
+      "learning_rate": 7.870022044630569e-06,
+      "loss": 0.416,
+      "step": 204
+    },
+    {
+      "epoch": 0.6137724550898204,
+      "grad_norm": 1.0379644199960554,
+      "learning_rate": 7.85073513589867e-06,
+      "loss": 0.3931,
+      "step": 205
+    },
+    {
+      "epoch": 0.6167664670658682,
+      "grad_norm": 1.4019618981062842,
+      "learning_rate": 7.831385174582901e-06,
+      "loss": 0.4022,
+      "step": 206
+    },
+    {
+      "epoch": 0.6197604790419161,
+      "grad_norm": 1.0791577660366212,
+      "learning_rate": 7.81197258866587e-06,
+      "loss": 0.3934,
+      "step": 207
+    },
+    {
+      "epoch": 0.6227544910179641,
+      "grad_norm": 1.1222440788383483,
+      "learning_rate": 7.792497807515317e-06,
+      "loss": 0.3542,
+      "step": 208
+    },
+    {
+      "epoch": 0.625748502994012,
+      "grad_norm": 1.155560270765335,
+      "learning_rate": 7.772961261874615e-06,
+      "loss": 0.3434,
+      "step": 209
+    },
+    {
+      "epoch": 0.6287425149700598,
+      "grad_norm": 1.1419669667397165,
+      "learning_rate": 7.75336338385325e-06,
+      "loss": 0.3792,
+      "step": 210
+    },
+    {
+      "epoch": 0.6317365269461078,
+      "grad_norm": 1.1240382220271334,
+      "learning_rate": 7.733704606917248e-06,
+      "loss": 0.4135,
+      "step": 211
+    },
+    {
+      "epoch": 0.6347305389221557,
+      "grad_norm": 1.1397664736616373,
+      "learning_rate": 7.713985365879607e-06,
+      "loss": 0.356,
+      "step": 212
+    },
+    {
+      "epoch": 0.6377245508982036,
+      "grad_norm": 1.1180198632885519,
+      "learning_rate": 7.694206096890667e-06,
+      "loss": 0.3698,
+      "step": 213
+    },
+    {
+      "epoch": 0.6407185628742516,
+      "grad_norm": 1.088320296312747,
+      "learning_rate": 7.674367237428467e-06,
+      "loss": 0.3502,
+      "step": 214
+    },
+    {
+      "epoch": 0.6437125748502994,
+      "grad_norm": 1.048807327112999,
+      "learning_rate": 7.654469226289068e-06,
+      "loss": 0.3356,
+      "step": 215
+    },
+    {
+      "epoch": 0.6467065868263473,
+      "grad_norm": 1.12927350812835,
+      "learning_rate": 7.63451250357685e-06,
+      "loss": 0.4058,
+      "step": 216
+    },
+    {
+      "epoch": 0.6497005988023952,
+      "grad_norm": 1.1247863713079453,
+      "learning_rate": 7.614497510694774e-06,
+      "loss": 0.4053,
+      "step": 217
+    },
+    {
+      "epoch": 0.6526946107784432,
+      "grad_norm": 1.2028599369113613,
+      "learning_rate": 7.5944246903346204e-06,
+      "loss": 0.3738,
+      "step": 218
+    },
+    {
+      "epoch": 0.655688622754491,
+      "grad_norm": 1.0864281064020682,
+      "learning_rate": 7.574294486467204e-06,
+      "loss": 0.3639,
+      "step": 219
+    },
+    {
+      "epoch": 0.6586826347305389,
+      "grad_norm": 1.122796070778498,
+      "learning_rate": 7.55410734433254e-06,
+      "loss": 0.4135,
+      "step": 220
+    },
+    {
+      "epoch": 0.6616766467065869,
+      "grad_norm": 1.1911986664040115,
+      "learning_rate": 7.533863710430011e-06,
+      "loss": 0.4306,
+      "step": 221
+    },
+    {
+      "epoch": 0.6646706586826348,
+      "grad_norm": 1.1429207410821842,
+      "learning_rate": 7.513564032508484e-06,
+      "loss": 0.3603,
+      "step": 222
+    },
+    {
+      "epoch": 0.6676646706586826,
+      "grad_norm": 1.255131460493026,
+      "learning_rate": 7.493208759556406e-06,
+      "loss": 0.4187,
+      "step": 223
+    },
+    {
+      "epoch": 0.6706586826347305,
+      "grad_norm": 1.2253223598034224,
+      "learning_rate": 7.472798341791877e-06,
+      "loss": 0.4525,
+      "step": 224
+    },
+    {
+      "epoch": 0.6736526946107785,
+      "grad_norm": 1.0880060470858857,
+      "learning_rate": 7.452333230652688e-06,
+      "loss": 0.3773,
+      "step": 225
+    },
+    {
+      "epoch": 0.6766467065868264,
+      "grad_norm": 1.12106842517365,
+      "learning_rate": 7.431813878786343e-06,
+      "loss": 0.367,
+      "step": 226
+    },
+    {
+      "epoch": 0.6796407185628742,
+      "grad_norm": 1.1978533976912833,
+      "learning_rate": 7.4112407400400395e-06,
+      "loss": 0.408,
+      "step": 227
+    },
+    {
+      "epoch": 0.6826347305389222,
+      "grad_norm": 1.0839021713515942,
+      "learning_rate": 7.390614269450633e-06,
+      "loss": 0.3788,
+      "step": 228
+    },
+    {
+      "epoch": 0.6856287425149701,
+      "grad_norm": 0.977543609935443,
+      "learning_rate": 7.369934923234577e-06,
+      "loss": 0.3542,
+      "step": 229
+    },
+    {
+      "epoch": 0.688622754491018,
+      "grad_norm": 1.2295578136990957,
+      "learning_rate": 7.349203158777826e-06,
+      "loss": 0.4477,
+      "step": 230
+    },
+    {
+      "epoch": 0.6916167664670658,
+      "grad_norm": 1.188980819466126,
+      "learning_rate": 7.32841943462572e-06,
+      "loss": 0.4001,
+      "step": 231
+    },
+    {
+      "epoch": 0.6946107784431138,
+      "grad_norm": 1.2126066312503347,
+      "learning_rate": 7.3075842104728445e-06,
+      "loss": 0.3969,
+      "step": 232
+    },
+    {
+      "epoch": 0.6976047904191617,
+      "grad_norm": 1.1592074213862762,
+      "learning_rate": 7.286697947152868e-06,
+      "loss": 0.4169,
+      "step": 233
+    },
+    {
+      "epoch": 0.7005988023952096,
+      "grad_norm": 1.1285084481638699,
+      "learning_rate": 7.265761106628338e-06,
+      "loss": 0.3795,
+      "step": 234
+    },
+    {
+      "epoch": 0.7035928143712575,
+      "grad_norm": 1.0378300254833732,
+      "learning_rate": 7.244774151980466e-06,
+      "loss": 0.3685,
+      "step": 235
+    },
+    {
+      "epoch": 0.7065868263473054,
+      "grad_norm": 1.3577946200558169,
+      "learning_rate": 7.223737547398898e-06,
+      "loss": 0.3785,
+      "step": 236
+    },
+    {
+      "epoch": 0.7095808383233533,
+      "grad_norm": 1.1267184381163402,
+      "learning_rate": 7.20265175817143e-06,
+      "loss": 0.3905,
+      "step": 237
+    },
+    {
+      "epoch": 0.7125748502994012,
+      "grad_norm": 1.2404457496704617,
+      "learning_rate": 7.181517250673729e-06,
+      "loss": 0.4275,
+      "step": 238
+    },
+    {
+      "epoch": 0.7155688622754491,
+      "grad_norm": 1.1486415642920504,
+      "learning_rate": 7.1603344923590065e-06,
+      "loss": 0.3757,
+      "step": 239
+    },
+    {
+      "epoch": 0.718562874251497,
+      "grad_norm": 1.063545683390086,
+      "learning_rate": 7.139103951747694e-06,
+      "loss": 0.4129,
+      "step": 240
+    },
+    {
+      "epoch": 0.7215568862275449,
+      "grad_norm": 1.0846163686526613,
+      "learning_rate": 7.1178260984170675e-06,
+      "loss": 0.3544,
+      "step": 241
+    },
+    {
+      "epoch": 0.7245508982035929,
+      "grad_norm": 1.1261583526578574,
+      "learning_rate": 7.0965014029908654e-06,
+      "loss": 0.3553,
+      "step": 242
+    },
+    {
+      "epoch": 0.7275449101796407,
+      "grad_norm": 1.020331742529792,
+      "learning_rate": 7.075130337128883e-06,
+      "loss": 0.387,
+      "step": 243
+    },
+    {
+      "epoch": 0.7305389221556886,
+      "grad_norm": 1.126460450902038,
+      "learning_rate": 7.053713373516538e-06,
+      "loss": 0.4066,
+      "step": 244
+    },
+    {
+      "epoch": 0.7335329341317365,
+      "grad_norm": 1.0779114477360106,
+      "learning_rate": 7.03225098585441e-06,
+      "loss": 0.4214,
+      "step": 245
+    },
+    {
+      "epoch": 0.7365269461077845,
+      "grad_norm": 1.0996097863606056,
+      "learning_rate": 7.0107436488477694e-06,
+      "loss": 0.4009,
+      "step": 246
+    },
+    {
+      "epoch": 0.7395209580838323,
+      "grad_norm": 1.1053513813267253,
+      "learning_rate": 6.989191838196083e-06,
+      "loss": 0.4146,
+      "step": 247
+    },
+    {
+      "epoch": 0.7425149700598802,
+      "grad_norm": 1.102337522649066,
+      "learning_rate": 6.9675960305824785e-06,
+      "loss": 0.3661,
+      "step": 248
+    },
+    {
+      "epoch": 0.7455089820359282,
+      "grad_norm": 1.282546879486523,
+      "learning_rate": 6.945956703663212e-06,
+      "loss": 0.4116,
+      "step": 249
+    },
+    {
+      "epoch": 0.7485029940119761,
+      "grad_norm": 1.0117028443061191,
+      "learning_rate": 6.9242743360570985e-06,
+      "loss": 0.3728,
+      "step": 250
+    },
+    {
+      "epoch": 0.7514970059880239,
+      "grad_norm": 1.256014110199748,
+      "learning_rate": 6.9025494073349284e-06,
+      "loss": 0.4259,
+      "step": 251
+    },
+    {
+      "epoch": 0.7544910179640718,
+      "grad_norm": 0.9930145550071506,
+      "learning_rate": 6.880782398008862e-06,
+      "loss": 0.3886,
+      "step": 252
+    },
+    {
+      "epoch": 0.7574850299401198,
+      "grad_norm": 1.174720504315703,
+      "learning_rate": 6.858973789521792e-06,
+      "loss": 0.3513,
+      "step": 253
+    },
+    {
+      "epoch": 0.7604790419161677,
+      "grad_norm": 1.1521932603814395,
+      "learning_rate": 6.837124064236709e-06,
+      "loss": 0.4158,
+      "step": 254
+    },
+    {
+      "epoch": 0.7634730538922155,
+      "grad_norm": 1.057417799552566,
+      "learning_rate": 6.815233705426019e-06,
+      "loss": 0.3835,
+      "step": 255
+    },
+    {
+      "epoch": 0.7664670658682635,
+      "grad_norm": 1.1199558891955381,
+      "learning_rate": 6.7933031972608644e-06,
+      "loss": 0.4144,
+      "step": 256
+    },
+    {
+      "epoch": 0.7694610778443114,
+      "grad_norm": 1.023379955504229,
+      "learning_rate": 6.771333024800411e-06,
+      "loss": 0.3922,
+      "step": 257
+    },
+    {
+      "epoch": 0.7724550898203593,
+      "grad_norm": 1.3083445788756933,
+      "learning_rate": 6.74932367398112e-06,
+      "loss": 0.3869,
+      "step": 258
+    },
+    {
+      "epoch": 0.7754491017964071,
+      "grad_norm": 1.0887488281339772,
+      "learning_rate": 6.727275631605996e-06,
+      "loss": 0.3404,
+      "step": 259
+    },
+    {
+      "epoch": 0.7784431137724551,
+      "grad_norm": 1.1784023860428798,
+      "learning_rate": 6.70518938533383e-06,
+      "loss": 0.3871,
+      "step": 260
+    },
+    {
+      "epoch": 0.781437125748503,
+      "grad_norm": 1.1193835581342824,
+      "learning_rate": 6.683065423668403e-06,
+      "loss": 0.3998,
+      "step": 261
+    },
+    {
+      "epoch": 0.7844311377245509,
+      "grad_norm": 1.1348145935462188,
+      "learning_rate": 6.660904235947687e-06,
+      "loss": 0.3876,
+      "step": 262
+    },
+    {
+      "epoch": 0.7874251497005988,
+      "grad_norm": 1.1263241588486157,
+      "learning_rate": 6.638706312333018e-06,
+      "loss": 0.4111,
+      "step": 263
+    },
+    {
+      "epoch": 0.7904191616766467,
+      "grad_norm": 1.1165913196710078,
+      "learning_rate": 6.61647214379826e-06,
+      "loss": 0.4467,
+      "step": 264
+    },
+    {
+      "epoch": 0.7934131736526946,
+      "grad_norm": 1.2420708970061312,
+      "learning_rate": 6.594202222118941e-06,
+      "loss": 0.4076,
+      "step": 265
+    },
+    {
+      "epoch": 0.7964071856287425,
+      "grad_norm": 1.1554874789619023,
+      "learning_rate": 6.571897039861377e-06,
+      "loss": 0.3887,
+      "step": 266
+    },
+    {
+      "epoch": 0.7994011976047904,
+      "grad_norm": 1.0664250417634766,
+      "learning_rate": 6.549557090371775e-06,
+      "loss": 0.3965,
+      "step": 267
+    },
+    {
+      "epoch": 0.8023952095808383,
+      "grad_norm": 1.0277517806594927,
+      "learning_rate": 6.527182867765333e-06,
+      "loss": 0.2948,
+      "step": 268
+    },
+    {
+      "epoch": 0.8053892215568862,
+      "grad_norm": 1.187700484201102,
+      "learning_rate": 6.504774866915291e-06,
+      "loss": 0.382,
+      "step": 269
+    },
+    {
+      "epoch": 0.8083832335329342,
+      "grad_norm": 1.2219208696784514,
+      "learning_rate": 6.482333583442002e-06,
+      "loss": 0.3539,
+      "step": 270
+    },
+    {
+      "epoch": 0.811377245508982,
+      "grad_norm": 1.007372801812248,
+      "learning_rate": 6.459859513701967e-06,
+      "loss": 0.3836,
+      "step": 271
+    },
+    {
+      "epoch": 0.8143712574850299,
+      "grad_norm": 1.3533632179372996,
+      "learning_rate": 6.437353154776848e-06,
+      "loss": 0.393,
+      "step": 272
+    },
+    {
+      "epoch": 0.8173652694610778,
+      "grad_norm": 1.1563157682673428,
+      "learning_rate": 6.414815004462483e-06,
+      "loss": 0.4226,
+      "step": 273
+    },
+    {
+      "epoch": 0.8203592814371258,
+      "grad_norm": 1.1032306960399896,
+      "learning_rate": 6.3922455612578715e-06,
+      "loss": 0.4165,
+      "step": 274
+    },
+    {
+      "epoch": 0.8233532934131736,
+      "grad_norm": 1.1102057753993912,
+      "learning_rate": 6.369645324354149e-06,
+      "loss": 0.3343,
+      "step": 275
+    },
+    {
+      "epoch": 0.8263473053892215,
+      "grad_norm": 1.0360686612504768,
+      "learning_rate": 6.3470147936235485e-06,
+      "loss": 0.3913,
+      "step": 276
+    },
+    {
+      "epoch": 0.8293413173652695,
+      "grad_norm": 0.9274487760976012,
+      "learning_rate": 6.3243544696083355e-06,
+      "loss": 0.3496,
+      "step": 277
+    },
+    {
+      "epoch": 0.8323353293413174,
+      "grad_norm": 1.0462358075954483,
+      "learning_rate": 6.301664853509755e-06,
+      "loss": 0.3314,
+      "step": 278
+    },
+    {
+      "epoch": 0.8353293413173652,
+      "grad_norm": 0.9916042060622593,
+      "learning_rate": 6.278946447176924e-06,
+      "loss": 0.3764,
+      "step": 279
+    },
+    {
+      "epoch": 0.8383233532934131,
+      "grad_norm": 1.0340902632598676,
+      "learning_rate": 6.256199753095745e-06,
+      "loss": 0.3565,
+      "step": 280
+    },
+    {
+      "epoch": 0.8413173652694611,
+      "grad_norm": 1.0719705879982018,
+      "learning_rate": 6.233425274377793e-06,
+      "loss": 0.3761,
+      "step": 281
+    },
+    {
+      "epoch": 0.844311377245509,
+      "grad_norm": 1.326589338818543,
+      "learning_rate": 6.21062351474918e-06,
+      "loss": 0.4309,
+      "step": 282
+    },
+    {
+      "epoch": 0.8473053892215568,
+      "grad_norm": 1.1312591703539556,
+      "learning_rate": 6.18779497853942e-06,
+      "loss": 0.401,
+      "step": 283
+    },
+    {
+      "epoch": 0.8502994011976048,
+      "grad_norm": 1.1177205936019006,
+      "learning_rate": 6.164940170670266e-06,
+      "loss": 0.4152,
+      "step": 284
+    },
+    {
+      "epoch": 0.8532934131736527,
+      "grad_norm": 0.964332210749677,
+      "learning_rate": 6.142059596644557e-06,
+      "loss": 0.3302,
+      "step": 285
+    },
+    {
+      "epoch": 0.8562874251497006,
+      "grad_norm": 1.0978681880015153,
+      "learning_rate": 6.11915376253502e-06,
+      "loss": 0.3616,
+      "step": 286
+    },
+    {
+      "epoch": 0.8592814371257484,
+      "grad_norm": 1.1226350482889762,
+      "learning_rate": 6.096223174973091e-06,
+      "loss": 0.3922,
+      "step": 287
+    },
+    {
+      "epoch": 0.8622754491017964,
+      "grad_norm": 1.0280132940644005,
+      "learning_rate": 6.073268341137694e-06,
+      "loss": 0.3276,
+      "step": 288
+    },
+    {
+      "epoch": 0.8652694610778443,
+      "grad_norm": 1.1499197494706406,
+      "learning_rate": 6.050289768744042e-06,
+      "loss": 0.3153,
+      "step": 289
+    },
+    {
+      "epoch": 0.8682634730538922,
+      "grad_norm": 1.2485245559017402,
+      "learning_rate": 6.0272879660323936e-06,
+      "loss": 0.4172,
+      "step": 290
+    },
+    {
+      "epoch": 0.8712574850299402,
+      "grad_norm": 1.1425976944474154,
+      "learning_rate": 6.004263441756815e-06,
+      "loss": 0.3901,
+      "step": 291
+    },
+    {
+      "epoch": 0.874251497005988,
+      "grad_norm": 1.1410181695352197,
+      "learning_rate": 5.98121670517393e-06,
+      "loss": 0.366,
+      "step": 292
+    },
+    {
+      "epoch": 0.8772455089820359,
+      "grad_norm": 0.9997129838690438,
+      "learning_rate": 5.958148266031654e-06,
+      "loss": 0.3349,
+      "step": 293
+    },
+    {
+      "epoch": 0.8802395209580839,
+      "grad_norm": 1.0908025268939436,
+      "learning_rate": 5.935058634557917e-06,
+      "loss": 0.3519,
+      "step": 294
+    },
+    {
+      "epoch": 0.8832335329341318,
+      "grad_norm": 0.9701503875036278,
+      "learning_rate": 5.911948321449384e-06,
+      "loss": 0.3769,
+      "step": 295
+    },
+    {
+      "epoch": 0.8862275449101796,
+      "grad_norm": 1.1805113050429954,
+      "learning_rate": 5.8888178378601565e-06,
+      "loss": 0.3972,
+      "step": 296
+    },
+    {
+      "epoch": 0.8892215568862275,
+      "grad_norm": 1.2046017883565834,
+      "learning_rate": 5.865667695390468e-06,
+      "loss": 0.3828,
+      "step": 297
+    },
+    {
+      "epoch": 0.8922155688622755,
+      "grad_norm": 1.0548858311144647,
+      "learning_rate": 5.842498406075363e-06,
+      "loss": 0.3605,
+      "step": 298
+    },
+    {
+      "epoch": 0.8952095808383234,
+      "grad_norm": 1.1362463789853134,
+      "learning_rate": 5.819310482373381e-06,
+      "loss": 0.3775,
+      "step": 299
+    },
+    {
+      "epoch": 0.8982035928143712,
+      "grad_norm": 1.0847854699637618,
+      "learning_rate": 5.796104437155213e-06,
+      "loss": 0.4242,
+      "step": 300
+    },
+    {
+      "epoch": 0.9011976047904192,
+      "grad_norm": 1.1448044438753795,
+      "learning_rate": 5.772880783692363e-06,
+      "loss": 0.3577,
+      "step": 301
+    },
+    {
+      "epoch": 0.9041916167664671,
+      "grad_norm": 1.1832587209873984,
+      "learning_rate": 5.749640035645798e-06,
+      "loss": 0.3529,
+      "step": 302
+    },
+    {
+      "epoch": 0.907185628742515,
+      "grad_norm": 1.1761920912189556,
+      "learning_rate": 5.726382707054578e-06,
+      "loss": 0.3792,
+      "step": 303
+    },
+    {
+      "epoch": 0.9101796407185628,
+      "grad_norm": 1.1526886407877144,
+      "learning_rate": 5.703109312324493e-06,
+      "loss": 0.3743,
+      "step": 304
+    },
+    {
+      "epoch": 0.9131736526946108,
+      "grad_norm": 1.2247851503316385,
+      "learning_rate": 5.679820366216684e-06,
+      "loss": 0.4548,
+      "step": 305
+    },
+    {
+      "epoch": 0.9161676646706587,
+      "grad_norm": 1.0502283874341918,
+      "learning_rate": 5.656516383836263e-06,
+      "loss": 0.331,
+      "step": 306
+    },
+    {
+      "epoch": 0.9191616766467066,
+      "grad_norm": 1.0937740393794946,
+      "learning_rate": 5.6331978806209044e-06,
+      "loss": 0.3945,
+      "step": 307
+    },
+    {
+      "epoch": 0.9221556886227545,
+      "grad_norm": 1.109376989761775,
+      "learning_rate": 5.609865372329461e-06,
+      "loss": 0.3606,
+      "step": 308
+    },
+    {
+      "epoch": 0.9251497005988024,
+      "grad_norm": 1.0643397774514596,
+      "learning_rate": 5.586519375030549e-06,
+      "loss": 0.3244,
+      "step": 309
+    },
+    {
+      "epoch": 0.9281437125748503,
+      "grad_norm": 1.0147266000825244,
+      "learning_rate": 5.5631604050911354e-06,
+      "loss": 0.3528,
+      "step": 310
+    },
+    {
+      "epoch": 0.9311377245508982,
+      "grad_norm": 1.2609062780278828,
+      "learning_rate": 5.539788979165115e-06,
+      "loss": 0.337,
+      "step": 311
+    },
+    {
+      "epoch": 0.9341317365269461,
+      "grad_norm": 0.9953690002226568,
+      "learning_rate": 5.516405614181883e-06,
+      "loss": 0.315,
+      "step": 312
+    },
+    {
+      "epoch": 0.937125748502994,
+      "grad_norm": 0.998446752455762,
+      "learning_rate": 5.4930108273349034e-06,
+      "loss": 0.352,
+      "step": 313
+    },
+    {
+      "epoch": 0.9401197604790419,
+      "grad_norm": 1.1221409376092786,
+      "learning_rate": 5.4696051360702725e-06,
+      "loss": 0.3389,
+      "step": 314
+    },
+    {
+      "epoch": 0.9431137724550899,
+      "grad_norm": 1.208847726716205,
+      "learning_rate": 5.446189058075265e-06,
+      "loss": 0.3977,
+      "step": 315
+    },
+    {
+      "epoch": 0.9461077844311377,
+      "grad_norm": 1.1225402007537828,
+      "learning_rate": 5.4227631112668955e-06,
+      "loss": 0.4058,
+      "step": 316
+    },
+    {
+      "epoch": 0.9491017964071856,
+      "grad_norm": 1.2009789331047636,
+      "learning_rate": 5.39932781378045e-06,
+      "loss": 0.4095,
+      "step": 317
+    },
+    {
+      "epoch": 0.9520958083832335,
+      "grad_norm": 0.992709027811188,
+      "learning_rate": 5.375883683958041e-06,
+      "loss": 0.3589,
+      "step": 318
+    },
+    {
+      "epoch": 0.9550898203592815,
+      "grad_norm": 1.1051317636593356,
+      "learning_rate": 5.3524312403371255e-06,
+      "loss": 0.3757,
+      "step": 319
+    },
+    {
+      "epoch": 0.9580838323353293,
+      "grad_norm": 1.256858633355358,
+      "learning_rate": 5.328971001639054e-06,
+      "loss": 0.335,
+      "step": 320
+    },
+    {
+      "epoch": 0.9610778443113772,
+      "grad_norm": 1.0940902027790071,
+      "learning_rate": 5.3055034867575825e-06,
+      "loss": 0.3422,
+      "step": 321
+    },
+    {
+      "epoch": 0.9640718562874252,
+      "grad_norm": 1.305524482407352,
+      "learning_rate": 5.282029214747404e-06,
+      "loss": 0.3985,
+      "step": 322
+    },
+    {
+      "epoch": 0.9670658682634731,
+      "grad_norm": 1.121978134882831,
+      "learning_rate": 5.258548704812667e-06,
+      "loss": 0.388,
+      "step": 323
+    },
+    {
+      "epoch": 0.9700598802395209,
+      "grad_norm": 1.1717652092862558,
+      "learning_rate": 5.235062476295488e-06,
+      "loss": 0.3713,
+      "step": 324
+    },
+    {
+      "epoch": 0.9730538922155688,
+      "grad_norm": 1.1488416652544782,
+      "learning_rate": 5.211571048664469e-06,
+      "loss": 0.328,
+      "step": 325
+    },
+    {
+      "epoch": 0.9760479041916168,
+      "grad_norm": 1.1222535144279493,
+      "learning_rate": 5.188074941503203e-06,
+      "loss": 0.3852,
+      "step": 326
+    },
+    {
+      "epoch": 0.9790419161676647,
+      "grad_norm": 1.1093510464041303,
+      "learning_rate": 5.164574674498788e-06,
+      "loss": 0.3771,
+      "step": 327
+    },
+    {
+      "epoch": 0.9820359281437125,
+      "grad_norm": 1.2042609761515466,
+      "learning_rate": 5.141070767430331e-06,
+      "loss": 0.3589,
+      "step": 328
+    },
+    {
+      "epoch": 0.9850299401197605,
+      "grad_norm": 1.1042356003334075,
+      "learning_rate": 5.117563740157444e-06,
+      "loss": 0.3637,
+      "step": 329
+    },
+    {
+      "epoch": 0.9880239520958084,
+      "grad_norm": 1.2223227352467507,
+      "learning_rate": 5.094054112608758e-06,
+      "loss": 0.4368,
+      "step": 330
+    },
+    {
+      "epoch": 0.9910179640718563,
+      "grad_norm": 1.0588393145709278,
+      "learning_rate": 5.070542404770413e-06,
+      "loss": 0.4087,
+      "step": 331
+    },
+    {
+      "epoch": 0.9940119760479041,
+      "grad_norm": 1.0125661943933963,
+      "learning_rate": 5.047029136674563e-06,
+      "loss": 0.3324,
+      "step": 332
+    },
+    {
+      "epoch": 0.9970059880239521,
+      "grad_norm": 1.2347879161144995,
+      "learning_rate": 5.023514828387868e-06,
+      "loss": 0.5011,
+      "step": 333
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.073632610937019,
+      "learning_rate": 5e-06,
+      "loss": 0.3002,
+      "step": 334
+    },
+    {
+      "epoch": 1.0029940119760479,
+      "grad_norm": 1.1476863960018056,
+      "learning_rate": 4.976485171612134e-06,
+      "loss": 0.241,
+      "step": 335
+    },
+    {
+      "epoch": 1.0059880239520957,
+      "grad_norm": 1.003038510122658,
+      "learning_rate": 4.95297086332544e-06,
+      "loss": 0.2597,
+      "step": 336
+    },
+    {
+      "epoch": 1.0089820359281436,
+      "grad_norm": 1.1718193950143698,
+      "learning_rate": 4.9294575952295896e-06,
+      "loss": 0.2781,
+      "step": 337
+    },
+    {
+      "epoch": 1.0119760479041917,
+      "grad_norm": 1.1674905598767986,
+      "learning_rate": 4.905945887391242e-06,
+      "loss": 0.3038,
+      "step": 338
+    },
+    {
+      "epoch": 1.0149700598802396,
+      "grad_norm": 1.144503338753483,
+      "learning_rate": 4.882436259842556e-06,
+      "loss": 0.2805,
+      "step": 339
+    },
+    {
+      "epoch": 1.0179640718562875,
+      "grad_norm": 1.036134384266589,
+      "learning_rate": 4.858929232569671e-06,
+      "loss": 0.2713,
+      "step": 340
+    },
+    {
+      "epoch": 1.0209580838323353,
+      "grad_norm": 1.1123685539991703,
+      "learning_rate": 4.835425325501214e-06,
+      "loss": 0.2827,
+      "step": 341
+    },
+    {
+      "epoch": 1.0239520958083832,
+      "grad_norm": 1.0263723051298212,
+      "learning_rate": 4.811925058496799e-06,
+      "loss": 0.2263,
+      "step": 342
+    },
+    {
+      "epoch": 1.026946107784431,
+      "grad_norm": 1.2677078473795624,
+      "learning_rate": 4.788428951335534e-06,
+      "loss": 0.2923,
+      "step": 343
+    },
+    {
+      "epoch": 1.029940119760479,
+      "grad_norm": 1.00978820050344,
+      "learning_rate": 4.7649375237045135e-06,
+      "loss": 0.2622,
+      "step": 344
+    },
+    {
+      "epoch": 1.032934131736527,
+      "grad_norm": 1.0899621607861625,
+      "learning_rate": 4.741451295187333e-06,
+      "loss": 0.2343,
+      "step": 345
+    },
+    {
+      "epoch": 1.035928143712575,
+      "grad_norm": 1.119075599907004,
+      "learning_rate": 4.717970785252596e-06,
+      "loss": 0.3206,
+      "step": 346
+    },
+    {
+      "epoch": 1.0389221556886228,
+      "grad_norm": 1.1072174024446133,
+      "learning_rate": 4.694496513242418e-06,
+      "loss": 0.3015,
+      "step": 347
+    },
+    {
+      "epoch": 1.0419161676646707,
+      "grad_norm": 1.21750606542217,
+      "learning_rate": 4.671028998360947e-06,
+      "loss": 0.2457,
+      "step": 348
+    },
+    {
+      "epoch": 1.0449101796407185,
+      "grad_norm": 1.2255087751895424,
+      "learning_rate": 4.647568759662876e-06,
+      "loss": 0.3055,
+      "step": 349
+    },
+    {
+      "epoch": 1.0479041916167664,
+      "grad_norm": 1.0828492062822344,
+      "learning_rate": 4.624116316041962e-06,
+      "loss": 0.2832,
+      "step": 350
+    },
+    {
+      "epoch": 1.0508982035928143,
+      "grad_norm": 1.1139322977906365,
+      "learning_rate": 4.600672186219551e-06,
+      "loss": 0.2927,
+      "step": 351
+    },
+    {
+      "epoch": 1.0538922155688624,
+      "grad_norm": 1.2275275119137352,
+      "learning_rate": 4.5772368887331044e-06,
+      "loss": 0.3232,
+      "step": 352
+    },
+    {
+      "epoch": 1.0568862275449102,
+      "grad_norm": 1.0873924118605698,
+      "learning_rate": 4.553810941924735e-06,
+      "loss": 0.3148,
+      "step": 353
+    },
+    {
+      "epoch": 1.0598802395209581,
+      "grad_norm": 1.1100215222525451,
+      "learning_rate": 4.530394863929728e-06,
+      "loss": 0.2835,
+      "step": 354
+    },
+    {
+      "epoch": 1.062874251497006,
+      "grad_norm": 1.16737267497979,
+      "learning_rate": 4.506989172665097e-06,
+      "loss": 0.2696,
+      "step": 355
+    },
+    {
+      "epoch": 1.0658682634730539,
+      "grad_norm": 1.203334209451657,
+      "learning_rate": 4.483594385818119e-06,
+      "loss": 0.264,
+      "step": 356
+    },
+    {
+      "epoch": 1.0688622754491017,
+      "grad_norm": 1.2011397338857834,
+      "learning_rate": 4.460211020834887e-06,
+      "loss": 0.2742,
+      "step": 357
+    },
+    {
+      "epoch": 1.0718562874251496,
+      "grad_norm": 1.2098118793414803,
+      "learning_rate": 4.436839594908866e-06,
+      "loss": 0.3472,
+      "step": 358
+    },
+    {
+      "epoch": 1.0748502994011977,
+      "grad_norm": 1.2419411055719038,
+      "learning_rate": 4.4134806249694514e-06,
+      "loss": 0.3117,
+      "step": 359
+    },
+    {
+      "epoch": 1.0778443113772456,
+      "grad_norm": 1.2044686504014432,
+      "learning_rate": 4.39013462767054e-06,
+      "loss": 0.3288,
+      "step": 360
+    },
+    {
+      "epoch": 1.0808383233532934,
+      "grad_norm": 1.0443061228699715,
+      "learning_rate": 4.366802119379098e-06,
+      "loss": 0.2205,
+      "step": 361
+    },
+    {
+      "epoch": 1.0838323353293413,
+      "grad_norm": 1.029641825281261,
+      "learning_rate": 4.34348361616374e-06,
+      "loss": 0.3258,
+      "step": 362
+    },
+    {
+      "epoch": 1.0868263473053892,
+      "grad_norm": 1.2744361327782119,
+      "learning_rate": 4.3201796337833165e-06,
+      "loss": 0.3032,
+      "step": 363
+    },
+    {
+      "epoch": 1.089820359281437,
+      "grad_norm": 1.3371810159203008,
+      "learning_rate": 4.29689068767551e-06,
+      "loss": 0.3284,
+      "step": 364
+    },
+    {
+      "epoch": 1.092814371257485,
+      "grad_norm": 1.0955594058792184,
+      "learning_rate": 4.273617292945425e-06,
+      "loss": 0.2808,
+      "step": 365
+    },
+    {
+      "epoch": 1.095808383233533,
+      "grad_norm": 1.0919014745998543,
+      "learning_rate": 4.250359964354203e-06,
+      "loss": 0.2909,
+      "step": 366
+    },
+    {
+      "epoch": 1.098802395209581,
+      "grad_norm": 1.200658274962031,
+      "learning_rate": 4.227119216307637e-06,
+      "loss": 0.2905,
+      "step": 367
+    },
+    {
+      "epoch": 1.1017964071856288,
+      "grad_norm": 1.1072200995399941,
+      "learning_rate": 4.203895562844789e-06,
+      "loss": 0.3254,
+      "step": 368
+    },
+    {
+      "epoch": 1.1047904191616766,
+      "grad_norm": 1.2286218801660436,
+      "learning_rate": 4.18068951762662e-06,
+      "loss": 0.3356,
+      "step": 369
+    },
+    {
+      "epoch": 1.1077844311377245,
+      "grad_norm": 1.1308056719224104,
+      "learning_rate": 4.157501593924638e-06,
+      "loss": 0.2607,
+      "step": 370
+    },
+    {
+      "epoch": 1.1107784431137724,
+      "grad_norm": 1.079515583156294,
+      "learning_rate": 4.134332304609533e-06,
+      "loss": 0.3071,
+      "step": 371
+    },
+    {
+      "epoch": 1.1137724550898203,
+      "grad_norm": 0.9233257552162477,
+      "learning_rate": 4.111182162139844e-06,
+      "loss": 0.2852,
+      "step": 372
+    },
+    {
+      "epoch": 1.1167664670658684,
+      "grad_norm": 1.198953593245219,
+      "learning_rate": 4.088051678550617e-06,
+      "loss": 0.2696,
+      "step": 373
+    },
+    {
+      "epoch": 1.1197604790419162,
+      "grad_norm": 1.1638800718828535,
+      "learning_rate": 4.064941365442084e-06,
+      "loss": 0.2903,
+      "step": 374
+    },
+    {
+      "epoch": 1.122754491017964,
+      "grad_norm": 1.336320945178432,
+      "learning_rate": 4.041851733968348e-06,
+      "loss": 0.3341,
+      "step": 375
+    },
+    {
+      "epoch": 1.125748502994012,
+      "grad_norm": 1.1811419620615347,
+      "learning_rate": 4.018783294826071e-06,
+      "loss": 0.308,
+      "step": 376
+    },
+    {
+      "epoch": 1.1287425149700598,
+      "grad_norm": 1.2071411260971012,
+      "learning_rate": 3.995736558243186e-06,
+      "loss": 0.3251,
+      "step": 377
+    },
+    {
+      "epoch": 1.1317365269461077,
+      "grad_norm": 1.0985724591777641,
+      "learning_rate": 3.972712033967608e-06,
+      "loss": 0.2932,
+      "step": 378
+    },
+    {
+      "epoch": 1.1347305389221556,
+      "grad_norm": 1.089029675226207,
+      "learning_rate": 3.949710231255961e-06,
+      "loss": 0.3416,
+      "step": 379
+    },
+    {
+      "epoch": 1.1377245508982037,
+      "grad_norm": 1.3475588894868682,
+      "learning_rate": 3.926731658862307e-06,
+      "loss": 0.2963,
+      "step": 380
+    },
+    {
+      "epoch": 1.1407185628742516,
+      "grad_norm": 1.07510976700889,
+      "learning_rate": 3.903776825026912e-06,
+      "loss": 0.3292,
+      "step": 381
+    },
+    {
+      "epoch": 1.1437125748502994,
+      "grad_norm": 1.1192969825792705,
+      "learning_rate": 3.8808462374649805e-06,
+      "loss": 0.3015,
+      "step": 382
+    },
+    {
+      "epoch": 1.1467065868263473,
+      "grad_norm": 1.114818831237709,
+      "learning_rate": 3.857940403355444e-06,
+      "loss": 0.2622,
+      "step": 383
+    },
+    {
+      "epoch": 1.1497005988023952,
+      "grad_norm": 1.1288635189538672,
+      "learning_rate": 3.8350598293297345e-06,
+      "loss": 0.3186,
+      "step": 384
+    },
+    {
+      "epoch": 1.152694610778443,
+      "grad_norm": 1.1360812577733934,
+      "learning_rate": 3.8122050214605822e-06,
+      "loss": 0.2698,
+      "step": 385
+    },
+    {
+      "epoch": 1.1556886227544911,
+      "grad_norm": 1.087706872824366,
+      "learning_rate": 3.7893764852508207e-06,
+      "loss": 0.3042,
+      "step": 386
+    },
+    {
+      "epoch": 1.158682634730539,
+      "grad_norm": 1.0790194414424001,
+      "learning_rate": 3.766574725622208e-06,
+      "loss": 0.2608,
+      "step": 387
+    },
+    {
+      "epoch": 1.1616766467065869,
+      "grad_norm": 1.07108700952082,
+      "learning_rate": 3.7438002469042567e-06,
+      "loss": 0.289,
+      "step": 388
+    },
+    {
+      "epoch": 1.1646706586826348,
+      "grad_norm": 1.1840600963763515,
+      "learning_rate": 3.721053552823078e-06,
+      "loss": 0.2977,
+      "step": 389
+    },
+    {
+      "epoch": 1.1676646706586826,
+      "grad_norm": 1.3695321248438133,
+      "learning_rate": 3.698335146490246e-06,
+      "loss": 0.2901,
+      "step": 390
+    },
+    {
+      "epoch": 1.1706586826347305,
+      "grad_norm": 1.30271603898251,
+      "learning_rate": 3.675645530391665e-06,
+      "loss": 0.2791,
+      "step": 391
+    },
+    {
+      "epoch": 1.1736526946107784,
+      "grad_norm": 1.14029429064732,
+      "learning_rate": 3.652985206376455e-06,
+      "loss": 0.26,
+      "step": 392
+    },
+    {
+      "epoch": 1.1766467065868262,
+      "grad_norm": 0.9441624990771004,
+      "learning_rate": 3.630354675645853e-06,
+      "loss": 0.2476,
+      "step": 393
+    },
+    {
+      "epoch": 1.1796407185628743,
+      "grad_norm": 1.1706819335800416,
+      "learning_rate": 3.6077544387421293e-06,
+      "loss": 0.3052,
+      "step": 394
+    },
+    {
+      "epoch": 1.1826347305389222,
+      "grad_norm": 1.2494283649761313,
+      "learning_rate": 3.5851849955375177e-06,
+      "loss": 0.254,
+      "step": 395
+    },
+    {
+      "epoch": 1.18562874251497,
+      "grad_norm": 1.1698085070618776,
+      "learning_rate": 3.5626468452231534e-06,
+      "loss": 0.2715,
+      "step": 396
+    },
+    {
+      "epoch": 1.188622754491018,
+      "grad_norm": 1.1377250135594723,
+      "learning_rate": 3.540140486298035e-06,
+      "loss": 0.2229,
+      "step": 397
+    },
+    {
+      "epoch": 1.1916167664670658,
+      "grad_norm": 1.04455633928348,
+      "learning_rate": 3.517666416557999e-06,
+      "loss": 0.325,
+      "step": 398
+    },
+    {
+      "epoch": 1.1946107784431137,
+      "grad_norm": 1.157095563464243,
+      "learning_rate": 3.495225133084712e-06,
+      "loss": 0.2887,
+      "step": 399
+    },
+    {
+      "epoch": 1.1976047904191618,
+      "grad_norm": 1.06102342485893,
+      "learning_rate": 3.472817132234669e-06,
+      "loss": 0.3094,
+      "step": 400
+    },
+    {
+      "epoch": 1.1976047904191618,
+      "eval_loss": 0.3810786306858063,
+      "eval_runtime": 2.2941,
+      "eval_samples_per_second": 23.539,
+      "eval_steps_per_second": 6.103,
+      "step": 400
+    },
+    {
+      "epoch": 1.2005988023952097,
+      "grad_norm": 1.12203303577262,
+      "learning_rate": 3.4504429096282246e-06,
+      "loss": 0.3236,
+      "step": 401
+    },
+    {
+      "epoch": 1.2035928143712575,
+      "grad_norm": 1.171093041503917,
+      "learning_rate": 3.428102960138625e-06,
+      "loss": 0.3136,
+      "step": 402
+    },
+    {
+      "epoch": 1.2065868263473054,
+      "grad_norm": 1.185142137340155,
+      "learning_rate": 3.405797777881059e-06,
+      "loss": 0.3241,
+      "step": 403
+    },
+    {
+      "epoch": 1.2095808383233533,
+      "grad_norm": 1.181722195419124,
+      "learning_rate": 3.3835278562017405e-06,
+      "loss": 0.2659,
+      "step": 404
+    },
+    {
+      "epoch": 1.2125748502994012,
+      "grad_norm": 1.0770823794077098,
+      "learning_rate": 3.3612936876669834e-06,
+      "loss": 0.2723,
+      "step": 405
+    },
+    {
+      "epoch": 1.215568862275449,
+      "grad_norm": 1.0255059722341144,
+      "learning_rate": 3.3390957640523147e-06,
+      "loss": 0.2704,
+      "step": 406
+    },
+    {
+      "epoch": 1.218562874251497,
+      "grad_norm": 1.056026336164162,
+      "learning_rate": 3.3169345763315986e-06,
+      "loss": 0.2902,
+      "step": 407
+    },
+    {
+      "epoch": 1.221556886227545,
+      "grad_norm": 1.146592434105594,
+      "learning_rate": 3.29481061466617e-06,
+      "loss": 0.3102,
+      "step": 408
+    },
+    {
+      "epoch": 1.2245508982035929,
+      "grad_norm": 1.2557145604398288,
+      "learning_rate": 3.2727243683940045e-06,
+      "loss": 0.288,
+      "step": 409
+    },
+    {
+      "epoch": 1.2275449101796407,
+      "grad_norm": 1.1527052196242946,
+      "learning_rate": 3.2506763260188824e-06,
+      "loss": 0.3141,
+      "step": 410
+    },
+    {
+      "epoch": 1.2305389221556886,
+      "grad_norm": 1.030276371245447,
+      "learning_rate": 3.2286669751995905e-06,
+      "loss": 0.2899,
+      "step": 411
+    },
+    {
+      "epoch": 1.2335329341317365,
+      "grad_norm": 1.248196342346568,
+      "learning_rate": 3.2066968027391377e-06,
+      "loss": 0.2948,
+      "step": 412
+    },
+    {
+      "epoch": 1.2365269461077844,
+      "grad_norm": 1.035759748259409,
+      "learning_rate": 3.1847662945739833e-06,
+      "loss": 0.2787,
+      "step": 413
+    },
+    {
+      "epoch": 1.2395209580838324,
+      "grad_norm": 1.1589434521519926,
+      "learning_rate": 3.1628759357632943e-06,
+      "loss": 0.3038,
+      "step": 414
+    },
+    {
+      "epoch": 1.2425149700598803,
+      "grad_norm": 1.2385783521968639,
+      "learning_rate": 3.1410262104782086e-06,
+      "loss": 0.258,
+      "step": 415
+    },
+    {
+      "epoch": 1.2455089820359282,
+      "grad_norm": 1.1944231097312643,
+      "learning_rate": 3.119217601991139e-06,
+      "loss": 0.2924,
+      "step": 416
+    },
+    {
+      "epoch": 1.248502994011976,
+      "grad_norm": 1.2177751222326658,
+      "learning_rate": 3.0974505926650724e-06,
+      "loss": 0.2482,
+      "step": 417
+    },
+    {
+      "epoch": 1.251497005988024,
+      "grad_norm": 1.0747426373567288,
+      "learning_rate": 3.0757256639429027e-06,
+      "loss": 0.284,
+      "step": 418
+    },
+    {
+      "epoch": 1.2544910179640718,
+      "grad_norm": 1.2181323642678046,
+      "learning_rate": 3.0540432963367907e-06,
+      "loss": 0.3248,
+      "step": 419
+    },
+    {
+      "epoch": 1.2574850299401197,
+      "grad_norm": 1.1499316870188003,
+      "learning_rate": 3.032403969417523e-06,
+      "loss": 0.2799,
+      "step": 420
+    },
+    {
+      "epoch": 1.2604790419161676,
+      "grad_norm": 1.0821360162870375,
+      "learning_rate": 3.010808161803917e-06,
+      "loss": 0.2663,
+      "step": 421
+    },
+    {
+      "epoch": 1.2634730538922156,
+      "grad_norm": 1.0698556103140462,
+      "learning_rate": 2.9892563511522305e-06,
+      "loss": 0.3203,
+      "step": 422
+    },
+    {
+      "epoch": 1.2664670658682635,
+      "grad_norm": 1.1391479675057465,
+      "learning_rate": 2.9677490141455915e-06,
+      "loss": 0.2697,
+      "step": 423
+    },
+    {
+      "epoch": 1.2694610778443114,
+      "grad_norm": 1.1529139086610662,
+      "learning_rate": 2.946286626483463e-06,
+      "loss": 0.3197,
+      "step": 424
+    },
+    {
+      "epoch": 1.2724550898203593,
+      "grad_norm": 1.2848035674658973,
+      "learning_rate": 2.924869662871117e-06,
+      "loss": 0.2821,
+      "step": 425
+    },
+    {
+      "epoch": 1.2754491017964071,
+      "grad_norm": 1.1750018473428654,
+      "learning_rate": 2.903498597009136e-06,
+      "loss": 0.3018,
+      "step": 426
+    },
+    {
+      "epoch": 1.278443113772455,
+      "grad_norm": 1.1825982724555755,
+      "learning_rate": 2.8821739015829338e-06,
+      "loss": 0.2942,
+      "step": 427
+    },
+    {
+      "epoch": 1.281437125748503,
+      "grad_norm": 1.1176369943201498,
+      "learning_rate": 2.8608960482523058e-06,
+      "loss": 0.2829,
+      "step": 428
+    },
+    {
+      "epoch": 1.284431137724551,
+      "grad_norm": 1.1061269162790748,
+      "learning_rate": 2.839665507640992e-06,
+      "loss": 0.2794,
+      "step": 429
+    },
+    {
+      "epoch": 1.2874251497005988,
+      "grad_norm": 1.0402680654570222,
+      "learning_rate": 2.818482749326272e-06,
+      "loss": 0.3363,
+      "step": 430
+    },
+    {
+      "epoch": 1.2904191616766467,
+      "grad_norm": 1.037333106602708,
+      "learning_rate": 2.797348241828569e-06,
+      "loss": 0.2914,
+      "step": 431
+    },
+    {
+      "epoch": 1.2934131736526946,
+      "grad_norm": 1.0104584217893617,
+      "learning_rate": 2.776262452601104e-06,
+      "loss": 0.3152,
+      "step": 432
+    },
+    {
+      "epoch": 1.2964071856287425,
+      "grad_norm": 1.1452974573449428,
+      "learning_rate": 2.7552258480195348e-06,
+      "loss": 0.2808,
+      "step": 433
+    },
+    {
+      "epoch": 1.2994011976047903,
+      "grad_norm": 1.0352674843476175,
+      "learning_rate": 2.734238893371667e-06,
+      "loss": 0.2759,
+      "step": 434
+    },
+    {
+      "epoch": 1.3023952095808382,
+      "grad_norm": 1.0804202013027306,
+      "learning_rate": 2.7133020528471322e-06,
+      "loss": 0.2796,
+      "step": 435
+    },
+    {
+      "epoch": 1.3053892215568863,
+      "grad_norm": 1.031777465285058,
+      "learning_rate": 2.6924157895271563e-06,
+      "loss": 0.2932,
+      "step": 436
+    },
+    {
+      "epoch": 1.3083832335329342,
+      "grad_norm": 1.1251370217128986,
+      "learning_rate": 2.671580565374282e-06,
+      "loss": 0.3206,
+      "step": 437
+    },
+    {
+      "epoch": 1.311377245508982,
+      "grad_norm": 1.121973623762262,
+      "learning_rate": 2.6507968412221763e-06,
+      "loss": 0.245,
+      "step": 438
+    },
+    {
+      "epoch": 1.31437125748503,
+      "grad_norm": 1.2366495719901756,
+      "learning_rate": 2.6300650767654234e-06,
+      "loss": 0.2774,
+      "step": 439
+    },
+    {
+      "epoch": 1.3173652694610778,
+      "grad_norm": 1.0405636103901026,
+      "learning_rate": 2.6093857305493666e-06,
+      "loss": 0.2874,
+      "step": 440
+    },
+    {
+      "epoch": 1.3203592814371259,
+      "grad_norm": 1.1249670538429395,
+      "learning_rate": 2.588759259959962e-06,
+      "loss": 0.2672,
+      "step": 441
+    },
+    {
+      "epoch": 1.3233532934131738,
+      "grad_norm": 1.1300925768145054,
+      "learning_rate": 2.568186121213658e-06,
+      "loss": 0.3094,
+      "step": 442
+    },
+    {
+      "epoch": 1.3263473053892216,
+      "grad_norm": 1.124638256427979,
+      "learning_rate": 2.547666769347312e-06,
+      "loss": 0.2916,
+      "step": 443
+    },
+    {
+      "epoch": 1.3293413173652695,
+      "grad_norm": 1.089824148425882,
+      "learning_rate": 2.5272016582081236e-06,
+      "loss": 0.352,
+      "step": 444
+    },
+    {
+      "epoch": 1.3323353293413174,
+      "grad_norm": 1.1857492610651894,
+      "learning_rate": 2.5067912404435952e-06,
+      "loss": 0.3144,
+      "step": 445
+    },
+    {
+      "epoch": 1.3353293413173652,
+      "grad_norm": 1.1902274094112208,
+      "learning_rate": 2.486435967491516e-06,
+      "loss": 0.2772,
+      "step": 446
+    },
+    {
+      "epoch": 1.3383233532934131,
+      "grad_norm": 1.0747316993137852,
+      "learning_rate": 2.4661362895699903e-06,
+      "loss": 0.2669,
+      "step": 447
+    },
+    {
+      "epoch": 1.341317365269461,
+      "grad_norm": 1.1826900378314504,
+      "learning_rate": 2.445892655667462e-06,
+      "loss": 0.3702,
+      "step": 448
+    },
+    {
+      "epoch": 1.3443113772455089,
+      "grad_norm": 1.2477765874215443,
+      "learning_rate": 2.425705513532798e-06,
+      "loss": 0.3146,
+      "step": 449
+    },
+    {
+      "epoch": 1.347305389221557,
+      "grad_norm": 1.1137627651397157,
+      "learning_rate": 2.4055753096653795e-06,
+      "loss": 0.3229,
+      "step": 450
+    },
+    {
+      "epoch": 1.3502994011976048,
+      "grad_norm": 1.0310460818533735,
+      "learning_rate": 2.3855024893052286e-06,
+      "loss": 0.255,
+      "step": 451
+    },
+    {
+      "epoch": 1.3532934131736527,
+      "grad_norm": 1.105665344718147,
+      "learning_rate": 2.365487496423152e-06,
+      "loss": 0.2788,
+      "step": 452
+    },
+    {
+      "epoch": 1.3562874251497006,
+      "grad_norm": 1.0083266887349462,
+      "learning_rate": 2.3455307737109338e-06,
+      "loss": 0.2382,
+      "step": 453
+    },
+    {
+      "epoch": 1.3592814371257484,
+      "grad_norm": 1.052141942485767,
+      "learning_rate": 2.3256327625715345e-06,
+      "loss": 0.2467,
+      "step": 454
+    },
+    {
+      "epoch": 1.3622754491017965,
+      "grad_norm": 1.2207633529227289,
+      "learning_rate": 2.3057939031093346e-06,
+      "loss": 0.2775,
+      "step": 455
+    },
+    {
+      "epoch": 1.3652694610778444,
+      "grad_norm": 1.0212532373015724,
+      "learning_rate": 2.2860146341203936e-06,
+      "loss": 0.2665,
+      "step": 456
+    },
+    {
+      "epoch": 1.3682634730538923,
+      "grad_norm": 1.1758122387574494,
+      "learning_rate": 2.2662953930827546e-06,
+      "loss": 0.3545,
+      "step": 457
+    },
+    {
+      "epoch": 1.3712574850299402,
+      "grad_norm": 1.1560608892190745,
+      "learning_rate": 2.2466366161467528e-06,
+      "loss": 0.2849,
+      "step": 458
+    },
+    {
+      "epoch": 1.374251497005988,
+      "grad_norm": 1.2483949703794486,
+      "learning_rate": 2.227038738125385e-06,
+      "loss": 0.2599,
+      "step": 459
+    },
+    {
+      "epoch": 1.377245508982036,
+      "grad_norm": 1.3329958160752777,
+      "learning_rate": 2.207502192484685e-06,
+      "loss": 0.2807,
+      "step": 460
+    },
+    {
+      "epoch": 1.3802395209580838,
+      "grad_norm": 1.1126882443443804,
+      "learning_rate": 2.188027411334131e-06,
+      "loss": 0.2764,
+      "step": 461
+    },
+    {
+      "epoch": 1.3832335329341316,
+      "grad_norm": 1.1555362704216332,
+      "learning_rate": 2.1686148254171012e-06,
+      "loss": 0.2731,
+      "step": 462
+    },
+    {
+      "epoch": 1.3862275449101795,
+      "grad_norm": 1.312114577239909,
+      "learning_rate": 2.1492648641013305e-06,
+      "loss": 0.2697,
+      "step": 463
+    },
+    {
+      "epoch": 1.3892215568862276,
+      "grad_norm": 0.9811945604433543,
+      "learning_rate": 2.1299779553694323e-06,
+      "loss": 0.263,
+      "step": 464
+    },
+    {
+      "epoch": 1.3922155688622755,
+      "grad_norm": 1.2668920992087422,
+      "learning_rate": 2.1107545258094135e-06,
+      "loss": 0.2819,
+      "step": 465
+    },
+    {
+      "epoch": 1.3952095808383234,
+      "grad_norm": 1.1418587097119621,
+      "learning_rate": 2.0915950006052555e-06,
+      "loss": 0.2364,
+      "step": 466
+    },
+    {
+      "epoch": 1.3982035928143712,
+      "grad_norm": 1.087333910478332,
+      "learning_rate": 2.0724998035274947e-06,
+      "loss": 0.2747,
+      "step": 467
+    },
+    {
+      "epoch": 1.401197604790419,
+      "grad_norm": 1.1042050937339147,
+      "learning_rate": 2.053469356923865e-06,
+      "loss": 0.2255,
+      "step": 468
+    },
+    {
+      "epoch": 1.4041916167664672,
+      "grad_norm": 1.1242524790939759,
+      "learning_rate": 2.0345040817099433e-06,
+      "loss": 0.2872,
+      "step": 469
+    },
+    {
+      "epoch": 1.407185628742515,
+      "grad_norm": 1.0530875984257457,
+      "learning_rate": 2.0156043973598475e-06,
+      "loss": 0.2607,
+      "step": 470
+    },
+    {
+      "epoch": 1.410179640718563,
+      "grad_norm": 1.2386379832955134,
+      "learning_rate": 1.996770721896957e-06,
+      "loss": 0.2922,
+      "step": 471
+    },
+    {
+      "epoch": 1.4131736526946108,
+      "grad_norm": 1.257160643872115,
+      "learning_rate": 1.9780034718846653e-06,
+      "loss": 0.2785,
+      "step": 472
+    },
+    {
+      "epoch": 1.4161676646706587,
+      "grad_norm": 1.1001357825904647,
+      "learning_rate": 1.9593030624171683e-06,
+      "loss": 0.31,
+      "step": 473
+    },
+    {
+      "epoch": 1.4191616766467066,
+      "grad_norm": 1.2149766352742957,
+      "learning_rate": 1.9406699071102774e-06,
+      "loss": 0.3172,
+      "step": 474
+    },
+    {
+      "epoch": 1.4221556886227544,
+      "grad_norm": 1.0453498072882739,
+      "learning_rate": 1.9221044180922833e-06,
+      "loss": 0.3136,
+      "step": 475
+    },
+    {
+      "epoch": 1.4251497005988023,
+      "grad_norm": 1.1209530663049792,
+      "learning_rate": 1.9036070059948253e-06,
+      "loss": 0.2913,
+      "step": 476
+    },
+    {
+      "epoch": 1.4281437125748502,
+      "grad_norm": 1.2267672477264093,
+      "learning_rate": 1.885178079943823e-06,
+      "loss": 0.3144,
+      "step": 477
+    },
+    {
+      "epoch": 1.4311377245508983,
+      "grad_norm": 1.19944383115661,
+      "learning_rate": 1.866818047550419e-06,
+      "loss": 0.3637,
+      "step": 478
+    },
+    {
+      "epoch": 1.4341317365269461,
+      "grad_norm": 1.0507166416466567,
+      "learning_rate": 1.8485273149019655e-06,
+      "loss": 0.2774,
+      "step": 479
+    },
+    {
+      "epoch": 1.437125748502994,
+      "grad_norm": 1.1734280188599726,
+      "learning_rate": 1.8303062865530407e-06,
+      "loss": 0.2582,
+      "step": 480
+    },
+    {
+      "epoch": 1.4401197604790419,
+      "grad_norm": 1.1812234956084324,
+      "learning_rate": 1.8121553655165058e-06,
+      "loss": 0.2748,
+      "step": 481
+    },
+    {
+      "epoch": 1.4431137724550898,
+      "grad_norm": 1.1241288457054772,
+      "learning_rate": 1.7940749532545832e-06,
+      "loss": 0.3234,
+      "step": 482
+    },
+    {
+      "epoch": 1.4461077844311379,
+      "grad_norm": 1.3129647827086377,
+      "learning_rate": 1.7760654496699876e-06,
+      "loss": 0.2808,
+      "step": 483
+    },
+    {
+      "epoch": 1.4491017964071857,
+      "grad_norm": 1.113491894756196,
+      "learning_rate": 1.7581272530970666e-06,
+      "loss": 0.2978,
+      "step": 484
+    },
+    {
+      "epoch": 1.4520958083832336,
+      "grad_norm": 1.05025634285768,
+      "learning_rate": 1.7402607602930106e-06,
+      "loss": 0.2794,
+      "step": 485
+    },
+    {
+      "epoch": 1.4550898203592815,
+      "grad_norm": 1.2591051264404973,
+      "learning_rate": 1.7224663664290537e-06,
+      "loss": 0.34,
+      "step": 486
+    },
+    {
+      "epoch": 1.4580838323353293,
+      "grad_norm": 1.1004870982248354,
+      "learning_rate": 1.7047444650817518e-06,
+      "loss": 0.2858,
+      "step": 487
+    },
+    {
+      "epoch": 1.4610778443113772,
+      "grad_norm": 1.2374130925821987,
+      "learning_rate": 1.6870954482242707e-06,
+      "loss": 0.2795,
+      "step": 488
+    },
+    {
+      "epoch": 1.464071856287425,
+      "grad_norm": 1.0792120180781104,
+      "learning_rate": 1.669519706217711e-06,
+      "loss": 0.2396,
+      "step": 489
+    },
+    {
+      "epoch": 1.467065868263473,
+      "grad_norm": 1.1844546788411854,
+      "learning_rate": 1.652017627802487e-06,
+      "loss": 0.3035,
+      "step": 490
+    },
+    {
+      "epoch": 1.4700598802395208,
+      "grad_norm": 1.0058535081951945,
+      "learning_rate": 1.6345896000897122e-06,
+      "loss": 0.2635,
+      "step": 491
+    },
+    {
+      "epoch": 1.473053892215569,
+      "grad_norm": 0.9929511792123022,
+      "learning_rate": 1.6172360085526567e-06,
+      "loss": 0.2993,
+      "step": 492
+    },
+    {
+      "epoch": 1.4760479041916168,
+      "grad_norm": 0.9844192650605211,
+      "learning_rate": 1.5999572370182016e-06,
+      "loss": 0.2938,
+      "step": 493
+    },
+    {
+      "epoch": 1.4790419161676647,
+      "grad_norm": 1.2057793179232426,
+      "learning_rate": 1.5827536676583643e-06,
+      "loss": 0.2991,
+      "step": 494
+    },
+    {
+      "epoch": 1.4820359281437125,
+      "grad_norm": 1.1377142557272997,
+      "learning_rate": 1.5656256809818343e-06,
+      "loss": 0.2761,
+      "step": 495
+    },
+    {
+      "epoch": 1.4850299401197604,
+      "grad_norm": 1.228052229869402,
+      "learning_rate": 1.54857365582557e-06,
+      "loss": 0.25,
+      "step": 496
+    },
+    {
+      "epoch": 1.4880239520958085,
+      "grad_norm": 1.147381047321433,
+      "learning_rate": 1.5315979693464039e-06,
+      "loss": 0.2994,
+      "step": 497
+    },
+    {
+      "epoch": 1.4910179640718564,
+      "grad_norm": 1.266804321002084,
+      "learning_rate": 1.5146989970127158e-06,
+      "loss": 0.2483,
+      "step": 498
+    },
+    {
+      "epoch": 1.4940119760479043,
+      "grad_norm": 1.1828347040447706,
+      "learning_rate": 1.4978771125961177e-06,
+      "loss": 0.2983,
+      "step": 499
+    },
+    {
+      "epoch": 1.4970059880239521,
+      "grad_norm": 1.0389099948255578,
+      "learning_rate": 1.4811326881631937e-06,
+      "loss": 0.2891,
+      "step": 500
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 1.0858972961766649,
+      "learning_rate": 1.4644660940672628e-06,
+      "loss": 0.2866,
+      "step": 501
+    },
+    {
+      "epoch": 1.5029940119760479,
+      "grad_norm": 1.1390648564203294,
+      "learning_rate": 1.4478776989401949e-06,
+      "loss": 0.303,
+      "step": 502
+    },
+    {
+      "epoch": 1.5059880239520957,
+      "grad_norm": 1.173938774623967,
+      "learning_rate": 1.4313678696842559e-06,
+      "loss": 0.2843,
+      "step": 503
+    },
+    {
+      "epoch": 1.5089820359281436,
+      "grad_norm": 1.1111740866903577,
+      "learning_rate": 1.4149369714639856e-06,
+      "loss": 0.2734,
+      "step": 504
+    },
+    {
+      "epoch": 1.5119760479041915,
+      "grad_norm": 1.162914058056962,
+      "learning_rate": 1.3985853676981316e-06,
+      "loss": 0.2898,
+      "step": 505
+    },
+    {
+      "epoch": 1.5149700598802394,
+      "grad_norm": 1.1623786598592987,
+      "learning_rate": 1.3823134200516043e-06,
+      "loss": 0.2742,
+      "step": 506
+    },
+    {
+      "epoch": 1.5179640718562875,
+      "grad_norm": 1.276961093425585,
+      "learning_rate": 1.366121488427481e-06,
+      "loss": 0.2771,
+      "step": 507
+    },
+    {
+      "epoch": 1.5209580838323353,
+      "grad_norm": 1.2311197822766862,
+      "learning_rate": 1.3500099309590397e-06,
+      "loss": 0.2954,
+      "step": 508
+    },
+    {
+      "epoch": 1.5239520958083832,
+      "grad_norm": 1.1878683954478704,
+      "learning_rate": 1.3339791040018479e-06,
+      "loss": 0.3327,
+      "step": 509
+    },
+    {
+      "epoch": 1.5269461077844313,
+      "grad_norm": 1.0638818666660017,
+      "learning_rate": 1.3180293621258694e-06,
+      "loss": 0.2798,
+      "step": 510
+    },
+    {
+      "epoch": 1.5299401197604792,
+      "grad_norm": 1.1678324378348284,
+      "learning_rate": 1.3021610581076316e-06,
+      "loss": 0.2374,
+      "step": 511
+    },
+    {
+      "epoch": 1.532934131736527,
+      "grad_norm": 1.3071648660018567,
+      "learning_rate": 1.2863745429224145e-06,
+      "loss": 0.2873,
+      "step": 512
+    },
+    {
+      "epoch": 1.535928143712575,
+      "grad_norm": 1.1256664532984102,
+      "learning_rate": 1.270670165736499e-06,
+      "loss": 0.2888,
+      "step": 513
+    },
+    {
+      "epoch": 1.5389221556886228,
+      "grad_norm": 1.3471690357271124,
+      "learning_rate": 1.2550482738994284e-06,
+      "loss": 0.3429,
+      "step": 514
+    },
+    {
+      "epoch": 1.5419161676646707,
+      "grad_norm": 1.1361366232878214,
+      "learning_rate": 1.239509212936343e-06,
+      "loss": 0.2827,
+      "step": 515
+    },
+    {
+      "epoch": 1.5449101796407185,
+      "grad_norm": 1.1886635703543298,
+      "learning_rate": 1.22405332654032e-06,
+      "loss": 0.2733,
+      "step": 516
+    },
+    {
+      "epoch": 1.5479041916167664,
+      "grad_norm": 0.9853710044335633,
+      "learning_rate": 1.2086809565647877e-06,
+      "loss": 0.2685,
+      "step": 517
+    },
+    {
+      "epoch": 1.5508982035928143,
+      "grad_norm": 1.1579914747515045,
+      "learning_rate": 1.1933924430159571e-06,
+      "loss": 0.281,
+      "step": 518
+    },
+    {
+      "epoch": 1.5538922155688621,
+      "grad_norm": 1.1518792387525134,
+      "learning_rate": 1.1781881240452958e-06,
+      "loss": 0.275,
+      "step": 519
+    },
+    {
+      "epoch": 1.55688622754491,
+      "grad_norm": 1.1457076754263529,
+      "learning_rate": 1.1630683359420653e-06,
+      "loss": 0.2931,
+      "step": 520
+    },
+    {
+      "epoch": 1.5598802395209581,
+      "grad_norm": 1.1968696932996832,
+      "learning_rate": 1.1480334131258626e-06,
+      "loss": 0.2252,
+      "step": 521
+    },
+    {
+      "epoch": 1.562874251497006,
+      "grad_norm": 1.258673507619603,
+      "learning_rate": 1.1330836881392405e-06,
+      "loss": 0.2714,
+      "step": 522
+    },
+    {
+      "epoch": 1.5658682634730539,
+      "grad_norm": 1.1183900349976967,
+      "learning_rate": 1.11821949164034e-06,
+      "loss": 0.2928,
+      "step": 523
+    },
+    {
+      "epoch": 1.568862275449102,
+      "grad_norm": 1.2467232265519301,
+      "learning_rate": 1.103441152395588e-06,
+      "loss": 0.3451,
+      "step": 524
+    },
+    {
+      "epoch": 1.5718562874251498,
+      "grad_norm": 1.1823094237229124,
+      "learning_rate": 1.088748997272414e-06,
+      "loss": 0.2969,
+      "step": 525
+    },
+    {
+      "epoch": 1.5748502994011977,
+      "grad_norm": 1.1274647456362425,
+      "learning_rate": 1.0741433512320316e-06,
+      "loss": 0.2333,
+      "step": 526
+    },
+    {
+      "epoch": 1.5778443113772456,
+      "grad_norm": 1.279211326867482,
+      "learning_rate": 1.0596245373222424e-06,
+      "loss": 0.2677,
+      "step": 527
+    },
+    {
+      "epoch": 1.5808383233532934,
+      "grad_norm": 1.1280493108443623,
+      "learning_rate": 1.045192876670298e-06,
+      "loss": 0.3069,
+      "step": 528
+    },
+    {
+      "epoch": 1.5838323353293413,
+      "grad_norm": 1.081967266611651,
+      "learning_rate": 1.0308486884757868e-06,
+      "loss": 0.3197,
+      "step": 529
+    },
+    {
+      "epoch": 1.5868263473053892,
+      "grad_norm": 1.0311128487985322,
+      "learning_rate": 1.0165922900035886e-06,
+      "loss": 0.291,
+      "step": 530
+    },
+    {
+      "epoch": 1.589820359281437,
+      "grad_norm": 1.1227288920854246,
+      "learning_rate": 1.0024239965768417e-06,
+      "loss": 0.2781,
+      "step": 531
+    },
+    {
+      "epoch": 1.592814371257485,
+      "grad_norm": 1.194456565248844,
+      "learning_rate": 9.883441215699824e-07,
+      "loss": 0.2181,
+      "step": 532
+    },
+    {
+      "epoch": 1.5958083832335328,
+      "grad_norm": 1.3757290606721246,
+      "learning_rate": 9.74352976401805e-07,
+      "loss": 0.3497,
+      "step": 533
+    },
+    {
+      "epoch": 1.5988023952095807,
+      "grad_norm": 1.3079982025930916,
+      "learning_rate": 9.604508705285765e-07,
+      "loss": 0.2908,
+      "step": 534
+    },
+    {
+      "epoch": 1.6017964071856288,
+      "grad_norm": 1.1924676398154075,
+      "learning_rate": 9.466381114371942e-07,
+      "loss": 0.2871,
+      "step": 535
+    },
+    {
+      "epoch": 1.6047904191616766,
+      "grad_norm": 1.1201405919102474,
+      "learning_rate": 9.329150046383773e-07,
+      "loss": 0.2781,
+      "step": 536
+    },
+    {
+      "epoch": 1.6077844311377245,
+      "grad_norm": 1.229711506202082,
+      "learning_rate": 9.192818536599213e-07,
+      "loss": 0.2941,
+      "step": 537
+    },
+    {
+      "epoch": 1.6107784431137726,
+      "grad_norm": 1.1088136087046185,
+      "learning_rate": 9.057389600399719e-07,
+      "loss": 0.2967,
+      "step": 538
+    },
+    {
+      "epoch": 1.6137724550898205,
+      "grad_norm": 1.0572404018380674,
+      "learning_rate": 8.922866233203681e-07,
+      "loss": 0.2615,
+      "step": 539
+    },
+    {
+      "epoch": 1.6167664670658684,
+      "grad_norm": 0.9423624755177453,
+      "learning_rate": 8.789251410400024e-07,
+      "loss": 0.216,
+      "step": 540
+    },
+    {
+      "epoch": 1.6197604790419162,
+      "grad_norm": 0.9340956179804273,
+      "learning_rate": 8.65654808728259e-07,
+      "loss": 0.277,
+      "step": 541
+    },
+    {
+      "epoch": 1.622754491017964,
+      "grad_norm": 1.4140114569532536,
+      "learning_rate": 8.524759198984567e-07,
+      "loss": 0.2594,
+      "step": 542
+    },
+    {
+      "epoch": 1.625748502994012,
+      "grad_norm": 1.1165641261607384,
+      "learning_rate": 8.393887660413719e-07,
+      "loss": 0.2835,
+      "step": 543
+    },
+    {
+      "epoch": 1.6287425149700598,
+      "grad_norm": 1.3330092034594856,
+      "learning_rate": 8.263936366187825e-07,
+      "loss": 0.334,
+      "step": 544
+    },
+    {
+      "epoch": 1.6317365269461077,
+      "grad_norm": 1.1337392054952442,
+      "learning_rate": 8.134908190570723e-07,
+      "loss": 0.3017,
+      "step": 545
+    },
+    {
+      "epoch": 1.6347305389221556,
+      "grad_norm": 1.0118426440711963,
+      "learning_rate": 8.006805987408705e-07,
+      "loss": 0.2118,
+      "step": 546
+    },
+    {
+      "epoch": 1.6377245508982035,
+      "grad_norm": 1.2024351673754978,
+      "learning_rate": 7.879632590067354e-07,
+      "loss": 0.2953,
+      "step": 547
+    },
+    {
+      "epoch": 1.6407185628742516,
+      "grad_norm": 1.2458713630955813,
+      "learning_rate": 7.753390811368972e-07,
+      "loss": 0.321,
+      "step": 548
+    },
+    {
+      "epoch": 1.6437125748502994,
+      "grad_norm": 1.1245529827343188,
+      "learning_rate": 7.628083443530287e-07,
+      "loss": 0.2761,
+      "step": 549
+    },
+    {
+      "epoch": 1.6467065868263473,
+      "grad_norm": 1.3223968958023358,
+      "learning_rate": 7.503713258100726e-07,
+      "loss": 0.3155,
+      "step": 550
+    },
+    {
+      "epoch": 1.6497005988023952,
+      "grad_norm": 1.0855870373166687,
+      "learning_rate": 7.380283005901084e-07,
+      "loss": 0.2899,
+      "step": 551
+    },
+    {
+      "epoch": 1.6526946107784433,
+      "grad_norm": 1.0802889579953756,
+      "learning_rate": 7.257795416962754e-07,
+      "loss": 0.2632,
+      "step": 552
+    },
+    {
+      "epoch": 1.6556886227544911,
+      "grad_norm": 1.1308722432182747,
+      "learning_rate": 7.136253200467231e-07,
+      "loss": 0.2875,
+      "step": 553
+    },
+    {
+      "epoch": 1.658682634730539,
+      "grad_norm": 1.1407372443180035,
+      "learning_rate": 7.015659044686307e-07,
+      "loss": 0.2736,
+      "step": 554
+    },
+    {
+      "epoch": 1.6616766467065869,
+      "grad_norm": 1.1551363344162962,
+      "learning_rate": 6.896015616922535e-07,
+      "loss": 0.2828,
+      "step": 555
+    },
+    {
+      "epoch": 1.6646706586826348,
+      "grad_norm": 1.044387355360179,
+      "learning_rate": 6.777325563450282e-07,
+      "loss": 0.3179,
+      "step": 556
+    },
+    {
+      "epoch": 1.6676646706586826,
+      "grad_norm": 1.1114438060031078,
+      "learning_rate": 6.659591509457125e-07,
+      "loss": 0.2764,
+      "step": 557
+    },
+    {
+      "epoch": 1.6706586826347305,
+      "grad_norm": 1.235696527352362,
+      "learning_rate": 6.542816058985896e-07,
+      "loss": 0.2944,
+      "step": 558
+    },
+    {
+      "epoch": 1.6736526946107784,
+      "grad_norm": 1.2177079996030908,
+      "learning_rate": 6.427001794876974e-07,
+      "loss": 0.2842,
+      "step": 559
+    },
+    {
+      "epoch": 1.6766467065868262,
+      "grad_norm": 1.0241332061694846,
+      "learning_rate": 6.312151278711237e-07,
+      "loss": 0.2982,
+      "step": 560
+    },
+    {
+      "epoch": 1.6796407185628741,
+      "grad_norm": 1.0529041030567587,
+      "learning_rate": 6.198267050753387e-07,
+      "loss": 0.2801,
+      "step": 561
+    },
+    {
+      "epoch": 1.6826347305389222,
+      "grad_norm": 1.1542211410143277,
+      "learning_rate": 6.085351629895736e-07,
+      "loss": 0.2824,
+      "step": 562
+    },
+    {
+      "epoch": 1.68562874251497,
+      "grad_norm": 1.2553813608934878,
+      "learning_rate": 5.973407513602514e-07,
+      "loss": 0.2566,
+      "step": 563
+    },
+    {
+      "epoch": 1.688622754491018,
+      "grad_norm": 1.1975871901332356,
+      "learning_rate": 5.862437177854629e-07,
+      "loss": 0.3033,
+      "step": 564
+    },
+    {
+      "epoch": 1.6916167664670658,
+      "grad_norm": 1.3331567624692788,
+      "learning_rate": 5.752443077094927e-07,
+      "loss": 0.2539,
+      "step": 565
+    },
+    {
+      "epoch": 1.694610778443114,
+      "grad_norm": 1.1371281266712578,
+      "learning_rate": 5.643427644173838e-07,
+      "loss": 0.2852,
+      "step": 566
+    },
+    {
+      "epoch": 1.6976047904191618,
+      "grad_norm": 1.0004104418942674,
+      "learning_rate": 5.535393290295643e-07,
+      "loss": 0.2628,
+      "step": 567
+    },
+    {
+      "epoch": 1.7005988023952097,
+      "grad_norm": 1.1787651247176714,
+      "learning_rate": 5.428342404965076e-07,
+      "loss": 0.2616,
+      "step": 568
+    },
+    {
+      "epoch": 1.7035928143712575,
+      "grad_norm": 1.2064618809836718,
+      "learning_rate": 5.322277355934557e-07,
+      "loss": 0.3215,
+      "step": 569
+    },
+    {
+      "epoch": 1.7065868263473054,
+      "grad_norm": 1.0888420519330986,
+      "learning_rate": 5.217200489151714e-07,
+      "loss": 0.2853,
+      "step": 570
+    },
+    {
+      "epoch": 1.7095808383233533,
+      "grad_norm": 1.0417964340089696,
+      "learning_rate": 5.113114128707592e-07,
+      "loss": 0.2324,
+      "step": 571
+    },
+    {
+      "epoch": 1.7125748502994012,
+      "grad_norm": 1.2051994409677669,
+      "learning_rate": 5.010020576785174e-07,
+      "loss": 0.3069,
+      "step": 572
+    },
+    {
+      "epoch": 1.715568862275449,
+      "grad_norm": 1.0446782589183807,
+      "learning_rate": 4.907922113608532e-07,
+      "loss": 0.2333,
+      "step": 573
+    },
+    {
+      "epoch": 1.718562874251497,
+      "grad_norm": 1.1447300778855605,
+      "learning_rate": 4.806820997392325e-07,
+      "loss": 0.2797,
+      "step": 574
+    },
+    {
+      "epoch": 1.7215568862275448,
+      "grad_norm": 1.0464475349884184,
+      "learning_rate": 4.7067194642919036e-07,
+      "loss": 0.2757,
+      "step": 575
+    },
+    {
+      "epoch": 1.7245508982035929,
+      "grad_norm": 1.2565762573571595,
+      "learning_rate": 4.607619728353818e-07,
+      "loss": 0.2689,
+      "step": 576
+    },
+    {
+      "epoch": 1.7275449101796407,
+      "grad_norm": 0.9898153758344741,
+      "learning_rate": 4.50952398146689e-07,
+      "loss": 0.2724,
+      "step": 577
+    },
+    {
+      "epoch": 1.7305389221556886,
+      "grad_norm": 1.0948862540321544,
+      "learning_rate": 4.4124343933136525e-07,
+      "loss": 0.2752,
+      "step": 578
+    },
+    {
+      "epoch": 1.7335329341317365,
+      "grad_norm": 1.009609811736011,
+      "learning_rate": 4.3163531113224466e-07,
+      "loss": 0.2484,
+      "step": 579
+    },
+    {
+      "epoch": 1.7365269461077846,
+      "grad_norm": 1.1388619535240432,
+      "learning_rate": 4.221282260619891e-07,
+      "loss": 0.2923,
+      "step": 580
+    },
+    {
+      "epoch": 1.7395209580838324,
+      "grad_norm": 1.0453890642177948,
+      "learning_rate": 4.127223943983849e-07,
+      "loss": 0.2615,
+      "step": 581
+    },
+    {
+      "epoch": 1.7425149700598803,
+      "grad_norm": 1.0590166016845914,
+      "learning_rate": 4.03418024179697e-07,
+      "loss": 0.2602,
+      "step": 582
+    },
+    {
+      "epoch": 1.7455089820359282,
+      "grad_norm": 1.099857421551844,
+      "learning_rate": 3.9421532120006544e-07,
+      "loss": 0.2543,
+      "step": 583
+    },
+    {
+      "epoch": 1.748502994011976,
+      "grad_norm": 1.0682149850946763,
+      "learning_rate": 3.851144890049535e-07,
+      "loss": 0.2686,
+      "step": 584
+    },
+    {
+      "epoch": 1.751497005988024,
+      "grad_norm": 1.2018283955979425,
+      "learning_rate": 3.761157288866418e-07,
+      "loss": 0.3353,
+      "step": 585
+    },
+    {
+      "epoch": 1.7544910179640718,
+      "grad_norm": 1.1277460529462433,
+      "learning_rate": 3.672192398797858e-07,
+      "loss": 0.2779,
+      "step": 586
+    },
+    {
+      "epoch": 1.7574850299401197,
+      "grad_norm": 1.166308082647788,
+      "learning_rate": 3.58425218757002e-07,
+      "loss": 0.2673,
+      "step": 587
+    },
+    {
+      "epoch": 1.7604790419161676,
+      "grad_norm": 1.0978045999301895,
+      "learning_rate": 3.497338600245254e-07,
+      "loss": 0.265,
+      "step": 588
+    },
+    {
+      "epoch": 1.7634730538922154,
+      "grad_norm": 1.0789310685581852,
+      "learning_rate": 3.4114535591790233e-07,
+      "loss": 0.257,
+      "step": 589
+    },
+    {
+      "epoch": 1.7664670658682635,
+      "grad_norm": 1.2402395600516338,
+      "learning_rate": 3.326598963977395e-07,
+      "loss": 0.2568,
+      "step": 590
+    },
+    {
+      "epoch": 1.7694610778443114,
+      "grad_norm": 1.1346008761053423,
+      "learning_rate": 3.242776691455013e-07,
+      "loss": 0.2972,
+      "step": 591
+    },
+    {
+      "epoch": 1.7724550898203593,
+      "grad_norm": 0.9307027403651428,
+      "learning_rate": 3.159988595593616e-07,
+      "loss": 0.2483,
+      "step": 592
+    },
+    {
+      "epoch": 1.7754491017964071,
+      "grad_norm": 1.097893673505579,
+      "learning_rate": 3.078236507501015e-07,
+      "loss": 0.2422,
+      "step": 593
+    },
+    {
+      "epoch": 1.7784431137724552,
+      "grad_norm": 1.099719258137491,
+      "learning_rate": 2.9975222353705757e-07,
+      "loss": 0.2972,
+      "step": 594
+    },
+    {
+      "epoch": 1.781437125748503,
+      "grad_norm": 1.1066544643170122,
+      "learning_rate": 2.917847564441256e-07,
+      "loss": 0.3143,
+      "step": 595
+    },
+    {
+      "epoch": 1.784431137724551,
+      "grad_norm": 1.1422520443318036,
+      "learning_rate": 2.839214256958106e-07,
+      "loss": 0.2761,
+      "step": 596
+    },
+    {
+      "epoch": 1.7874251497005988,
+      "grad_norm": 1.2600305029810888,
+      "learning_rate": 2.7616240521332884e-07,
+      "loss": 0.2632,
+      "step": 597
+    },
+    {
+      "epoch": 1.7904191616766467,
+      "grad_norm": 0.997598729167516,
+      "learning_rate": 2.6850786661076047e-07,
+      "loss": 0.2781,
+      "step": 598
+    },
+    {
+      "epoch": 1.7934131736526946,
+      "grad_norm": 1.0972039239845834,
+      "learning_rate": 2.6095797919125533e-07,
+      "loss": 0.2582,
+      "step": 599
+    },
+    {
+      "epoch": 1.7964071856287425,
+      "grad_norm": 1.0501639196968615,
+      "learning_rate": 2.5351290994328703e-07,
+      "loss": 0.2983,
+      "step": 600
+    },
+    {
+      "epoch": 1.7964071856287425,
+      "eval_loss": 0.37384727597236633,
+      "eval_runtime": 2.2853,
+      "eval_samples_per_second": 23.63,
+      "eval_steps_per_second": 6.126,
+      "step": 600
+    },
+    {
+      "epoch": 1.7994011976047903,
+      "grad_norm": 1.0351299997916499,
+      "learning_rate": 2.4617282353696093e-07,
+      "loss": 0.2495,
+      "step": 601
+    },
+    {
+      "epoch": 1.8023952095808382,
+      "grad_norm": 1.223547417035152,
+      "learning_rate": 2.3893788232036807e-07,
+      "loss": 0.3499,
+      "step": 602
+    },
+    {
+      "epoch": 1.805389221556886,
+      "grad_norm": 1.1024244427389678,
+      "learning_rate": 2.318082463160032e-07,
+      "loss": 0.3062,
+      "step": 603
+    },
+    {
+      "epoch": 1.8083832335329342,
+      "grad_norm": 0.9593982963823812,
+      "learning_rate": 2.2478407321721295e-07,
+      "loss": 0.2618,
+      "step": 604
+    },
+    {
+      "epoch": 1.811377245508982,
+      "grad_norm": 1.0531965673132127,
+      "learning_rate": 2.1786551838471892e-07,
+      "loss": 0.2701,
+      "step": 605
+    },
+    {
+      "epoch": 1.81437125748503,
+      "grad_norm": 1.2554375606368706,
+      "learning_rate": 2.1105273484317402e-07,
+      "loss": 0.2358,
+      "step": 606
+    },
+    {
+      "epoch": 1.8173652694610778,
+      "grad_norm": 1.0524001021397076,
+      "learning_rate": 2.043458732777831e-07,
+      "loss": 0.2453,
+      "step": 607
+    },
+    {
+      "epoch": 1.8203592814371259,
+      "grad_norm": 1.199369907179867,
+      "learning_rate": 1.9774508203096843e-07,
+      "loss": 0.2841,
+      "step": 608
+    },
+    {
+      "epoch": 1.8233532934131738,
+      "grad_norm": 1.0397890244394785,
+      "learning_rate": 1.9125050709908388e-07,
+      "loss": 0.2474,
+      "step": 609
+    },
+    {
+      "epoch": 1.8263473053892216,
+      "grad_norm": 1.1490564686055238,
+      "learning_rate": 1.8486229212919482e-07,
+      "loss": 0.2894,
+      "step": 610
+    },
+    {
+      "epoch": 1.8293413173652695,
+      "grad_norm": 1.0660953211554132,
+      "learning_rate": 1.7858057841589281e-07,
+      "loss": 0.2753,
+      "step": 611
+    },
+    {
+      "epoch": 1.8323353293413174,
+      "grad_norm": 1.2840819770381964,
+      "learning_rate": 1.7240550489817652e-07,
+      "loss": 0.2998,
+      "step": 612
+    },
+    {
+      "epoch": 1.8353293413173652,
+      "grad_norm": 1.2838773699186365,
+      "learning_rate": 1.66337208156373e-07,
+      "loss": 0.2709,
+      "step": 613
+    },
+    {
+      "epoch": 1.8383233532934131,
+      "grad_norm": 1.1566312716867309,
+      "learning_rate": 1.6037582240912175e-07,
+      "loss": 0.2872,
+      "step": 614
+    },
+    {
+      "epoch": 1.841317365269461,
+      "grad_norm": 1.1109113028726003,
+      "learning_rate": 1.5452147951040165e-07,
+      "loss": 0.2859,
+      "step": 615
+    },
+    {
+      "epoch": 1.8443113772455089,
+      "grad_norm": 1.33828947447816,
+      "learning_rate": 1.4877430894662037e-07,
+      "loss": 0.3318,
+      "step": 616
+    },
+    {
+      "epoch": 1.8473053892215567,
+      "grad_norm": 1.1348765073710838,
+      "learning_rate": 1.4313443783374405e-07,
+      "loss": 0.2577,
+      "step": 617
+    },
+    {
+      "epoch": 1.8502994011976048,
+      "grad_norm": 1.1584762643735678,
+      "learning_rate": 1.3760199091449045e-07,
+      "loss": 0.2432,
+      "step": 618
+    },
+    {
+      "epoch": 1.8532934131736527,
+      "grad_norm": 1.0012712527017773,
+      "learning_rate": 1.3217709055556638e-07,
+      "loss": 0.2393,
+      "step": 619
+    },
+    {
+      "epoch": 1.8562874251497006,
+      "grad_norm": 1.097933155229744,
+      "learning_rate": 1.268598567449647e-07,
+      "loss": 0.2748,
+      "step": 620
+    },
+    {
+      "epoch": 1.8592814371257484,
+      "grad_norm": 1.2667620861179836,
+      "learning_rate": 1.2165040708930763e-07,
+      "loss": 0.2799,
+      "step": 621
+    },
+    {
+      "epoch": 1.8622754491017965,
+      "grad_norm": 0.9829508979611353,
+      "learning_rate": 1.1654885681124661e-07,
+      "loss": 0.2651,
+      "step": 622
+    },
+    {
+      "epoch": 1.8652694610778444,
+      "grad_norm": 1.181542123359679,
+      "learning_rate": 1.1155531874691372e-07,
+      "loss": 0.3149,
+      "step": 623
+    },
+    {
+      "epoch": 1.8682634730538923,
+      "grad_norm": 1.0797619877117075,
+      "learning_rate": 1.0666990334342708e-07,
+      "loss": 0.2492,
+      "step": 624
+    },
+    {
+      "epoch": 1.8712574850299402,
+      "grad_norm": 1.0895760923433364,
+      "learning_rate": 1.0189271865644445e-07,
+      "loss": 0.2783,
+      "step": 625
+    },
+    {
+      "epoch": 1.874251497005988,
+      "grad_norm": 1.4375826103032592,
+      "learning_rate": 9.722387034777847e-08,
+      "loss": 0.3303,
+      "step": 626
+    },
+    {
+      "epoch": 1.877245508982036,
+      "grad_norm": 1.1741164553978873,
+      "learning_rate": 9.266346168305518e-08,
+      "loss": 0.3286,
+      "step": 627
+    },
+    {
+      "epoch": 1.8802395209580838,
+      "grad_norm": 1.0460672015084689,
+      "learning_rate": 8.821159352943142e-08,
+      "loss": 0.2694,
+      "step": 628
+    },
+    {
+      "epoch": 1.8832335329341316,
+      "grad_norm": 1.1486974063864275,
+      "learning_rate": 8.38683643533661e-08,
+      "loss": 0.2851,
+      "step": 629
+    },
+    {
+      "epoch": 1.8862275449101795,
+      "grad_norm": 1.210891212928368,
+      "learning_rate": 7.963387021843683e-08,
+      "loss": 0.2914,
+      "step": 630
+    },
+    {
+      "epoch": 1.8892215568862274,
+      "grad_norm": 1.1279024237744695,
+      "learning_rate": 7.550820478322285e-08,
+      "loss": 0.3081,
+      "step": 631
+    },
+    {
+      "epoch": 1.8922155688622755,
+      "grad_norm": 1.0675114721678558,
+      "learning_rate": 7.149145929922607e-08,
+      "loss": 0.2375,
+      "step": 632
+    },
+    {
+      "epoch": 1.8952095808383234,
+      "grad_norm": 1.165125413843238,
+      "learning_rate": 6.758372260885714e-08,
+      "loss": 0.2933,
+      "step": 633
+    },
+    {
+      "epoch": 1.8982035928143712,
+      "grad_norm": 1.2767226603081276,
+      "learning_rate": 6.378508114346982e-08,
+      "loss": 0.2699,
+      "step": 634
+    },
+    {
+      "epoch": 1.9011976047904193,
+      "grad_norm": 1.3193499505864115,
+      "learning_rate": 6.009561892144744e-08,
+      "loss": 0.2633,
+      "step": 635
+    },
+    {
+      "epoch": 1.9041916167664672,
+      "grad_norm": 1.145028006477466,
+      "learning_rate": 5.651541754634726e-08,
+      "loss": 0.2571,
+      "step": 636
+    },
+    {
+      "epoch": 1.907185628742515,
+      "grad_norm": 1.1844047732432987,
+      "learning_rate": 5.304455620509297e-08,
+      "loss": 0.288,
+      "step": 637
+    },
+    {
+      "epoch": 1.910179640718563,
+      "grad_norm": 1.1588827018880847,
+      "learning_rate": 4.968311166622553e-08,
+      "loss": 0.2823,
+      "step": 638
+    },
+    {
+      "epoch": 1.9131736526946108,
+      "grad_norm": 1.2913426741033982,
+      "learning_rate": 4.643115827820399e-08,
+      "loss": 0.2503,
+      "step": 639
+    },
+    {
+      "epoch": 1.9161676646706587,
+      "grad_norm": 1.2439237791245041,
+      "learning_rate": 4.328876796776071e-08,
+      "loss": 0.315,
+      "step": 640
+    },
+    {
+      "epoch": 1.9191616766467066,
+      "grad_norm": 1.2149792253415745,
+      "learning_rate": 4.0256010238310936e-08,
+      "loss": 0.2664,
+      "step": 641
+    },
+    {
+      "epoch": 1.9221556886227544,
+      "grad_norm": 1.0987835132213755,
+      "learning_rate": 3.733295216841626e-08,
+      "loss": 0.2954,
+      "step": 642
+    },
+    {
+      "epoch": 1.9251497005988023,
+      "grad_norm": 1.1419793240442702,
+      "learning_rate": 3.451965841029914e-08,
+      "loss": 0.2768,
+      "step": 643
+    },
+    {
+      "epoch": 1.9281437125748502,
+      "grad_norm": 1.0349651310776806,
+      "learning_rate": 3.181619118841517e-08,
+      "loss": 0.3183,
+      "step": 644
+    },
+    {
+      "epoch": 1.931137724550898,
+      "grad_norm": 1.2879336442522813,
+      "learning_rate": 2.9222610298074717e-08,
+      "loss": 0.2541,
+      "step": 645
+    },
+    {
+      "epoch": 1.9341317365269461,
+      "grad_norm": 1.124918533241315,
+      "learning_rate": 2.673897310412288e-08,
+      "loss": 0.3076,
+      "step": 646
+    },
+    {
+      "epoch": 1.937125748502994,
+      "grad_norm": 0.9656481385246801,
+      "learning_rate": 2.4365334539667717e-08,
+      "loss": 0.2412,
+      "step": 647
+    },
+    {
+      "epoch": 1.9401197604790419,
+      "grad_norm": 1.178454834064512,
+      "learning_rate": 2.210174710486679e-08,
+      "loss": 0.2685,
+      "step": 648
+    },
+    {
+      "epoch": 1.94311377245509,
+      "grad_norm": 1.05307953402916,
+      "learning_rate": 1.99482608657664e-08,
+      "loss": 0.2647,
+      "step": 649
+    },
+    {
+      "epoch": 1.9461077844311379,
+      "grad_norm": 1.214139385138789,
+      "learning_rate": 1.7904923453193056e-08,
+      "loss": 0.3097,
+      "step": 650
+    },
+    {
+      "epoch": 1.9491017964071857,
+      "grad_norm": 1.1009388064541303,
+      "learning_rate": 1.5971780061701524e-08,
+      "loss": 0.2612,
+      "step": 651
+    },
+    {
+      "epoch": 1.9520958083832336,
+      "grad_norm": 1.150462404574678,
+      "learning_rate": 1.4148873448573408e-08,
+      "loss": 0.253,
+      "step": 652
+    },
+    {
+      "epoch": 1.9550898203592815,
+      "grad_norm": 1.0447259942981557,
+      "learning_rate": 1.2436243932872349e-08,
+      "loss": 0.2885,
+      "step": 653
+    },
+    {
+      "epoch": 1.9580838323353293,
+      "grad_norm": 1.122567359700312,
+      "learning_rate": 1.0833929394552523e-08,
+      "loss": 0.2452,
+      "step": 654
+    },
+    {
+      "epoch": 1.9610778443113772,
+      "grad_norm": 1.027601114488911,
+      "learning_rate": 9.341965273621522e-09,
+      "loss": 0.2893,
+      "step": 655
+    },
+    {
+      "epoch": 1.964071856287425,
+      "grad_norm": 1.1888392795209928,
+      "learning_rate": 7.96038456935322e-09,
+      "loss": 0.2934,
+      "step": 656
+    },
+    {
+      "epoch": 1.967065868263473,
+      "grad_norm": 1.1390150234084202,
+      "learning_rate": 6.6892178395611125e-09,
+      "loss": 0.319,
+      "step": 657
+    },
+    {
+      "epoch": 1.9700598802395208,
+      "grad_norm": 1.3088856587532898,
+      "learning_rate": 5.528493199922769e-09,
+      "loss": 0.3162,
+      "step": 658
+    },
+    {
+      "epoch": 1.9730538922155687,
+      "grad_norm": 1.1438085838572427,
+      "learning_rate": 4.478236323355312e-09,
+      "loss": 0.2511,
+      "step": 659
+    },
+    {
+      "epoch": 1.9760479041916168,
+      "grad_norm": 1.2284135560984777,
+      "learning_rate": 3.538470439448105e-09,
+      "loss": 0.3412,
+      "step": 660
+    },
+    {
+      "epoch": 1.9790419161676647,
+      "grad_norm": 1.1126818458876289,
+      "learning_rate": 2.709216333952602e-09,
+      "loss": 0.2904,
+      "step": 661
+    },
+    {
+      "epoch": 1.9820359281437125,
+      "grad_norm": 1.033514167646948,
+      "learning_rate": 1.9904923483171632e-09,
+      "loss": 0.2412,
+      "step": 662
+    },
+    {
+      "epoch": 1.9850299401197606,
+      "grad_norm": 1.18690751895419,
+      "learning_rate": 1.3823143792851545e-09,
+      "loss": 0.331,
+      "step": 663
+    },
+    {
+      "epoch": 1.9880239520958085,
+      "grad_norm": 1.21187037624878,
+      "learning_rate": 8.846958785418969e-10,
+      "loss": 0.3228,
+      "step": 664
+    },
+    {
+      "epoch": 1.9910179640718564,
+      "grad_norm": 1.115825785507024,
+      "learning_rate": 4.97647852417682e-10,
+      "loss": 0.2515,
+      "step": 665
+    },
+    {
+      "epoch": 1.9940119760479043,
+      "grad_norm": 1.0598510542494015,
+      "learning_rate": 2.2117886164407797e-10,
+      "loss": 0.2614,
+      "step": 666
+    },
+    {
+      "epoch": 1.9970059880239521,
+      "grad_norm": 0.9825317690024875,
+      "learning_rate": 5.529502116519148e-11,
+      "loss": 0.243,
+      "step": 667
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.0780801157479016,
+      "learning_rate": 0.0,
+      "loss": 0.1864,
+      "step": 668
+    },
+    {
+      "epoch": 2.0,
+      "step": 668,
+      "total_flos": 21199392964608.0,
+      "train_loss": 0.34515638107400454,
+      "train_runtime": 902.9082,
+      "train_samples_per_second": 11.804,
+      "train_steps_per_second": 0.74
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 668,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 5000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 21199392964608.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}