{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.07077975698950101,
"eval_steps": 200,
"global_step": 2200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.00011796626164916834,
"grad_norm": 93.0,
"learning_rate": 1.8912529550827425e-07,
"loss": 7.9641,
"num_input_tokens_seen": 1179648,
"step": 1
},
{
"epoch": 0.0002359325232983367,
"grad_norm": 95.0,
"learning_rate": 3.782505910165485e-07,
"loss": 7.9866,
"num_input_tokens_seen": 2359296,
"step": 2
},
{
"epoch": 0.000353898784947505,
"grad_norm": 97.5,
"learning_rate": 5.673758865248227e-07,
"loss": 7.983,
"num_input_tokens_seen": 3538944,
"step": 3
},
{
"epoch": 0.0004718650465966734,
"grad_norm": 92.0,
"learning_rate": 7.56501182033097e-07,
"loss": 7.9018,
"num_input_tokens_seen": 4718592,
"step": 4
},
{
"epoch": 0.0005898313082458417,
"grad_norm": 102.5,
"learning_rate": 9.456264775413712e-07,
"loss": 8.0784,
"num_input_tokens_seen": 5898240,
"step": 5
},
{
"epoch": 0.00070779756989501,
"grad_norm": 95.0,
"learning_rate": 1.1347517730496454e-06,
"loss": 7.9578,
"num_input_tokens_seen": 7077888,
"step": 6
},
{
"epoch": 0.0008257638315441783,
"grad_norm": 93.5,
"learning_rate": 1.3238770685579196e-06,
"loss": 7.975,
"num_input_tokens_seen": 8257536,
"step": 7
},
{
"epoch": 0.0009437300931933467,
"grad_norm": 85.0,
"learning_rate": 1.513002364066194e-06,
"loss": 7.8405,
"num_input_tokens_seen": 9437184,
"step": 8
},
{
"epoch": 0.001061696354842515,
"grad_norm": 83.0,
"learning_rate": 1.7021276595744682e-06,
"loss": 7.841,
"num_input_tokens_seen": 10616832,
"step": 9
},
{
"epoch": 0.0011796626164916834,
"grad_norm": 82.5,
"learning_rate": 1.8912529550827423e-06,
"loss": 7.9491,
"num_input_tokens_seen": 11796480,
"step": 10
},
{
"epoch": 0.0012976288781408518,
"grad_norm": 68.0,
"learning_rate": 2.0803782505910165e-06,
"loss": 7.5612,
"num_input_tokens_seen": 12976128,
"step": 11
},
{
"epoch": 0.00141559513979002,
"grad_norm": 66.5,
"learning_rate": 2.269503546099291e-06,
"loss": 7.5808,
"num_input_tokens_seen": 14155776,
"step": 12
},
{
"epoch": 0.0015335614014391884,
"grad_norm": 57.5,
"learning_rate": 2.4586288416075653e-06,
"loss": 7.4494,
"num_input_tokens_seen": 15335424,
"step": 13
},
{
"epoch": 0.0016515276630883566,
"grad_norm": 51.25,
"learning_rate": 2.6477541371158392e-06,
"loss": 7.4606,
"num_input_tokens_seen": 16515072,
"step": 14
},
{
"epoch": 0.001769493924737525,
"grad_norm": 42.75,
"learning_rate": 2.836879432624114e-06,
"loss": 7.2804,
"num_input_tokens_seen": 17694720,
"step": 15
},
{
"epoch": 0.0018874601863866935,
"grad_norm": 38.0,
"learning_rate": 3.026004728132388e-06,
"loss": 7.2173,
"num_input_tokens_seen": 18874368,
"step": 16
},
{
"epoch": 0.0020054264480358617,
"grad_norm": 33.75,
"learning_rate": 3.2151300236406624e-06,
"loss": 7.0934,
"num_input_tokens_seen": 20054016,
"step": 17
},
{
"epoch": 0.00212339270968503,
"grad_norm": 30.5,
"learning_rate": 3.4042553191489363e-06,
"loss": 7.1284,
"num_input_tokens_seen": 21233664,
"step": 18
},
{
"epoch": 0.0022413589713341986,
"grad_norm": 26.5,
"learning_rate": 3.5933806146572107e-06,
"loss": 6.932,
"num_input_tokens_seen": 22413312,
"step": 19
},
{
"epoch": 0.0023593252329833668,
"grad_norm": 23.875,
"learning_rate": 3.7825059101654847e-06,
"loss": 6.9163,
"num_input_tokens_seen": 23592960,
"step": 20
},
{
"epoch": 0.002477291494632535,
"grad_norm": 21.625,
"learning_rate": 3.9716312056737595e-06,
"loss": 6.8335,
"num_input_tokens_seen": 24772608,
"step": 21
},
{
"epoch": 0.0025952577562817036,
"grad_norm": 18.625,
"learning_rate": 4.160756501182033e-06,
"loss": 6.6761,
"num_input_tokens_seen": 25952256,
"step": 22
},
{
"epoch": 0.002713224017930872,
"grad_norm": 16.0,
"learning_rate": 4.349881796690308e-06,
"loss": 6.6474,
"num_input_tokens_seen": 27131904,
"step": 23
},
{
"epoch": 0.00283119027958004,
"grad_norm": 18.0,
"learning_rate": 4.539007092198582e-06,
"loss": 6.6062,
"num_input_tokens_seen": 28311552,
"step": 24
},
{
"epoch": 0.0029491565412292082,
"grad_norm": 17.5,
"learning_rate": 4.728132387706856e-06,
"loss": 6.6435,
"num_input_tokens_seen": 29491200,
"step": 25
},
{
"epoch": 0.003067122802878377,
"grad_norm": 14.25,
"learning_rate": 4.9172576832151305e-06,
"loss": 6.57,
"num_input_tokens_seen": 30670848,
"step": 26
},
{
"epoch": 0.003185089064527545,
"grad_norm": 13.3125,
"learning_rate": 5.106382978723404e-06,
"loss": 6.4037,
"num_input_tokens_seen": 31850496,
"step": 27
},
{
"epoch": 0.0033030553261767133,
"grad_norm": 10.8125,
"learning_rate": 5.2955082742316784e-06,
"loss": 6.2927,
"num_input_tokens_seen": 33030144,
"step": 28
},
{
"epoch": 0.003421021587825882,
"grad_norm": 9.75,
"learning_rate": 5.484633569739954e-06,
"loss": 6.2084,
"num_input_tokens_seen": 34209792,
"step": 29
},
{
"epoch": 0.00353898784947505,
"grad_norm": 10.25,
"learning_rate": 5.673758865248228e-06,
"loss": 6.2349,
"num_input_tokens_seen": 35389440,
"step": 30
},
{
"epoch": 0.0036569541111242184,
"grad_norm": 10.8125,
"learning_rate": 5.862884160756502e-06,
"loss": 6.155,
"num_input_tokens_seen": 36569088,
"step": 31
},
{
"epoch": 0.003774920372773387,
"grad_norm": 10.0,
"learning_rate": 6.052009456264776e-06,
"loss": 6.1846,
"num_input_tokens_seen": 37748736,
"step": 32
},
{
"epoch": 0.003892886634422555,
"grad_norm": 8.0,
"learning_rate": 6.24113475177305e-06,
"loss": 6.0656,
"num_input_tokens_seen": 38928384,
"step": 33
},
{
"epoch": 0.004010852896071723,
"grad_norm": 8.0625,
"learning_rate": 6.430260047281325e-06,
"loss": 5.9873,
"num_input_tokens_seen": 40108032,
"step": 34
},
{
"epoch": 0.004128819157720892,
"grad_norm": 8.875,
"learning_rate": 6.619385342789598e-06,
"loss": 6.0008,
"num_input_tokens_seen": 41287680,
"step": 35
},
{
"epoch": 0.00424678541937006,
"grad_norm": 7.34375,
"learning_rate": 6.808510638297873e-06,
"loss": 6.0852,
"num_input_tokens_seen": 42467328,
"step": 36
},
{
"epoch": 0.004364751681019229,
"grad_norm": 6.28125,
"learning_rate": 6.997635933806147e-06,
"loss": 5.8264,
"num_input_tokens_seen": 43646976,
"step": 37
},
{
"epoch": 0.004482717942668397,
"grad_norm": 6.75,
"learning_rate": 7.186761229314421e-06,
"loss": 5.9511,
"num_input_tokens_seen": 44826624,
"step": 38
},
{
"epoch": 0.004600684204317565,
"grad_norm": 6.96875,
"learning_rate": 7.375886524822695e-06,
"loss": 5.8042,
"num_input_tokens_seen": 46006272,
"step": 39
},
{
"epoch": 0.0047186504659667335,
"grad_norm": 5.59375,
"learning_rate": 7.565011820330969e-06,
"loss": 5.8955,
"num_input_tokens_seen": 47185920,
"step": 40
},
{
"epoch": 0.004836616727615902,
"grad_norm": 5.3125,
"learning_rate": 7.754137115839244e-06,
"loss": 5.7764,
"num_input_tokens_seen": 48365568,
"step": 41
},
{
"epoch": 0.00495458298926507,
"grad_norm": 5.40625,
"learning_rate": 7.943262411347519e-06,
"loss": 5.7784,
"num_input_tokens_seen": 49545216,
"step": 42
},
{
"epoch": 0.005072549250914238,
"grad_norm": 5.03125,
"learning_rate": 8.132387706855792e-06,
"loss": 5.7388,
"num_input_tokens_seen": 50724864,
"step": 43
},
{
"epoch": 0.005190515512563407,
"grad_norm": 4.84375,
"learning_rate": 8.321513002364066e-06,
"loss": 5.7821,
"num_input_tokens_seen": 51904512,
"step": 44
},
{
"epoch": 0.0053084817742125754,
"grad_norm": 4.78125,
"learning_rate": 8.510638297872341e-06,
"loss": 5.6466,
"num_input_tokens_seen": 53084160,
"step": 45
},
{
"epoch": 0.005426448035861744,
"grad_norm": 4.125,
"learning_rate": 8.699763593380616e-06,
"loss": 5.6224,
"num_input_tokens_seen": 54263808,
"step": 46
},
{
"epoch": 0.005544414297510912,
"grad_norm": 5.0,
"learning_rate": 8.888888888888888e-06,
"loss": 5.9019,
"num_input_tokens_seen": 55443456,
"step": 47
},
{
"epoch": 0.00566238055916008,
"grad_norm": 4.0,
"learning_rate": 9.078014184397164e-06,
"loss": 5.5812,
"num_input_tokens_seen": 56623104,
"step": 48
},
{
"epoch": 0.005780346820809248,
"grad_norm": 3.796875,
"learning_rate": 9.267139479905439e-06,
"loss": 5.633,
"num_input_tokens_seen": 57802752,
"step": 49
},
{
"epoch": 0.0058983130824584165,
"grad_norm": 3.84375,
"learning_rate": 9.456264775413712e-06,
"loss": 5.5031,
"num_input_tokens_seen": 58982400,
"step": 50
},
{
"epoch": 0.0060162793441075856,
"grad_norm": 3.4375,
"learning_rate": 9.645390070921986e-06,
"loss": 5.5146,
"num_input_tokens_seen": 60162048,
"step": 51
},
{
"epoch": 0.006134245605756754,
"grad_norm": 3.609375,
"learning_rate": 9.834515366430261e-06,
"loss": 5.3805,
"num_input_tokens_seen": 61341696,
"step": 52
},
{
"epoch": 0.006252211867405922,
"grad_norm": 3.234375,
"learning_rate": 1.0023640661938535e-05,
"loss": 5.4098,
"num_input_tokens_seen": 62521344,
"step": 53
},
{
"epoch": 0.00637017812905509,
"grad_norm": 3.40625,
"learning_rate": 1.0212765957446808e-05,
"loss": 5.3773,
"num_input_tokens_seen": 63700992,
"step": 54
},
{
"epoch": 0.006488144390704258,
"grad_norm": 3.1875,
"learning_rate": 1.0401891252955083e-05,
"loss": 5.3287,
"num_input_tokens_seen": 64880640,
"step": 55
},
{
"epoch": 0.006606110652353427,
"grad_norm": 3.046875,
"learning_rate": 1.0591016548463357e-05,
"loss": 5.2282,
"num_input_tokens_seen": 66060288,
"step": 56
},
{
"epoch": 0.006724076914002596,
"grad_norm": 3.0625,
"learning_rate": 1.0780141843971632e-05,
"loss": 5.2967,
"num_input_tokens_seen": 67239936,
"step": 57
},
{
"epoch": 0.006842043175651764,
"grad_norm": 2.859375,
"learning_rate": 1.0969267139479907e-05,
"loss": 5.1126,
"num_input_tokens_seen": 68419584,
"step": 58
},
{
"epoch": 0.006960009437300932,
"grad_norm": 3.0625,
"learning_rate": 1.1158392434988181e-05,
"loss": 5.3309,
"num_input_tokens_seen": 69599232,
"step": 59
},
{
"epoch": 0.0070779756989501,
"grad_norm": 3.0625,
"learning_rate": 1.1347517730496456e-05,
"loss": 5.2134,
"num_input_tokens_seen": 70778880,
"step": 60
},
{
"epoch": 0.0071959419605992685,
"grad_norm": 3.546875,
"learning_rate": 1.153664302600473e-05,
"loss": 5.1972,
"num_input_tokens_seen": 71958528,
"step": 61
},
{
"epoch": 0.007313908222248437,
"grad_norm": 2.828125,
"learning_rate": 1.1725768321513003e-05,
"loss": 5.0123,
"num_input_tokens_seen": 73138176,
"step": 62
},
{
"epoch": 0.007431874483897605,
"grad_norm": 3.203125,
"learning_rate": 1.1914893617021277e-05,
"loss": 5.1108,
"num_input_tokens_seen": 74317824,
"step": 63
},
{
"epoch": 0.007549840745546774,
"grad_norm": 3.609375,
"learning_rate": 1.2104018912529552e-05,
"loss": 5.2287,
"num_input_tokens_seen": 75497472,
"step": 64
},
{
"epoch": 0.007667807007195942,
"grad_norm": 2.703125,
"learning_rate": 1.2293144208037825e-05,
"loss": 5.0902,
"num_input_tokens_seen": 76677120,
"step": 65
},
{
"epoch": 0.00778577326884511,
"grad_norm": 3.1875,
"learning_rate": 1.24822695035461e-05,
"loss": 5.1526,
"num_input_tokens_seen": 77856768,
"step": 66
},
{
"epoch": 0.007903739530494279,
"grad_norm": 3.359375,
"learning_rate": 1.2671394799054376e-05,
"loss": 4.8855,
"num_input_tokens_seen": 79036416,
"step": 67
},
{
"epoch": 0.008021705792143447,
"grad_norm": 2.765625,
"learning_rate": 1.286052009456265e-05,
"loss": 5.0226,
"num_input_tokens_seen": 80216064,
"step": 68
},
{
"epoch": 0.008139672053792615,
"grad_norm": 3.078125,
"learning_rate": 1.3049645390070925e-05,
"loss": 5.1366,
"num_input_tokens_seen": 81395712,
"step": 69
},
{
"epoch": 0.008257638315441783,
"grad_norm": 3.71875,
"learning_rate": 1.3238770685579197e-05,
"loss": 5.0749,
"num_input_tokens_seen": 82575360,
"step": 70
},
{
"epoch": 0.008375604577090951,
"grad_norm": 2.765625,
"learning_rate": 1.3427895981087472e-05,
"loss": 4.8899,
"num_input_tokens_seen": 83755008,
"step": 71
},
{
"epoch": 0.00849357083874012,
"grad_norm": 3.140625,
"learning_rate": 1.3617021276595745e-05,
"loss": 4.8809,
"num_input_tokens_seen": 84934656,
"step": 72
},
{
"epoch": 0.008611537100389288,
"grad_norm": 3.296875,
"learning_rate": 1.380614657210402e-05,
"loss": 4.8839,
"num_input_tokens_seen": 86114304,
"step": 73
},
{
"epoch": 0.008729503362038458,
"grad_norm": 2.953125,
"learning_rate": 1.3995271867612294e-05,
"loss": 4.9931,
"num_input_tokens_seen": 87293952,
"step": 74
},
{
"epoch": 0.008847469623687626,
"grad_norm": 2.9375,
"learning_rate": 1.418439716312057e-05,
"loss": 4.8041,
"num_input_tokens_seen": 88473600,
"step": 75
},
{
"epoch": 0.008965435885336794,
"grad_norm": 3.5625,
"learning_rate": 1.4373522458628843e-05,
"loss": 4.8611,
"num_input_tokens_seen": 89653248,
"step": 76
},
{
"epoch": 0.009083402146985962,
"grad_norm": 3.734375,
"learning_rate": 1.4562647754137118e-05,
"loss": 4.7747,
"num_input_tokens_seen": 90832896,
"step": 77
},
{
"epoch": 0.00920136840863513,
"grad_norm": 2.984375,
"learning_rate": 1.475177304964539e-05,
"loss": 4.6884,
"num_input_tokens_seen": 92012544,
"step": 78
},
{
"epoch": 0.009319334670284299,
"grad_norm": 2.671875,
"learning_rate": 1.4940898345153665e-05,
"loss": 4.6617,
"num_input_tokens_seen": 93192192,
"step": 79
},
{
"epoch": 0.009437300931933467,
"grad_norm": 2.84375,
"learning_rate": 1.5130023640661939e-05,
"loss": 4.6174,
"num_input_tokens_seen": 94371840,
"step": 80
},
{
"epoch": 0.009555267193582635,
"grad_norm": 3.71875,
"learning_rate": 1.5319148936170214e-05,
"loss": 4.5483,
"num_input_tokens_seen": 95551488,
"step": 81
},
{
"epoch": 0.009673233455231803,
"grad_norm": 4.34375,
"learning_rate": 1.5508274231678487e-05,
"loss": 4.604,
"num_input_tokens_seen": 96731136,
"step": 82
},
{
"epoch": 0.009791199716880972,
"grad_norm": 3.09375,
"learning_rate": 1.5697399527186764e-05,
"loss": 4.5806,
"num_input_tokens_seen": 97910784,
"step": 83
},
{
"epoch": 0.00990916597853014,
"grad_norm": 3.203125,
"learning_rate": 1.5886524822695038e-05,
"loss": 4.4723,
"num_input_tokens_seen": 99090432,
"step": 84
},
{
"epoch": 0.010027132240179308,
"grad_norm": 3.8125,
"learning_rate": 1.607565011820331e-05,
"loss": 4.4723,
"num_input_tokens_seen": 100270080,
"step": 85
},
{
"epoch": 0.010145098501828476,
"grad_norm": 3.03125,
"learning_rate": 1.6264775413711585e-05,
"loss": 4.4519,
"num_input_tokens_seen": 101449728,
"step": 86
},
{
"epoch": 0.010263064763477646,
"grad_norm": 5.53125,
"learning_rate": 1.645390070921986e-05,
"loss": 4.4077,
"num_input_tokens_seen": 102629376,
"step": 87
},
{
"epoch": 0.010381031025126814,
"grad_norm": 4.03125,
"learning_rate": 1.6643026004728132e-05,
"loss": 4.5295,
"num_input_tokens_seen": 103809024,
"step": 88
},
{
"epoch": 0.010498997286775983,
"grad_norm": 3.484375,
"learning_rate": 1.683215130023641e-05,
"loss": 4.3235,
"num_input_tokens_seen": 104988672,
"step": 89
},
{
"epoch": 0.010616963548425151,
"grad_norm": 5.78125,
"learning_rate": 1.7021276595744682e-05,
"loss": 4.3245,
"num_input_tokens_seen": 106168320,
"step": 90
},
{
"epoch": 0.010734929810074319,
"grad_norm": 3.0625,
"learning_rate": 1.7210401891252956e-05,
"loss": 4.2554,
"num_input_tokens_seen": 107347968,
"step": 91
},
{
"epoch": 0.010852896071723487,
"grad_norm": 3.671875,
"learning_rate": 1.7399527186761233e-05,
"loss": 4.2572,
"num_input_tokens_seen": 108527616,
"step": 92
},
{
"epoch": 0.010970862333372655,
"grad_norm": 5.0625,
"learning_rate": 1.7588652482269506e-05,
"loss": 4.2442,
"num_input_tokens_seen": 109707264,
"step": 93
},
{
"epoch": 0.011088828595021824,
"grad_norm": 2.6875,
"learning_rate": 1.7777777777777777e-05,
"loss": 4.3006,
"num_input_tokens_seen": 110886912,
"step": 94
},
{
"epoch": 0.011206794856670992,
"grad_norm": 3.171875,
"learning_rate": 1.7966903073286054e-05,
"loss": 4.2813,
"num_input_tokens_seen": 112066560,
"step": 95
},
{
"epoch": 0.01132476111832016,
"grad_norm": 2.765625,
"learning_rate": 1.8156028368794327e-05,
"loss": 4.1186,
"num_input_tokens_seen": 113246208,
"step": 96
},
{
"epoch": 0.011442727379969328,
"grad_norm": 3.1875,
"learning_rate": 1.83451536643026e-05,
"loss": 4.1652,
"num_input_tokens_seen": 114425856,
"step": 97
},
{
"epoch": 0.011560693641618497,
"grad_norm": 4.71875,
"learning_rate": 1.8534278959810878e-05,
"loss": 4.1923,
"num_input_tokens_seen": 115605504,
"step": 98
},
{
"epoch": 0.011678659903267665,
"grad_norm": 3.359375,
"learning_rate": 1.872340425531915e-05,
"loss": 4.0728,
"num_input_tokens_seen": 116785152,
"step": 99
},
{
"epoch": 0.011796626164916833,
"grad_norm": 2.484375,
"learning_rate": 1.8912529550827425e-05,
"loss": 4.2189,
"num_input_tokens_seen": 117964800,
"step": 100
},
{
"epoch": 0.011914592426566003,
"grad_norm": 3.109375,
"learning_rate": 1.91016548463357e-05,
"loss": 4.0632,
"num_input_tokens_seen": 119144448,
"step": 101
},
{
"epoch": 0.012032558688215171,
"grad_norm": 4.40625,
"learning_rate": 1.929078014184397e-05,
"loss": 3.8775,
"num_input_tokens_seen": 120324096,
"step": 102
},
{
"epoch": 0.01215052494986434,
"grad_norm": 2.5625,
"learning_rate": 1.9479905437352245e-05,
"loss": 3.8658,
"num_input_tokens_seen": 121503744,
"step": 103
},
{
"epoch": 0.012268491211513508,
"grad_norm": 3.703125,
"learning_rate": 1.9669030732860522e-05,
"loss": 3.9888,
"num_input_tokens_seen": 122683392,
"step": 104
},
{
"epoch": 0.012386457473162676,
"grad_norm": 2.78125,
"learning_rate": 1.9858156028368796e-05,
"loss": 3.9439,
"num_input_tokens_seen": 123863040,
"step": 105
},
{
"epoch": 0.012504423734811844,
"grad_norm": 3.4375,
"learning_rate": 2.004728132387707e-05,
"loss": 4.0098,
"num_input_tokens_seen": 125042688,
"step": 106
},
{
"epoch": 0.012622389996461012,
"grad_norm": 2.71875,
"learning_rate": 2.0236406619385343e-05,
"loss": 3.9454,
"num_input_tokens_seen": 126222336,
"step": 107
},
{
"epoch": 0.01274035625811018,
"grad_norm": 3.03125,
"learning_rate": 2.0425531914893616e-05,
"loss": 3.8491,
"num_input_tokens_seen": 127401984,
"step": 108
},
{
"epoch": 0.012858322519759349,
"grad_norm": 3.59375,
"learning_rate": 2.0614657210401893e-05,
"loss": 3.9602,
"num_input_tokens_seen": 128581632,
"step": 109
},
{
"epoch": 0.012976288781408517,
"grad_norm": 5.25,
"learning_rate": 2.0803782505910167e-05,
"loss": 3.9542,
"num_input_tokens_seen": 129761280,
"step": 110
},
{
"epoch": 0.013094255043057685,
"grad_norm": 2.34375,
"learning_rate": 2.099290780141844e-05,
"loss": 3.7846,
"num_input_tokens_seen": 130940928,
"step": 111
},
{
"epoch": 0.013212221304706853,
"grad_norm": 3.484375,
"learning_rate": 2.1182033096926714e-05,
"loss": 3.9128,
"num_input_tokens_seen": 132120576,
"step": 112
},
{
"epoch": 0.013330187566356021,
"grad_norm": 7.4375,
"learning_rate": 2.137115839243499e-05,
"loss": 3.7365,
"num_input_tokens_seen": 133300224,
"step": 113
},
{
"epoch": 0.013448153828005191,
"grad_norm": 3.6875,
"learning_rate": 2.1560283687943264e-05,
"loss": 3.8298,
"num_input_tokens_seen": 134479872,
"step": 114
},
{
"epoch": 0.01356612008965436,
"grad_norm": 11.25,
"learning_rate": 2.1749408983451538e-05,
"loss": 3.9192,
"num_input_tokens_seen": 135659520,
"step": 115
},
{
"epoch": 0.013684086351303528,
"grad_norm": 9.875,
"learning_rate": 2.1938534278959815e-05,
"loss": 3.7628,
"num_input_tokens_seen": 136839168,
"step": 116
},
{
"epoch": 0.013802052612952696,
"grad_norm": 4.90625,
"learning_rate": 2.2127659574468088e-05,
"loss": 3.7212,
"num_input_tokens_seen": 138018816,
"step": 117
},
{
"epoch": 0.013920018874601864,
"grad_norm": 6.875,
"learning_rate": 2.2316784869976362e-05,
"loss": 3.8291,
"num_input_tokens_seen": 139198464,
"step": 118
},
{
"epoch": 0.014037985136251032,
"grad_norm": 3.984375,
"learning_rate": 2.2505910165484635e-05,
"loss": 3.7104,
"num_input_tokens_seen": 140378112,
"step": 119
},
{
"epoch": 0.0141559513979002,
"grad_norm": 5.0625,
"learning_rate": 2.2695035460992912e-05,
"loss": 3.6898,
"num_input_tokens_seen": 141557760,
"step": 120
},
{
"epoch": 0.014273917659549369,
"grad_norm": 3.828125,
"learning_rate": 2.2884160756501186e-05,
"loss": 3.61,
"num_input_tokens_seen": 142737408,
"step": 121
},
{
"epoch": 0.014391883921198537,
"grad_norm": 4.875,
"learning_rate": 2.307328605200946e-05,
"loss": 3.6886,
"num_input_tokens_seen": 143917056,
"step": 122
},
{
"epoch": 0.014509850182847705,
"grad_norm": 4.75,
"learning_rate": 2.326241134751773e-05,
"loss": 3.6435,
"num_input_tokens_seen": 145096704,
"step": 123
},
{
"epoch": 0.014627816444496873,
"grad_norm": 3.75,
"learning_rate": 2.3451536643026006e-05,
"loss": 3.6619,
"num_input_tokens_seen": 146276352,
"step": 124
},
{
"epoch": 0.014745782706146042,
"grad_norm": 5.375,
"learning_rate": 2.364066193853428e-05,
"loss": 3.631,
"num_input_tokens_seen": 147456000,
"step": 125
},
{
"epoch": 0.01486374896779521,
"grad_norm": 4.125,
"learning_rate": 2.3829787234042553e-05,
"loss": 3.6482,
"num_input_tokens_seen": 148635648,
"step": 126
},
{
"epoch": 0.014981715229444378,
"grad_norm": 7.78125,
"learning_rate": 2.4018912529550827e-05,
"loss": 3.6785,
"num_input_tokens_seen": 149815296,
"step": 127
},
{
"epoch": 0.015099681491093548,
"grad_norm": 7.0,
"learning_rate": 2.4208037825059104e-05,
"loss": 3.6113,
"num_input_tokens_seen": 150994944,
"step": 128
},
{
"epoch": 0.015217647752742716,
"grad_norm": 4.46875,
"learning_rate": 2.4397163120567377e-05,
"loss": 3.6015,
"num_input_tokens_seen": 152174592,
"step": 129
},
{
"epoch": 0.015335614014391884,
"grad_norm": 4.1875,
"learning_rate": 2.458628841607565e-05,
"loss": 3.5241,
"num_input_tokens_seen": 153354240,
"step": 130
},
{
"epoch": 0.015453580276041053,
"grad_norm": 4.0625,
"learning_rate": 2.4775413711583928e-05,
"loss": 3.6007,
"num_input_tokens_seen": 154533888,
"step": 131
},
{
"epoch": 0.01557154653769022,
"grad_norm": 3.046875,
"learning_rate": 2.49645390070922e-05,
"loss": 3.5949,
"num_input_tokens_seen": 155713536,
"step": 132
},
{
"epoch": 0.01568951279933939,
"grad_norm": 3.265625,
"learning_rate": 2.5153664302600475e-05,
"loss": 3.5403,
"num_input_tokens_seen": 156893184,
"step": 133
},
{
"epoch": 0.015807479060988557,
"grad_norm": 2.859375,
"learning_rate": 2.5342789598108752e-05,
"loss": 3.5399,
"num_input_tokens_seen": 158072832,
"step": 134
},
{
"epoch": 0.015925445322637725,
"grad_norm": 2.6875,
"learning_rate": 2.5531914893617025e-05,
"loss": 3.5408,
"num_input_tokens_seen": 159252480,
"step": 135
},
{
"epoch": 0.016043411584286894,
"grad_norm": 2.03125,
"learning_rate": 2.57210401891253e-05,
"loss": 3.4678,
"num_input_tokens_seen": 160432128,
"step": 136
},
{
"epoch": 0.016161377845936062,
"grad_norm": 2.796875,
"learning_rate": 2.5910165484633572e-05,
"loss": 3.6019,
"num_input_tokens_seen": 161611776,
"step": 137
},
{
"epoch": 0.01627934410758523,
"grad_norm": 1.96875,
"learning_rate": 2.609929078014185e-05,
"loss": 3.4525,
"num_input_tokens_seen": 162791424,
"step": 138
},
{
"epoch": 0.0163973103692344,
"grad_norm": 2.390625,
"learning_rate": 2.628841607565012e-05,
"loss": 3.557,
"num_input_tokens_seen": 163971072,
"step": 139
},
{
"epoch": 0.016515276630883566,
"grad_norm": 2.734375,
"learning_rate": 2.6477541371158393e-05,
"loss": 3.5324,
"num_input_tokens_seen": 165150720,
"step": 140
},
{
"epoch": 0.016633242892532735,
"grad_norm": 3.703125,
"learning_rate": 2.6666666666666667e-05,
"loss": 3.4707,
"num_input_tokens_seen": 166330368,
"step": 141
},
{
"epoch": 0.016751209154181903,
"grad_norm": 3.53125,
"learning_rate": 2.6855791962174944e-05,
"loss": 3.3545,
"num_input_tokens_seen": 167510016,
"step": 142
},
{
"epoch": 0.01686917541583107,
"grad_norm": 1.7890625,
"learning_rate": 2.7044917257683217e-05,
"loss": 3.4051,
"num_input_tokens_seen": 168689664,
"step": 143
},
{
"epoch": 0.01698714167748024,
"grad_norm": 4.78125,
"learning_rate": 2.723404255319149e-05,
"loss": 3.4119,
"num_input_tokens_seen": 169869312,
"step": 144
},
{
"epoch": 0.017105107939129408,
"grad_norm": 3.0625,
"learning_rate": 2.7423167848699764e-05,
"loss": 3.5149,
"num_input_tokens_seen": 171048960,
"step": 145
},
{
"epoch": 0.017223074200778576,
"grad_norm": 3.671875,
"learning_rate": 2.761229314420804e-05,
"loss": 3.3442,
"num_input_tokens_seen": 172228608,
"step": 146
},
{
"epoch": 0.017341040462427744,
"grad_norm": 3.453125,
"learning_rate": 2.7801418439716315e-05,
"loss": 3.3277,
"num_input_tokens_seen": 173408256,
"step": 147
},
{
"epoch": 0.017459006724076916,
"grad_norm": 3.4375,
"learning_rate": 2.7990543735224588e-05,
"loss": 3.3905,
"num_input_tokens_seen": 174587904,
"step": 148
},
{
"epoch": 0.017576972985726084,
"grad_norm": 2.0,
"learning_rate": 2.8179669030732865e-05,
"loss": 3.276,
"num_input_tokens_seen": 175767552,
"step": 149
},
{
"epoch": 0.017694939247375252,
"grad_norm": 4.0,
"learning_rate": 2.836879432624114e-05,
"loss": 3.3211,
"num_input_tokens_seen": 176947200,
"step": 150
},
{
"epoch": 0.01781290550902442,
"grad_norm": 4.03125,
"learning_rate": 2.8557919621749412e-05,
"loss": 3.3483,
"num_input_tokens_seen": 178126848,
"step": 151
},
{
"epoch": 0.01793087177067359,
"grad_norm": 2.53125,
"learning_rate": 2.8747044917257686e-05,
"loss": 3.3391,
"num_input_tokens_seen": 179306496,
"step": 152
},
{
"epoch": 0.018048838032322757,
"grad_norm": 5.5625,
"learning_rate": 2.8936170212765963e-05,
"loss": 3.4719,
"num_input_tokens_seen": 180486144,
"step": 153
},
{
"epoch": 0.018166804293971925,
"grad_norm": 3.171875,
"learning_rate": 2.9125295508274236e-05,
"loss": 3.2727,
"num_input_tokens_seen": 181665792,
"step": 154
},
{
"epoch": 0.018284770555621093,
"grad_norm": 5.09375,
"learning_rate": 2.9314420803782506e-05,
"loss": 3.284,
"num_input_tokens_seen": 182845440,
"step": 155
},
{
"epoch": 0.01840273681727026,
"grad_norm": 3.25,
"learning_rate": 2.950354609929078e-05,
"loss": 3.2279,
"num_input_tokens_seen": 184025088,
"step": 156
},
{
"epoch": 0.01852070307891943,
"grad_norm": 3.71875,
"learning_rate": 2.9692671394799057e-05,
"loss": 3.2438,
"num_input_tokens_seen": 185204736,
"step": 157
},
{
"epoch": 0.018638669340568598,
"grad_norm": 3.875,
"learning_rate": 2.988179669030733e-05,
"loss": 3.3257,
"num_input_tokens_seen": 186384384,
"step": 158
},
{
"epoch": 0.018756635602217766,
"grad_norm": 2.21875,
"learning_rate": 3.0070921985815604e-05,
"loss": 3.2727,
"num_input_tokens_seen": 187564032,
"step": 159
},
{
"epoch": 0.018874601863866934,
"grad_norm": 4.125,
"learning_rate": 3.0260047281323877e-05,
"loss": 3.245,
"num_input_tokens_seen": 188743680,
"step": 160
},
{
"epoch": 0.018992568125516102,
"grad_norm": 3.640625,
"learning_rate": 3.0449172576832154e-05,
"loss": 3.1904,
"num_input_tokens_seen": 189923328,
"step": 161
},
{
"epoch": 0.01911053438716527,
"grad_norm": 2.625,
"learning_rate": 3.063829787234043e-05,
"loss": 3.2754,
"num_input_tokens_seen": 191102976,
"step": 162
},
{
"epoch": 0.01922850064881444,
"grad_norm": 3.578125,
"learning_rate": 3.0827423167848705e-05,
"loss": 3.1889,
"num_input_tokens_seen": 192282624,
"step": 163
},
{
"epoch": 0.019346466910463607,
"grad_norm": 2.953125,
"learning_rate": 3.1016548463356975e-05,
"loss": 3.2809,
"num_input_tokens_seen": 193462272,
"step": 164
},
{
"epoch": 0.019464433172112775,
"grad_norm": 2.234375,
"learning_rate": 3.120567375886525e-05,
"loss": 3.1929,
"num_input_tokens_seen": 194641920,
"step": 165
},
{
"epoch": 0.019582399433761943,
"grad_norm": 2.515625,
"learning_rate": 3.139479905437353e-05,
"loss": 3.1474,
"num_input_tokens_seen": 195821568,
"step": 166
},
{
"epoch": 0.01970036569541111,
"grad_norm": 4.09375,
"learning_rate": 3.15839243498818e-05,
"loss": 3.0433,
"num_input_tokens_seen": 197001216,
"step": 167
},
{
"epoch": 0.01981833195706028,
"grad_norm": 2.6875,
"learning_rate": 3.1773049645390076e-05,
"loss": 3.1527,
"num_input_tokens_seen": 198180864,
"step": 168
},
{
"epoch": 0.019936298218709448,
"grad_norm": 3.5,
"learning_rate": 3.196217494089835e-05,
"loss": 3.2145,
"num_input_tokens_seen": 199360512,
"step": 169
},
{
"epoch": 0.020054264480358616,
"grad_norm": 4.9375,
"learning_rate": 3.215130023640662e-05,
"loss": 3.2642,
"num_input_tokens_seen": 200540160,
"step": 170
},
{
"epoch": 0.020172230742007784,
"grad_norm": 2.484375,
"learning_rate": 3.234042553191489e-05,
"loss": 3.1286,
"num_input_tokens_seen": 201719808,
"step": 171
},
{
"epoch": 0.020290197003656953,
"grad_norm": 8.1875,
"learning_rate": 3.252955082742317e-05,
"loss": 3.1274,
"num_input_tokens_seen": 202899456,
"step": 172
},
{
"epoch": 0.02040816326530612,
"grad_norm": 6.375,
"learning_rate": 3.271867612293144e-05,
"loss": 3.096,
"num_input_tokens_seen": 204079104,
"step": 173
},
{
"epoch": 0.020526129526955292,
"grad_norm": 5.8125,
"learning_rate": 3.290780141843972e-05,
"loss": 3.1647,
"num_input_tokens_seen": 205258752,
"step": 174
},
{
"epoch": 0.02064409578860446,
"grad_norm": 5.6875,
"learning_rate": 3.3096926713947994e-05,
"loss": 3.269,
"num_input_tokens_seen": 206438400,
"step": 175
},
{
"epoch": 0.02076206205025363,
"grad_norm": 4.4375,
"learning_rate": 3.3286052009456264e-05,
"loss": 3.2106,
"num_input_tokens_seen": 207618048,
"step": 176
},
{
"epoch": 0.020880028311902797,
"grad_norm": 3.109375,
"learning_rate": 3.347517730496454e-05,
"loss": 3.1545,
"num_input_tokens_seen": 208797696,
"step": 177
},
{
"epoch": 0.020997994573551965,
"grad_norm": 6.6875,
"learning_rate": 3.366430260047282e-05,
"loss": 3.1045,
"num_input_tokens_seen": 209977344,
"step": 178
},
{
"epoch": 0.021115960835201134,
"grad_norm": 6.0625,
"learning_rate": 3.385342789598109e-05,
"loss": 3.0496,
"num_input_tokens_seen": 211156992,
"step": 179
},
{
"epoch": 0.021233927096850302,
"grad_norm": 4.03125,
"learning_rate": 3.4042553191489365e-05,
"loss": 3.081,
"num_input_tokens_seen": 212336640,
"step": 180
},
{
"epoch": 0.02135189335849947,
"grad_norm": 3.921875,
"learning_rate": 3.423167848699764e-05,
"loss": 3.0552,
"num_input_tokens_seen": 213516288,
"step": 181
},
{
"epoch": 0.021469859620148638,
"grad_norm": 4.5,
"learning_rate": 3.442080378250591e-05,
"loss": 3.0172,
"num_input_tokens_seen": 214695936,
"step": 182
},
{
"epoch": 0.021587825881797806,
"grad_norm": 4.0625,
"learning_rate": 3.460992907801419e-05,
"loss": 3.1379,
"num_input_tokens_seen": 215875584,
"step": 183
},
{
"epoch": 0.021705792143446975,
"grad_norm": 4.9375,
"learning_rate": 3.4799054373522466e-05,
"loss": 3.1235,
"num_input_tokens_seen": 217055232,
"step": 184
},
{
"epoch": 0.021823758405096143,
"grad_norm": 3.78125,
"learning_rate": 3.4988179669030736e-05,
"loss": 3.1189,
"num_input_tokens_seen": 218234880,
"step": 185
},
{
"epoch": 0.02194172466674531,
"grad_norm": 5.0625,
"learning_rate": 3.517730496453901e-05,
"loss": 3.0035,
"num_input_tokens_seen": 219414528,
"step": 186
},
{
"epoch": 0.02205969092839448,
"grad_norm": 4.28125,
"learning_rate": 3.536643026004728e-05,
"loss": 3.0478,
"num_input_tokens_seen": 220594176,
"step": 187
},
{
"epoch": 0.022177657190043647,
"grad_norm": 5.8125,
"learning_rate": 3.555555555555555e-05,
"loss": 3.0777,
"num_input_tokens_seen": 221773824,
"step": 188
},
{
"epoch": 0.022295623451692816,
"grad_norm": 5.0625,
"learning_rate": 3.574468085106383e-05,
"loss": 3.0665,
"num_input_tokens_seen": 222953472,
"step": 189
},
{
"epoch": 0.022413589713341984,
"grad_norm": 4.25,
"learning_rate": 3.593380614657211e-05,
"loss": 3.0271,
"num_input_tokens_seen": 224133120,
"step": 190
},
{
"epoch": 0.022531555974991152,
"grad_norm": 4.03125,
"learning_rate": 3.612293144208038e-05,
"loss": 3.033,
"num_input_tokens_seen": 225312768,
"step": 191
},
{
"epoch": 0.02264952223664032,
"grad_norm": 5.09375,
"learning_rate": 3.6312056737588654e-05,
"loss": 3.143,
"num_input_tokens_seen": 226492416,
"step": 192
},
{
"epoch": 0.02276748849828949,
"grad_norm": 4.09375,
"learning_rate": 3.650118203309693e-05,
"loss": 3.0347,
"num_input_tokens_seen": 227672064,
"step": 193
},
{
"epoch": 0.022885454759938657,
"grad_norm": 4.65625,
"learning_rate": 3.66903073286052e-05,
"loss": 3.07,
"num_input_tokens_seen": 228851712,
"step": 194
},
{
"epoch": 0.023003421021587825,
"grad_norm": 3.90625,
"learning_rate": 3.687943262411348e-05,
"loss": 3.0225,
"num_input_tokens_seen": 230031360,
"step": 195
},
{
"epoch": 0.023121387283236993,
"grad_norm": 4.96875,
"learning_rate": 3.7068557919621755e-05,
"loss": 3.0222,
"num_input_tokens_seen": 231211008,
"step": 196
},
{
"epoch": 0.02323935354488616,
"grad_norm": 3.84375,
"learning_rate": 3.7257683215130025e-05,
"loss": 2.9755,
"num_input_tokens_seen": 232390656,
"step": 197
},
{
"epoch": 0.02335731980653533,
"grad_norm": 5.0,
"learning_rate": 3.74468085106383e-05,
"loss": 3.0163,
"num_input_tokens_seen": 233570304,
"step": 198
},
{
"epoch": 0.023475286068184498,
"grad_norm": 4.53125,
"learning_rate": 3.763593380614658e-05,
"loss": 2.9553,
"num_input_tokens_seen": 234749952,
"step": 199
},
{
"epoch": 0.023593252329833666,
"grad_norm": 4.6875,
"learning_rate": 3.782505910165485e-05,
"loss": 3.015,
"num_input_tokens_seen": 235929600,
"step": 200
},
{
"epoch": 0.023593252329833666,
"eval_wikipedia_loss": 2.9615590572357178,
"eval_wikipedia_runtime": 172.3085,
"eval_wikipedia_samples_per_second": 4.074,
"eval_wikipedia_steps_per_second": 0.174,
"num_input_tokens_seen": 235929600,
"step": 200
},
{
"epoch": 0.023593252329833666,
"eval_toxicity_loss": 4.73836088180542,
"eval_toxicity_runtime": 0.999,
"eval_toxicity_samples_per_second": 2.002,
"eval_toxicity_steps_per_second": 1.001,
"num_input_tokens_seen": 235929600,
"step": 200
},
{
"epoch": 0.023711218591482838,
"grad_norm": 4.0,
"learning_rate": 3.8014184397163126e-05,
"loss": 2.9448,
"num_input_tokens_seen": 237109248,
"step": 201
},
{
"epoch": 0.023829184853132006,
"grad_norm": 4.78125,
"learning_rate": 3.82033096926714e-05,
"loss": 3.1131,
"num_input_tokens_seen": 238288896,
"step": 202
},
{
"epoch": 0.023947151114781174,
"grad_norm": 3.828125,
"learning_rate": 3.839243498817967e-05,
"loss": 2.9217,
"num_input_tokens_seen": 239468544,
"step": 203
},
{
"epoch": 0.024065117376430342,
"grad_norm": 4.46875,
"learning_rate": 3.858156028368794e-05,
"loss": 2.9658,
"num_input_tokens_seen": 240648192,
"step": 204
},
{
"epoch": 0.02418308363807951,
"grad_norm": 3.703125,
"learning_rate": 3.877068557919622e-05,
"loss": 2.9857,
"num_input_tokens_seen": 241827840,
"step": 205
},
{
"epoch": 0.02430104989972868,
"grad_norm": 4.9375,
"learning_rate": 3.895981087470449e-05,
"loss": 2.9761,
"num_input_tokens_seen": 243007488,
"step": 206
},
{
"epoch": 0.024419016161377847,
"grad_norm": 4.28125,
"learning_rate": 3.914893617021277e-05,
"loss": 2.981,
"num_input_tokens_seen": 244187136,
"step": 207
},
{
"epoch": 0.024536982423027015,
"grad_norm": 4.3125,
"learning_rate": 3.9338061465721044e-05,
"loss": 3.0308,
"num_input_tokens_seen": 245366784,
"step": 208
},
{
"epoch": 0.024654948684676183,
"grad_norm": 4.0,
"learning_rate": 3.9527186761229314e-05,
"loss": 2.9529,
"num_input_tokens_seen": 246546432,
"step": 209
},
{
"epoch": 0.02477291494632535,
"grad_norm": 4.28125,
"learning_rate": 3.971631205673759e-05,
"loss": 2.9297,
"num_input_tokens_seen": 247726080,
"step": 210
},
{
"epoch": 0.02489088120797452,
"grad_norm": 3.328125,
"learning_rate": 3.990543735224587e-05,
"loss": 2.8328,
"num_input_tokens_seen": 248905728,
"step": 211
},
{
"epoch": 0.025008847469623688,
"grad_norm": 4.6875,
"learning_rate": 4.009456264775414e-05,
"loss": 2.9565,
"num_input_tokens_seen": 250085376,
"step": 212
},
{
"epoch": 0.025126813731272856,
"grad_norm": 3.734375,
"learning_rate": 4.028368794326241e-05,
"loss": 2.9276,
"num_input_tokens_seen": 251265024,
"step": 213
},
{
"epoch": 0.025244779992922024,
"grad_norm": 4.90625,
"learning_rate": 4.0472813238770685e-05,
"loss": 2.8872,
"num_input_tokens_seen": 252444672,
"step": 214
},
{
"epoch": 0.025362746254571193,
"grad_norm": 3.984375,
"learning_rate": 4.066193853427896e-05,
"loss": 2.9423,
"num_input_tokens_seen": 253624320,
"step": 215
},
{
"epoch": 0.02548071251622036,
"grad_norm": 4.28125,
"learning_rate": 4.085106382978723e-05,
"loss": 2.9417,
"num_input_tokens_seen": 254803968,
"step": 216
},
{
"epoch": 0.02559867877786953,
"grad_norm": 3.53125,
"learning_rate": 4.104018912529551e-05,
"loss": 2.878,
"num_input_tokens_seen": 255983616,
"step": 217
},
{
"epoch": 0.025716645039518697,
"grad_norm": 3.71875,
"learning_rate": 4.1229314420803786e-05,
"loss": 2.8925,
"num_input_tokens_seen": 257163264,
"step": 218
},
{
"epoch": 0.025834611301167865,
"grad_norm": 3.109375,
"learning_rate": 4.1418439716312056e-05,
"loss": 2.9432,
"num_input_tokens_seen": 258342912,
"step": 219
},
{
"epoch": 0.025952577562817034,
"grad_norm": 4.875,
"learning_rate": 4.1607565011820333e-05,
"loss": 2.8552,
"num_input_tokens_seen": 259522560,
"step": 220
},
{
"epoch": 0.026070543824466202,
"grad_norm": 3.09375,
"learning_rate": 4.1796690307328604e-05,
"loss": 2.8685,
"num_input_tokens_seen": 260702208,
"step": 221
},
{
"epoch": 0.02618851008611537,
"grad_norm": 4.6875,
"learning_rate": 4.198581560283688e-05,
"loss": 2.9284,
"num_input_tokens_seen": 261881856,
"step": 222
},
{
"epoch": 0.026306476347764538,
"grad_norm": 3.265625,
"learning_rate": 4.217494089834516e-05,
"loss": 2.8862,
"num_input_tokens_seen": 263061504,
"step": 223
},
{
"epoch": 0.026424442609413706,
"grad_norm": 4.5625,
"learning_rate": 4.236406619385343e-05,
"loss": 2.8105,
"num_input_tokens_seen": 264241152,
"step": 224
},
{
"epoch": 0.026542408871062875,
"grad_norm": 3.546875,
"learning_rate": 4.2553191489361704e-05,
"loss": 2.9406,
"num_input_tokens_seen": 265420800,
"step": 225
},
{
"epoch": 0.026660375132712043,
"grad_norm": 3.640625,
"learning_rate": 4.274231678486998e-05,
"loss": 2.907,
"num_input_tokens_seen": 266600448,
"step": 226
},
{
"epoch": 0.02677834139436121,
"grad_norm": 3.015625,
"learning_rate": 4.293144208037825e-05,
"loss": 3.0143,
"num_input_tokens_seen": 267780096,
"step": 227
},
{
"epoch": 0.026896307656010383,
"grad_norm": 4.25,
"learning_rate": 4.312056737588653e-05,
"loss": 2.9126,
"num_input_tokens_seen": 268959744,
"step": 228
},
{
"epoch": 0.02701427391765955,
"grad_norm": 2.5,
"learning_rate": 4.3309692671394805e-05,
"loss": 2.8879,
"num_input_tokens_seen": 270139392,
"step": 229
},
{
"epoch": 0.02713224017930872,
"grad_norm": 3.34375,
"learning_rate": 4.3498817966903076e-05,
"loss": 2.8174,
"num_input_tokens_seen": 271319040,
"step": 230
},
{
"epoch": 0.027250206440957887,
"grad_norm": 4.40625,
"learning_rate": 4.368794326241135e-05,
"loss": 2.895,
"num_input_tokens_seen": 272498688,
"step": 231
},
{
"epoch": 0.027368172702607056,
"grad_norm": 2.140625,
"learning_rate": 4.387706855791963e-05,
"loss": 2.8573,
"num_input_tokens_seen": 273678336,
"step": 232
},
{
"epoch": 0.027486138964256224,
"grad_norm": 3.609375,
"learning_rate": 4.40661938534279e-05,
"loss": 2.7662,
"num_input_tokens_seen": 274857984,
"step": 233
},
{
"epoch": 0.027604105225905392,
"grad_norm": 3.484375,
"learning_rate": 4.4255319148936176e-05,
"loss": 2.8103,
"num_input_tokens_seen": 276037632,
"step": 234
},
{
"epoch": 0.02772207148755456,
"grad_norm": 1.9765625,
"learning_rate": 4.444444444444445e-05,
"loss": 2.8271,
"num_input_tokens_seen": 277217280,
"step": 235
},
{
"epoch": 0.02784003774920373,
"grad_norm": 3.328125,
"learning_rate": 4.4633569739952723e-05,
"loss": 2.8881,
"num_input_tokens_seen": 278396928,
"step": 236
},
{
"epoch": 0.027958004010852897,
"grad_norm": 3.9375,
"learning_rate": 4.4822695035461e-05,
"loss": 2.9018,
"num_input_tokens_seen": 279576576,
"step": 237
},
{
"epoch": 0.028075970272502065,
"grad_norm": 5.25,
"learning_rate": 4.501182033096927e-05,
"loss": 2.8584,
"num_input_tokens_seen": 280756224,
"step": 238
},
{
"epoch": 0.028193936534151233,
"grad_norm": 2.328125,
"learning_rate": 4.520094562647755e-05,
"loss": 2.8435,
"num_input_tokens_seen": 281935872,
"step": 239
},
{
"epoch": 0.0283119027958004,
"grad_norm": 3.375,
"learning_rate": 4.5390070921985824e-05,
"loss": 2.8827,
"num_input_tokens_seen": 283115520,
"step": 240
},
{
"epoch": 0.02842986905744957,
"grad_norm": 2.3125,
"learning_rate": 4.5579196217494095e-05,
"loss": 2.8415,
"num_input_tokens_seen": 284295168,
"step": 241
},
{
"epoch": 0.028547835319098738,
"grad_norm": 3.828125,
"learning_rate": 4.576832151300237e-05,
"loss": 2.8305,
"num_input_tokens_seen": 285474816,
"step": 242
},
{
"epoch": 0.028665801580747906,
"grad_norm": 2.5625,
"learning_rate": 4.595744680851065e-05,
"loss": 2.7947,
"num_input_tokens_seen": 286654464,
"step": 243
},
{
"epoch": 0.028783767842397074,
"grad_norm": 2.375,
"learning_rate": 4.614657210401892e-05,
"loss": 2.812,
"num_input_tokens_seen": 287834112,
"step": 244
},
{
"epoch": 0.028901734104046242,
"grad_norm": 2.578125,
"learning_rate": 4.633569739952719e-05,
"loss": 2.8157,
"num_input_tokens_seen": 289013760,
"step": 245
},
{
"epoch": 0.02901970036569541,
"grad_norm": 3.625,
"learning_rate": 4.652482269503546e-05,
"loss": 2.8768,
"num_input_tokens_seen": 290193408,
"step": 246
},
{
"epoch": 0.02913766662734458,
"grad_norm": 2.328125,
"learning_rate": 4.6713947990543736e-05,
"loss": 2.834,
"num_input_tokens_seen": 291373056,
"step": 247
},
{
"epoch": 0.029255632888993747,
"grad_norm": 2.1875,
"learning_rate": 4.690307328605201e-05,
"loss": 2.8125,
"num_input_tokens_seen": 292552704,
"step": 248
},
{
"epoch": 0.029373599150642915,
"grad_norm": 2.921875,
"learning_rate": 4.709219858156028e-05,
"loss": 2.7906,
"num_input_tokens_seen": 293732352,
"step": 249
},
{
"epoch": 0.029491565412292083,
"grad_norm": 1.921875,
"learning_rate": 4.728132387706856e-05,
"loss": 2.8163,
"num_input_tokens_seen": 294912000,
"step": 250
},
{
"epoch": 0.02960953167394125,
"grad_norm": 2.75,
"learning_rate": 4.747044917257684e-05,
"loss": 2.7826,
"num_input_tokens_seen": 296091648,
"step": 251
},
{
"epoch": 0.02972749793559042,
"grad_norm": 2.671875,
"learning_rate": 4.765957446808511e-05,
"loss": 2.8368,
"num_input_tokens_seen": 297271296,
"step": 252
},
{
"epoch": 0.029845464197239588,
"grad_norm": 2.796875,
"learning_rate": 4.7848699763593384e-05,
"loss": 2.7188,
"num_input_tokens_seen": 298450944,
"step": 253
},
{
"epoch": 0.029963430458888756,
"grad_norm": 1.984375,
"learning_rate": 4.8037825059101654e-05,
"loss": 2.7609,
"num_input_tokens_seen": 299630592,
"step": 254
},
{
"epoch": 0.030081396720537928,
"grad_norm": 2.484375,
"learning_rate": 4.822695035460993e-05,
"loss": 2.742,
"num_input_tokens_seen": 300810240,
"step": 255
},
{
"epoch": 0.030199362982187096,
"grad_norm": 1.84375,
"learning_rate": 4.841607565011821e-05,
"loss": 2.7751,
"num_input_tokens_seen": 301989888,
"step": 256
},
{
"epoch": 0.030317329243836264,
"grad_norm": 3.078125,
"learning_rate": 4.860520094562648e-05,
"loss": 2.6949,
"num_input_tokens_seen": 303169536,
"step": 257
},
{
"epoch": 0.030435295505485432,
"grad_norm": 3.25,
"learning_rate": 4.8794326241134755e-05,
"loss": 2.8174,
"num_input_tokens_seen": 304349184,
"step": 258
},
{
"epoch": 0.0305532617671346,
"grad_norm": 1.4921875,
"learning_rate": 4.898345153664303e-05,
"loss": 2.6962,
"num_input_tokens_seen": 305528832,
"step": 259
},
{
"epoch": 0.03067122802878377,
"grad_norm": 3.046875,
"learning_rate": 4.91725768321513e-05,
"loss": 2.8484,
"num_input_tokens_seen": 306708480,
"step": 260
},
{
"epoch": 0.030789194290432937,
"grad_norm": 2.78125,
"learning_rate": 4.936170212765958e-05,
"loss": 2.632,
"num_input_tokens_seen": 307888128,
"step": 261
},
{
"epoch": 0.030907160552082105,
"grad_norm": 2.546875,
"learning_rate": 4.9550827423167856e-05,
"loss": 2.8744,
"num_input_tokens_seen": 309067776,
"step": 262
},
{
"epoch": 0.031025126813731273,
"grad_norm": 2.359375,
"learning_rate": 4.9739952718676126e-05,
"loss": 2.7714,
"num_input_tokens_seen": 310247424,
"step": 263
},
{
"epoch": 0.03114309307538044,
"grad_norm": 1.8515625,
"learning_rate": 4.99290780141844e-05,
"loss": 2.8042,
"num_input_tokens_seen": 311427072,
"step": 264
},
{
"epoch": 0.031261059337029606,
"grad_norm": 2.265625,
"learning_rate": 5.011820330969268e-05,
"loss": 2.7448,
"num_input_tokens_seen": 312606720,
"step": 265
},
{
"epoch": 0.03137902559867878,
"grad_norm": 1.78125,
"learning_rate": 5.030732860520095e-05,
"loss": 2.678,
"num_input_tokens_seen": 313786368,
"step": 266
},
{
"epoch": 0.03149699186032794,
"grad_norm": 2.203125,
"learning_rate": 5.049645390070923e-05,
"loss": 2.676,
"num_input_tokens_seen": 314966016,
"step": 267
},
{
"epoch": 0.031614958121977114,
"grad_norm": 1.90625,
"learning_rate": 5.0685579196217504e-05,
"loss": 2.6778,
"num_input_tokens_seen": 316145664,
"step": 268
},
{
"epoch": 0.031732924383626286,
"grad_norm": 2.203125,
"learning_rate": 5.0874704491725774e-05,
"loss": 2.7624,
"num_input_tokens_seen": 317325312,
"step": 269
},
{
"epoch": 0.03185089064527545,
"grad_norm": 3.53125,
"learning_rate": 5.106382978723405e-05,
"loss": 2.6462,
"num_input_tokens_seen": 318504960,
"step": 270
},
{
"epoch": 0.03196885690692462,
"grad_norm": 2.1875,
"learning_rate": 5.125295508274232e-05,
"loss": 2.6475,
"num_input_tokens_seen": 319684608,
"step": 271
},
{
"epoch": 0.03208682316857379,
"grad_norm": 1.703125,
"learning_rate": 5.14420803782506e-05,
"loss": 2.7214,
"num_input_tokens_seen": 320864256,
"step": 272
},
{
"epoch": 0.03220478943022296,
"grad_norm": 2.25,
"learning_rate": 5.1631205673758875e-05,
"loss": 2.6597,
"num_input_tokens_seen": 322043904,
"step": 273
},
{
"epoch": 0.032322755691872124,
"grad_norm": 2.171875,
"learning_rate": 5.1820330969267145e-05,
"loss": 2.6412,
"num_input_tokens_seen": 323223552,
"step": 274
},
{
"epoch": 0.032440721953521295,
"grad_norm": 2.25,
"learning_rate": 5.200945626477542e-05,
"loss": 2.7664,
"num_input_tokens_seen": 324403200,
"step": 275
},
{
"epoch": 0.03255868821517046,
"grad_norm": 3.296875,
"learning_rate": 5.21985815602837e-05,
"loss": 2.7168,
"num_input_tokens_seen": 325582848,
"step": 276
},
{
"epoch": 0.03267665447681963,
"grad_norm": 1.515625,
"learning_rate": 5.238770685579196e-05,
"loss": 2.698,
"num_input_tokens_seen": 326762496,
"step": 277
},
{
"epoch": 0.0327946207384688,
"grad_norm": 1.75,
"learning_rate": 5.257683215130024e-05,
"loss": 2.7068,
"num_input_tokens_seen": 327942144,
"step": 278
},
{
"epoch": 0.03291258700011797,
"grad_norm": 2.859375,
"learning_rate": 5.276595744680851e-05,
"loss": 2.6487,
"num_input_tokens_seen": 329121792,
"step": 279
},
{
"epoch": 0.03303055326176713,
"grad_norm": 2.515625,
"learning_rate": 5.2955082742316786e-05,
"loss": 2.5925,
"num_input_tokens_seen": 330301440,
"step": 280
},
{
"epoch": 0.033148519523416305,
"grad_norm": 2.40625,
"learning_rate": 5.314420803782506e-05,
"loss": 2.6826,
"num_input_tokens_seen": 331481088,
"step": 281
},
{
"epoch": 0.03326648578506547,
"grad_norm": 2.015625,
"learning_rate": 5.333333333333333e-05,
"loss": 2.6997,
"num_input_tokens_seen": 332660736,
"step": 282
},
{
"epoch": 0.03338445204671464,
"grad_norm": 2.234375,
"learning_rate": 5.352245862884161e-05,
"loss": 2.6378,
"num_input_tokens_seen": 333840384,
"step": 283
},
{
"epoch": 0.033502418308363806,
"grad_norm": 1.984375,
"learning_rate": 5.371158392434989e-05,
"loss": 2.588,
"num_input_tokens_seen": 335020032,
"step": 284
},
{
"epoch": 0.03362038457001298,
"grad_norm": 2.546875,
"learning_rate": 5.390070921985816e-05,
"loss": 2.6051,
"num_input_tokens_seen": 336199680,
"step": 285
},
{
"epoch": 0.03373835083166214,
"grad_norm": 1.7421875,
"learning_rate": 5.4089834515366434e-05,
"loss": 2.6874,
"num_input_tokens_seen": 337379328,
"step": 286
},
{
"epoch": 0.033856317093311314,
"grad_norm": 2.09375,
"learning_rate": 5.4278959810874704e-05,
"loss": 2.7186,
"num_input_tokens_seen": 338558976,
"step": 287
},
{
"epoch": 0.03397428335496048,
"grad_norm": 1.796875,
"learning_rate": 5.446808510638298e-05,
"loss": 2.6698,
"num_input_tokens_seen": 339738624,
"step": 288
},
{
"epoch": 0.03409224961660965,
"grad_norm": 2.3125,
"learning_rate": 5.465721040189126e-05,
"loss": 2.603,
"num_input_tokens_seen": 340918272,
"step": 289
},
{
"epoch": 0.034210215878258815,
"grad_norm": 2.140625,
"learning_rate": 5.484633569739953e-05,
"loss": 2.6479,
"num_input_tokens_seen": 342097920,
"step": 290
},
{
"epoch": 0.03432818213990799,
"grad_norm": 2.140625,
"learning_rate": 5.5035460992907805e-05,
"loss": 2.7037,
"num_input_tokens_seen": 343277568,
"step": 291
},
{
"epoch": 0.03444614840155715,
"grad_norm": 1.90625,
"learning_rate": 5.522458628841608e-05,
"loss": 2.681,
"num_input_tokens_seen": 344457216,
"step": 292
},
{
"epoch": 0.03456411466320632,
"grad_norm": 2.046875,
"learning_rate": 5.541371158392435e-05,
"loss": 2.5947,
"num_input_tokens_seen": 345636864,
"step": 293
},
{
"epoch": 0.03468208092485549,
"grad_norm": 1.734375,
"learning_rate": 5.560283687943263e-05,
"loss": 2.6439,
"num_input_tokens_seen": 346816512,
"step": 294
},
{
"epoch": 0.03480004718650466,
"grad_norm": 2.28125,
"learning_rate": 5.5791962174940906e-05,
"loss": 2.6404,
"num_input_tokens_seen": 347996160,
"step": 295
},
{
"epoch": 0.03491801344815383,
"grad_norm": 2.03125,
"learning_rate": 5.5981087470449176e-05,
"loss": 2.6686,
"num_input_tokens_seen": 349175808,
"step": 296
},
{
"epoch": 0.035035979709802996,
"grad_norm": 1.7265625,
"learning_rate": 5.617021276595745e-05,
"loss": 2.666,
"num_input_tokens_seen": 350355456,
"step": 297
},
{
"epoch": 0.03515394597145217,
"grad_norm": 2.109375,
"learning_rate": 5.635933806146573e-05,
"loss": 2.7266,
"num_input_tokens_seen": 351535104,
"step": 298
},
{
"epoch": 0.03527191223310133,
"grad_norm": 2.03125,
"learning_rate": 5.6548463356974e-05,
"loss": 2.6655,
"num_input_tokens_seen": 352714752,
"step": 299
},
{
"epoch": 0.035389878494750504,
"grad_norm": 1.9296875,
"learning_rate": 5.673758865248228e-05,
"loss": 2.6975,
"num_input_tokens_seen": 353894400,
"step": 300
},
{
"epoch": 0.03550784475639967,
"grad_norm": 2.15625,
"learning_rate": 5.692671394799055e-05,
"loss": 2.6232,
"num_input_tokens_seen": 355074048,
"step": 301
},
{
"epoch": 0.03562581101804884,
"grad_norm": 2.140625,
"learning_rate": 5.7115839243498824e-05,
"loss": 2.6344,
"num_input_tokens_seen": 356253696,
"step": 302
},
{
"epoch": 0.035743777279698005,
"grad_norm": 1.7109375,
"learning_rate": 5.73049645390071e-05,
"loss": 2.5703,
"num_input_tokens_seen": 357433344,
"step": 303
},
{
"epoch": 0.03586174354134718,
"grad_norm": 2.125,
"learning_rate": 5.749408983451537e-05,
"loss": 2.5405,
"num_input_tokens_seen": 358612992,
"step": 304
},
{
"epoch": 0.03597970980299634,
"grad_norm": 2.546875,
"learning_rate": 5.768321513002365e-05,
"loss": 2.6079,
"num_input_tokens_seen": 359792640,
"step": 305
},
{
"epoch": 0.03609767606464551,
"grad_norm": 2.015625,
"learning_rate": 5.7872340425531925e-05,
"loss": 2.6958,
"num_input_tokens_seen": 360972288,
"step": 306
},
{
"epoch": 0.03621564232629468,
"grad_norm": 1.46875,
"learning_rate": 5.8061465721040195e-05,
"loss": 2.5541,
"num_input_tokens_seen": 362151936,
"step": 307
},
{
"epoch": 0.03633360858794385,
"grad_norm": 1.796875,
"learning_rate": 5.825059101654847e-05,
"loss": 2.6427,
"num_input_tokens_seen": 363331584,
"step": 308
},
{
"epoch": 0.036451574849593014,
"grad_norm": 2.171875,
"learning_rate": 5.843971631205675e-05,
"loss": 2.5853,
"num_input_tokens_seen": 364511232,
"step": 309
},
{
"epoch": 0.036569541111242186,
"grad_norm": 1.8046875,
"learning_rate": 5.862884160756501e-05,
"loss": 2.5426,
"num_input_tokens_seen": 365690880,
"step": 310
},
{
"epoch": 0.03668750737289135,
"grad_norm": 1.625,
"learning_rate": 5.881796690307329e-05,
"loss": 2.6835,
"num_input_tokens_seen": 366870528,
"step": 311
},
{
"epoch": 0.03680547363454052,
"grad_norm": 2.640625,
"learning_rate": 5.900709219858156e-05,
"loss": 2.626,
"num_input_tokens_seen": 368050176,
"step": 312
},
{
"epoch": 0.03692343989618969,
"grad_norm": 1.4765625,
"learning_rate": 5.9196217494089836e-05,
"loss": 2.5361,
"num_input_tokens_seen": 369229824,
"step": 313
},
{
"epoch": 0.03704140615783886,
"grad_norm": 3.03125,
"learning_rate": 5.938534278959811e-05,
"loss": 2.5762,
"num_input_tokens_seen": 370409472,
"step": 314
},
{
"epoch": 0.037159372419488024,
"grad_norm": 1.890625,
"learning_rate": 5.9574468085106384e-05,
"loss": 2.5874,
"num_input_tokens_seen": 371589120,
"step": 315
},
{
"epoch": 0.037277338681137195,
"grad_norm": 2.6875,
"learning_rate": 5.976359338061466e-05,
"loss": 2.5874,
"num_input_tokens_seen": 372768768,
"step": 316
},
{
"epoch": 0.03739530494278636,
"grad_norm": 1.8359375,
"learning_rate": 5.995271867612294e-05,
"loss": 2.4888,
"num_input_tokens_seen": 373948416,
"step": 317
},
{
"epoch": 0.03751327120443553,
"grad_norm": 2.25,
"learning_rate": 6.014184397163121e-05,
"loss": 2.5391,
"num_input_tokens_seen": 375128064,
"step": 318
},
{
"epoch": 0.0376312374660847,
"grad_norm": 1.890625,
"learning_rate": 6.0330969267139484e-05,
"loss": 2.6462,
"num_input_tokens_seen": 376307712,
"step": 319
},
{
"epoch": 0.03774920372773387,
"grad_norm": 1.875,
"learning_rate": 6.0520094562647755e-05,
"loss": 2.4726,
"num_input_tokens_seen": 377487360,
"step": 320
},
{
"epoch": 0.03786716998938304,
"grad_norm": 2.296875,
"learning_rate": 6.070921985815603e-05,
"loss": 2.5759,
"num_input_tokens_seen": 378667008,
"step": 321
},
{
"epoch": 0.037985136251032205,
"grad_norm": 2.703125,
"learning_rate": 6.089834515366431e-05,
"loss": 2.5199,
"num_input_tokens_seen": 379846656,
"step": 322
},
{
"epoch": 0.038103102512681376,
"grad_norm": 2.390625,
"learning_rate": 6.108747044917259e-05,
"loss": 2.6243,
"num_input_tokens_seen": 381026304,
"step": 323
},
{
"epoch": 0.03822106877433054,
"grad_norm": 1.65625,
"learning_rate": 6.127659574468086e-05,
"loss": 2.5922,
"num_input_tokens_seen": 382205952,
"step": 324
},
{
"epoch": 0.03833903503597971,
"grad_norm": 1.640625,
"learning_rate": 6.146572104018913e-05,
"loss": 2.5376,
"num_input_tokens_seen": 383385600,
"step": 325
},
{
"epoch": 0.03845700129762888,
"grad_norm": 2.03125,
"learning_rate": 6.165484633569741e-05,
"loss": 2.547,
"num_input_tokens_seen": 384565248,
"step": 326
},
{
"epoch": 0.03857496755927805,
"grad_norm": 2.765625,
"learning_rate": 6.184397163120568e-05,
"loss": 2.5095,
"num_input_tokens_seen": 385744896,
"step": 327
},
{
"epoch": 0.038692933820927214,
"grad_norm": 1.40625,
"learning_rate": 6.203309692671395e-05,
"loss": 2.5431,
"num_input_tokens_seen": 386924544,
"step": 328
},
{
"epoch": 0.038810900082576386,
"grad_norm": 2.203125,
"learning_rate": 6.222222222222223e-05,
"loss": 2.5744,
"num_input_tokens_seen": 388104192,
"step": 329
},
{
"epoch": 0.03892886634422555,
"grad_norm": 1.9921875,
"learning_rate": 6.24113475177305e-05,
"loss": 2.6511,
"num_input_tokens_seen": 389283840,
"step": 330
},
{
"epoch": 0.03904683260587472,
"grad_norm": 3.15625,
"learning_rate": 6.260047281323877e-05,
"loss": 2.5556,
"num_input_tokens_seen": 390463488,
"step": 331
},
{
"epoch": 0.03916479886752389,
"grad_norm": 1.953125,
"learning_rate": 6.278959810874706e-05,
"loss": 2.5616,
"num_input_tokens_seen": 391643136,
"step": 332
},
{
"epoch": 0.03928276512917306,
"grad_norm": 1.6484375,
"learning_rate": 6.297872340425533e-05,
"loss": 2.5747,
"num_input_tokens_seen": 392822784,
"step": 333
},
{
"epoch": 0.03940073139082222,
"grad_norm": 3.15625,
"learning_rate": 6.31678486997636e-05,
"loss": 2.4747,
"num_input_tokens_seen": 394002432,
"step": 334
},
{
"epoch": 0.039518697652471395,
"grad_norm": 1.765625,
"learning_rate": 6.335697399527188e-05,
"loss": 2.5124,
"num_input_tokens_seen": 395182080,
"step": 335
},
{
"epoch": 0.03963666391412056,
"grad_norm": 2.15625,
"learning_rate": 6.354609929078015e-05,
"loss": 2.552,
"num_input_tokens_seen": 396361728,
"step": 336
},
{
"epoch": 0.03975463017576973,
"grad_norm": 1.8125,
"learning_rate": 6.373522458628842e-05,
"loss": 2.4541,
"num_input_tokens_seen": 397541376,
"step": 337
},
{
"epoch": 0.039872596437418896,
"grad_norm": 3.15625,
"learning_rate": 6.39243498817967e-05,
"loss": 2.4496,
"num_input_tokens_seen": 398721024,
"step": 338
},
{
"epoch": 0.03999056269906807,
"grad_norm": 2.015625,
"learning_rate": 6.411347517730498e-05,
"loss": 2.5182,
"num_input_tokens_seen": 399900672,
"step": 339
},
{
"epoch": 0.04010852896071723,
"grad_norm": 1.8671875,
"learning_rate": 6.430260047281325e-05,
"loss": 2.5118,
"num_input_tokens_seen": 401080320,
"step": 340
},
{
"epoch": 0.040226495222366404,
"grad_norm": 4.46875,
"learning_rate": 6.449172576832153e-05,
"loss": 2.5161,
"num_input_tokens_seen": 402259968,
"step": 341
},
{
"epoch": 0.04034446148401557,
"grad_norm": 1.8828125,
"learning_rate": 6.468085106382979e-05,
"loss": 2.5411,
"num_input_tokens_seen": 403439616,
"step": 342
},
{
"epoch": 0.04046242774566474,
"grad_norm": 6.53125,
"learning_rate": 6.486997635933806e-05,
"loss": 2.5341,
"num_input_tokens_seen": 404619264,
"step": 343
},
{
"epoch": 0.040580394007313905,
"grad_norm": 5.0,
"learning_rate": 6.505910165484634e-05,
"loss": 2.5139,
"num_input_tokens_seen": 405798912,
"step": 344
},
{
"epoch": 0.04069836026896308,
"grad_norm": 6.9375,
"learning_rate": 6.524822695035461e-05,
"loss": 2.542,
"num_input_tokens_seen": 406978560,
"step": 345
},
{
"epoch": 0.04081632653061224,
"grad_norm": 6.625,
"learning_rate": 6.543735224586288e-05,
"loss": 2.6481,
"num_input_tokens_seen": 408158208,
"step": 346
},
{
"epoch": 0.04093429279226141,
"grad_norm": 1.9375,
"learning_rate": 6.562647754137116e-05,
"loss": 2.5413,
"num_input_tokens_seen": 409337856,
"step": 347
},
{
"epoch": 0.041052259053910585,
"grad_norm": 3.25,
"learning_rate": 6.581560283687943e-05,
"loss": 2.5869,
"num_input_tokens_seen": 410517504,
"step": 348
},
{
"epoch": 0.04117022531555975,
"grad_norm": 2.171875,
"learning_rate": 6.60047281323877e-05,
"loss": 2.5717,
"num_input_tokens_seen": 411697152,
"step": 349
},
{
"epoch": 0.04128819157720892,
"grad_norm": 2.296875,
"learning_rate": 6.619385342789599e-05,
"loss": 2.569,
"num_input_tokens_seen": 412876800,
"step": 350
},
{
"epoch": 0.041406157838858086,
"grad_norm": 2.53125,
"learning_rate": 6.638297872340426e-05,
"loss": 2.5656,
"num_input_tokens_seen": 414056448,
"step": 351
},
{
"epoch": 0.04152412410050726,
"grad_norm": 1.5625,
"learning_rate": 6.657210401891253e-05,
"loss": 2.4558,
"num_input_tokens_seen": 415236096,
"step": 352
},
{
"epoch": 0.04164209036215642,
"grad_norm": 3.046875,
"learning_rate": 6.676122931442081e-05,
"loss": 2.5107,
"num_input_tokens_seen": 416415744,
"step": 353
},
{
"epoch": 0.041760056623805594,
"grad_norm": 2.03125,
"learning_rate": 6.695035460992908e-05,
"loss": 2.5576,
"num_input_tokens_seen": 417595392,
"step": 354
},
{
"epoch": 0.04187802288545476,
"grad_norm": 2.015625,
"learning_rate": 6.713947990543735e-05,
"loss": 2.5029,
"num_input_tokens_seen": 418775040,
"step": 355
},
{
"epoch": 0.04199598914710393,
"grad_norm": 1.84375,
"learning_rate": 6.732860520094564e-05,
"loss": 2.547,
"num_input_tokens_seen": 419954688,
"step": 356
},
{
"epoch": 0.042113955408753095,
"grad_norm": 1.9453125,
"learning_rate": 6.75177304964539e-05,
"loss": 2.4942,
"num_input_tokens_seen": 421134336,
"step": 357
},
{
"epoch": 0.04223192167040227,
"grad_norm": 2.109375,
"learning_rate": 6.770685579196218e-05,
"loss": 2.4546,
"num_input_tokens_seen": 422313984,
"step": 358
},
{
"epoch": 0.04234988793205143,
"grad_norm": 1.875,
"learning_rate": 6.789598108747046e-05,
"loss": 2.6115,
"num_input_tokens_seen": 423493632,
"step": 359
},
{
"epoch": 0.042467854193700603,
"grad_norm": 1.5,
"learning_rate": 6.808510638297873e-05,
"loss": 2.4977,
"num_input_tokens_seen": 424673280,
"step": 360
},
{
"epoch": 0.04258582045534977,
"grad_norm": 1.84375,
"learning_rate": 6.8274231678487e-05,
"loss": 2.4208,
"num_input_tokens_seen": 425852928,
"step": 361
},
{
"epoch": 0.04270378671699894,
"grad_norm": 1.421875,
"learning_rate": 6.846335697399528e-05,
"loss": 2.5416,
"num_input_tokens_seen": 427032576,
"step": 362
},
{
"epoch": 0.042821752978648105,
"grad_norm": 1.78125,
"learning_rate": 6.865248226950355e-05,
"loss": 2.4938,
"num_input_tokens_seen": 428212224,
"step": 363
},
{
"epoch": 0.042939719240297276,
"grad_norm": 1.5859375,
"learning_rate": 6.884160756501182e-05,
"loss": 2.5203,
"num_input_tokens_seen": 429391872,
"step": 364
},
{
"epoch": 0.04305768550194644,
"grad_norm": 1.9296875,
"learning_rate": 6.903073286052011e-05,
"loss": 2.515,
"num_input_tokens_seen": 430571520,
"step": 365
},
{
"epoch": 0.04317565176359561,
"grad_norm": 2.25,
"learning_rate": 6.921985815602838e-05,
"loss": 2.5211,
"num_input_tokens_seen": 431751168,
"step": 366
},
{
"epoch": 0.04329361802524478,
"grad_norm": 1.796875,
"learning_rate": 6.940898345153665e-05,
"loss": 2.4887,
"num_input_tokens_seen": 432930816,
"step": 367
},
{
"epoch": 0.04341158428689395,
"grad_norm": 1.5234375,
"learning_rate": 6.959810874704493e-05,
"loss": 2.4896,
"num_input_tokens_seen": 434110464,
"step": 368
},
{
"epoch": 0.043529550548543114,
"grad_norm": 1.6171875,
"learning_rate": 6.97872340425532e-05,
"loss": 2.4172,
"num_input_tokens_seen": 435290112,
"step": 369
},
{
"epoch": 0.043647516810192286,
"grad_norm": 2.1875,
"learning_rate": 6.997635933806147e-05,
"loss": 2.4538,
"num_input_tokens_seen": 436469760,
"step": 370
},
{
"epoch": 0.04376548307184145,
"grad_norm": 2.015625,
"learning_rate": 7.016548463356976e-05,
"loss": 2.4439,
"num_input_tokens_seen": 437649408,
"step": 371
},
{
"epoch": 0.04388344933349062,
"grad_norm": 1.828125,
"learning_rate": 7.035460992907803e-05,
"loss": 2.4898,
"num_input_tokens_seen": 438829056,
"step": 372
},
{
"epoch": 0.04400141559513979,
"grad_norm": 1.6171875,
"learning_rate": 7.05437352245863e-05,
"loss": 2.4337,
"num_input_tokens_seen": 440008704,
"step": 373
},
{
"epoch": 0.04411938185678896,
"grad_norm": 1.53125,
"learning_rate": 7.073286052009457e-05,
"loss": 2.4729,
"num_input_tokens_seen": 441188352,
"step": 374
},
{
"epoch": 0.04423734811843813,
"grad_norm": 1.9765625,
"learning_rate": 7.092198581560284e-05,
"loss": 2.4801,
"num_input_tokens_seen": 442368000,
"step": 375
},
{
"epoch": 0.044355314380087295,
"grad_norm": 2.0,
"learning_rate": 7.11111111111111e-05,
"loss": 2.5573,
"num_input_tokens_seen": 443547648,
"step": 376
},
{
"epoch": 0.044473280641736467,
"grad_norm": 1.640625,
"learning_rate": 7.130023640661939e-05,
"loss": 2.4675,
"num_input_tokens_seen": 444727296,
"step": 377
},
{
"epoch": 0.04459124690338563,
"grad_norm": 1.8984375,
"learning_rate": 7.148936170212766e-05,
"loss": 2.4965,
"num_input_tokens_seen": 445906944,
"step": 378
},
{
"epoch": 0.0447092131650348,
"grad_norm": 1.2578125,
"learning_rate": 7.167848699763593e-05,
"loss": 2.4161,
"num_input_tokens_seen": 447086592,
"step": 379
},
{
"epoch": 0.04482717942668397,
"grad_norm": 1.8359375,
"learning_rate": 7.186761229314421e-05,
"loss": 2.4437,
"num_input_tokens_seen": 448266240,
"step": 380
},
{
"epoch": 0.04494514568833314,
"grad_norm": 1.8046875,
"learning_rate": 7.205673758865248e-05,
"loss": 2.4978,
"num_input_tokens_seen": 449445888,
"step": 381
},
{
"epoch": 0.045063111949982304,
"grad_norm": 1.53125,
"learning_rate": 7.224586288416075e-05,
"loss": 2.4316,
"num_input_tokens_seen": 450625536,
"step": 382
},
{
"epoch": 0.045181078211631476,
"grad_norm": 2.0625,
"learning_rate": 7.243498817966904e-05,
"loss": 2.5152,
"num_input_tokens_seen": 451805184,
"step": 383
},
{
"epoch": 0.04529904447328064,
"grad_norm": 1.65625,
"learning_rate": 7.262411347517731e-05,
"loss": 2.4161,
"num_input_tokens_seen": 452984832,
"step": 384
},
{
"epoch": 0.04541701073492981,
"grad_norm": 1.8515625,
"learning_rate": 7.281323877068558e-05,
"loss": 2.4274,
"num_input_tokens_seen": 454164480,
"step": 385
},
{
"epoch": 0.04553497699657898,
"grad_norm": 2.0625,
"learning_rate": 7.300236406619386e-05,
"loss": 2.4693,
"num_input_tokens_seen": 455344128,
"step": 386
},
{
"epoch": 0.04565294325822815,
"grad_norm": 1.765625,
"learning_rate": 7.319148936170213e-05,
"loss": 2.4907,
"num_input_tokens_seen": 456523776,
"step": 387
},
{
"epoch": 0.04577090951987731,
"grad_norm": 1.71875,
"learning_rate": 7.33806146572104e-05,
"loss": 2.4603,
"num_input_tokens_seen": 457703424,
"step": 388
},
{
"epoch": 0.045888875781526485,
"grad_norm": 1.7734375,
"learning_rate": 7.356973995271869e-05,
"loss": 2.4651,
"num_input_tokens_seen": 458883072,
"step": 389
},
{
"epoch": 0.04600684204317565,
"grad_norm": 1.9765625,
"learning_rate": 7.375886524822696e-05,
"loss": 2.4041,
"num_input_tokens_seen": 460062720,
"step": 390
},
{
"epoch": 0.04612480830482482,
"grad_norm": 2.90625,
"learning_rate": 7.394799054373523e-05,
"loss": 2.3896,
"num_input_tokens_seen": 461242368,
"step": 391
},
{
"epoch": 0.046242774566473986,
"grad_norm": 1.515625,
"learning_rate": 7.413711583924351e-05,
"loss": 2.3966,
"num_input_tokens_seen": 462422016,
"step": 392
},
{
"epoch": 0.04636074082812316,
"grad_norm": 1.984375,
"learning_rate": 7.432624113475178e-05,
"loss": 2.5009,
"num_input_tokens_seen": 463601664,
"step": 393
},
{
"epoch": 0.04647870708977232,
"grad_norm": 1.8984375,
"learning_rate": 7.451536643026005e-05,
"loss": 2.4524,
"num_input_tokens_seen": 464781312,
"step": 394
},
{
"epoch": 0.046596673351421494,
"grad_norm": 2.46875,
"learning_rate": 7.470449172576833e-05,
"loss": 2.4169,
"num_input_tokens_seen": 465960960,
"step": 395
},
{
"epoch": 0.04671463961307066,
"grad_norm": 1.609375,
"learning_rate": 7.48936170212766e-05,
"loss": 2.5313,
"num_input_tokens_seen": 467140608,
"step": 396
},
{
"epoch": 0.04683260587471983,
"grad_norm": 1.7890625,
"learning_rate": 7.508274231678487e-05,
"loss": 2.5893,
"num_input_tokens_seen": 468320256,
"step": 397
},
{
"epoch": 0.046950572136368995,
"grad_norm": 1.9453125,
"learning_rate": 7.527186761229316e-05,
"loss": 2.453,
"num_input_tokens_seen": 469499904,
"step": 398
},
{
"epoch": 0.04706853839801817,
"grad_norm": 1.78125,
"learning_rate": 7.546099290780143e-05,
"loss": 2.4462,
"num_input_tokens_seen": 470679552,
"step": 399
},
{
"epoch": 0.04718650465966733,
"grad_norm": 2.140625,
"learning_rate": 7.56501182033097e-05,
"loss": 2.5393,
"num_input_tokens_seen": 471859200,
"step": 400
},
{
"epoch": 0.04718650465966733,
"eval_wikipedia_loss": 2.4841043949127197,
"eval_wikipedia_runtime": 173.8643,
"eval_wikipedia_samples_per_second": 4.038,
"eval_wikipedia_steps_per_second": 0.173,
"num_input_tokens_seen": 471859200,
"step": 400
},
{
"epoch": 0.04718650465966733,
"eval_toxicity_loss": 4.2398834228515625,
"eval_toxicity_runtime": 0.9943,
"eval_toxicity_samples_per_second": 2.011,
"eval_toxicity_steps_per_second": 1.006,
"num_input_tokens_seen": 471859200,
"step": 400
},
{
"epoch": 0.047304470921316503,
"grad_norm": 1.9609375,
"learning_rate": 7.583924349881798e-05,
"loss": 2.4697,
"num_input_tokens_seen": 473038848,
"step": 401
},
{
"epoch": 0.047422437182965675,
"grad_norm": 1.703125,
"learning_rate": 7.602836879432625e-05,
"loss": 2.4191,
"num_input_tokens_seen": 474218496,
"step": 402
},
{
"epoch": 0.04754040344461484,
"grad_norm": 1.2421875,
"learning_rate": 7.621749408983452e-05,
"loss": 2.3892,
"num_input_tokens_seen": 475398144,
"step": 403
},
{
"epoch": 0.04765836970626401,
"grad_norm": 1.6640625,
"learning_rate": 7.64066193853428e-05,
"loss": 2.4324,
"num_input_tokens_seen": 476577792,
"step": 404
},
{
"epoch": 0.047776335967913176,
"grad_norm": 2.015625,
"learning_rate": 7.659574468085108e-05,
"loss": 2.4291,
"num_input_tokens_seen": 477757440,
"step": 405
},
{
"epoch": 0.04789430222956235,
"grad_norm": 2.65625,
"learning_rate": 7.678486997635935e-05,
"loss": 2.4577,
"num_input_tokens_seen": 478937088,
"step": 406
},
{
"epoch": 0.04801226849121151,
"grad_norm": 1.2265625,
"learning_rate": 7.697399527186762e-05,
"loss": 2.4554,
"num_input_tokens_seen": 480116736,
"step": 407
},
{
"epoch": 0.048130234752860684,
"grad_norm": 3.453125,
"learning_rate": 7.716312056737589e-05,
"loss": 2.3692,
"num_input_tokens_seen": 481296384,
"step": 408
},
{
"epoch": 0.04824820101450985,
"grad_norm": 1.953125,
"learning_rate": 7.735224586288416e-05,
"loss": 2.518,
"num_input_tokens_seen": 482476032,
"step": 409
},
{
"epoch": 0.04836616727615902,
"grad_norm": 2.046875,
"learning_rate": 7.754137115839244e-05,
"loss": 2.495,
"num_input_tokens_seen": 483655680,
"step": 410
},
{
"epoch": 0.048484133537808186,
"grad_norm": 1.9765625,
"learning_rate": 7.773049645390071e-05,
"loss": 2.4344,
"num_input_tokens_seen": 484835328,
"step": 411
},
{
"epoch": 0.04860209979945736,
"grad_norm": 1.9296875,
"learning_rate": 7.791962174940898e-05,
"loss": 2.4095,
"num_input_tokens_seen": 486014976,
"step": 412
},
{
"epoch": 0.04872006606110652,
"grad_norm": 2.234375,
"learning_rate": 7.810874704491726e-05,
"loss": 2.4562,
"num_input_tokens_seen": 487194624,
"step": 413
},
{
"epoch": 0.048838032322755694,
"grad_norm": 1.2109375,
"learning_rate": 7.829787234042553e-05,
"loss": 2.3863,
"num_input_tokens_seen": 488374272,
"step": 414
},
{
"epoch": 0.04895599858440486,
"grad_norm": 1.96875,
"learning_rate": 7.84869976359338e-05,
"loss": 2.4577,
"num_input_tokens_seen": 489553920,
"step": 415
},
{
"epoch": 0.04907396484605403,
"grad_norm": 1.375,
"learning_rate": 7.867612293144209e-05,
"loss": 2.4166,
"num_input_tokens_seen": 490733568,
"step": 416
},
{
"epoch": 0.049191931107703195,
"grad_norm": 1.453125,
"learning_rate": 7.886524822695036e-05,
"loss": 2.4524,
"num_input_tokens_seen": 491913216,
"step": 417
},
{
"epoch": 0.049309897369352367,
"grad_norm": 1.390625,
"learning_rate": 7.905437352245863e-05,
"loss": 2.396,
"num_input_tokens_seen": 493092864,
"step": 418
},
{
"epoch": 0.04942786363100153,
"grad_norm": 1.21875,
"learning_rate": 7.924349881796691e-05,
"loss": 2.3969,
"num_input_tokens_seen": 494272512,
"step": 419
},
{
"epoch": 0.0495458298926507,
"grad_norm": 1.625,
"learning_rate": 7.943262411347518e-05,
"loss": 2.4247,
"num_input_tokens_seen": 495452160,
"step": 420
},
{
"epoch": 0.04966379615429987,
"grad_norm": 2.3125,
"learning_rate": 7.962174940898345e-05,
"loss": 2.5668,
"num_input_tokens_seen": 496631808,
"step": 421
},
{
"epoch": 0.04978176241594904,
"grad_norm": 2.078125,
"learning_rate": 7.981087470449174e-05,
"loss": 2.3802,
"num_input_tokens_seen": 497811456,
"step": 422
},
{
"epoch": 0.049899728677598204,
"grad_norm": 1.1953125,
"learning_rate": 8e-05,
"loss": 2.4226,
"num_input_tokens_seen": 498991104,
"step": 423
},
{
"epoch": 0.050017694939247376,
"grad_norm": 1.453125,
"learning_rate": 7.999999695696824e-05,
"loss": 2.4269,
"num_input_tokens_seen": 500170752,
"step": 424
},
{
"epoch": 0.05013566120089654,
"grad_norm": 1.4609375,
"learning_rate": 7.999998782787343e-05,
"loss": 2.4702,
"num_input_tokens_seen": 501350400,
"step": 425
},
{
"epoch": 0.05025362746254571,
"grad_norm": 1.96875,
"learning_rate": 7.999997261271695e-05,
"loss": 2.3265,
"num_input_tokens_seen": 502530048,
"step": 426
},
{
"epoch": 0.05037159372419488,
"grad_norm": 2.265625,
"learning_rate": 7.999995131150112e-05,
"loss": 2.4,
"num_input_tokens_seen": 503709696,
"step": 427
},
{
"epoch": 0.05048955998584405,
"grad_norm": 1.3359375,
"learning_rate": 7.999992392422917e-05,
"loss": 2.4076,
"num_input_tokens_seen": 504889344,
"step": 428
},
{
"epoch": 0.05060752624749322,
"grad_norm": 1.140625,
"learning_rate": 7.999989045090529e-05,
"loss": 2.3864,
"num_input_tokens_seen": 506068992,
"step": 429
},
{
"epoch": 0.050725492509142385,
"grad_norm": 1.4765625,
"learning_rate": 7.999985089153455e-05,
"loss": 2.4194,
"num_input_tokens_seen": 507248640,
"step": 430
},
{
"epoch": 0.05084345877079156,
"grad_norm": 2.0625,
"learning_rate": 7.999980524612299e-05,
"loss": 2.4248,
"num_input_tokens_seen": 508428288,
"step": 431
},
{
"epoch": 0.05096142503244072,
"grad_norm": 1.953125,
"learning_rate": 7.999975351467752e-05,
"loss": 2.3898,
"num_input_tokens_seen": 509607936,
"step": 432
},
{
"epoch": 0.05107939129408989,
"grad_norm": 1.421875,
"learning_rate": 7.999969569720605e-05,
"loss": 2.3805,
"num_input_tokens_seen": 510787584,
"step": 433
},
{
"epoch": 0.05119735755573906,
"grad_norm": 1.1328125,
"learning_rate": 7.999963179371735e-05,
"loss": 2.4378,
"num_input_tokens_seen": 511967232,
"step": 434
},
{
"epoch": 0.05131532381738823,
"grad_norm": 1.3984375,
"learning_rate": 7.999956180422118e-05,
"loss": 2.4174,
"num_input_tokens_seen": 513146880,
"step": 435
},
{
"epoch": 0.051433290079037394,
"grad_norm": 1.6171875,
"learning_rate": 7.999948572872813e-05,
"loss": 2.3132,
"num_input_tokens_seen": 514326528,
"step": 436
},
{
"epoch": 0.051551256340686566,
"grad_norm": 2.0,
"learning_rate": 7.999940356724983e-05,
"loss": 2.3784,
"num_input_tokens_seen": 515506176,
"step": 437
},
{
"epoch": 0.05166922260233573,
"grad_norm": 1.3359375,
"learning_rate": 7.999931531979876e-05,
"loss": 2.3361,
"num_input_tokens_seen": 516685824,
"step": 438
},
{
"epoch": 0.0517871888639849,
"grad_norm": 1.328125,
"learning_rate": 7.999922098638835e-05,
"loss": 2.4646,
"num_input_tokens_seen": 517865472,
"step": 439
},
{
"epoch": 0.05190515512563407,
"grad_norm": 1.453125,
"learning_rate": 7.999912056703294e-05,
"loss": 2.3374,
"num_input_tokens_seen": 519045120,
"step": 440
},
{
"epoch": 0.05202312138728324,
"grad_norm": 1.9296875,
"learning_rate": 7.999901406174781e-05,
"loss": 2.3778,
"num_input_tokens_seen": 520224768,
"step": 441
},
{
"epoch": 0.052141087648932403,
"grad_norm": 1.3203125,
"learning_rate": 7.999890147054918e-05,
"loss": 2.328,
"num_input_tokens_seen": 521404416,
"step": 442
},
{
"epoch": 0.052259053910581575,
"grad_norm": 1.453125,
"learning_rate": 7.999878279345418e-05,
"loss": 2.3145,
"num_input_tokens_seen": 522584064,
"step": 443
},
{
"epoch": 0.05237702017223074,
"grad_norm": 1.3046875,
"learning_rate": 7.999865803048087e-05,
"loss": 2.3785,
"num_input_tokens_seen": 523763712,
"step": 444
},
{
"epoch": 0.05249498643387991,
"grad_norm": 1.3359375,
"learning_rate": 7.99985271816482e-05,
"loss": 2.3462,
"num_input_tokens_seen": 524943360,
"step": 445
},
{
"epoch": 0.052612952695529076,
"grad_norm": 1.453125,
"learning_rate": 7.999839024697612e-05,
"loss": 2.5261,
"num_input_tokens_seen": 526123008,
"step": 446
},
{
"epoch": 0.05273091895717825,
"grad_norm": 1.4140625,
"learning_rate": 7.999824722648545e-05,
"loss": 2.3022,
"num_input_tokens_seen": 527302656,
"step": 447
},
{
"epoch": 0.05284888521882741,
"grad_norm": 1.3359375,
"learning_rate": 7.999809812019794e-05,
"loss": 2.4073,
"num_input_tokens_seen": 528482304,
"step": 448
},
{
"epoch": 0.052966851480476584,
"grad_norm": 1.1015625,
"learning_rate": 7.999794292813627e-05,
"loss": 2.3876,
"num_input_tokens_seen": 529661952,
"step": 449
},
{
"epoch": 0.05308481774212575,
"grad_norm": 1.8203125,
"learning_rate": 7.99977816503241e-05,
"loss": 2.3443,
"num_input_tokens_seen": 530841600,
"step": 450
},
{
"epoch": 0.05320278400377492,
"grad_norm": 2.375,
"learning_rate": 7.999761428678591e-05,
"loss": 2.4333,
"num_input_tokens_seen": 532021248,
"step": 451
},
{
"epoch": 0.053320750265424086,
"grad_norm": 1.796875,
"learning_rate": 7.999744083754721e-05,
"loss": 2.3346,
"num_input_tokens_seen": 533200896,
"step": 452
},
{
"epoch": 0.05343871652707326,
"grad_norm": 1.046875,
"learning_rate": 7.999726130263437e-05,
"loss": 2.356,
"num_input_tokens_seen": 534380544,
"step": 453
},
{
"epoch": 0.05355668278872242,
"grad_norm": 1.578125,
"learning_rate": 7.99970756820747e-05,
"loss": 2.2924,
"num_input_tokens_seen": 535560192,
"step": 454
},
{
"epoch": 0.053674649050371594,
"grad_norm": 1.9375,
"learning_rate": 7.999688397589647e-05,
"loss": 2.3714,
"num_input_tokens_seen": 536739840,
"step": 455
},
{
"epoch": 0.053792615312020765,
"grad_norm": 1.546875,
"learning_rate": 7.99966861841288e-05,
"loss": 2.3435,
"num_input_tokens_seen": 537919488,
"step": 456
},
{
"epoch": 0.05391058157366993,
"grad_norm": 1.4375,
"learning_rate": 7.999648230680182e-05,
"loss": 2.4296,
"num_input_tokens_seen": 539099136,
"step": 457
},
{
"epoch": 0.0540285478353191,
"grad_norm": 1.2890625,
"learning_rate": 7.999627234394656e-05,
"loss": 2.4,
"num_input_tokens_seen": 540278784,
"step": 458
},
{
"epoch": 0.054146514096968267,
"grad_norm": 1.5625,
"learning_rate": 7.999605629559493e-05,
"loss": 2.3882,
"num_input_tokens_seen": 541458432,
"step": 459
},
{
"epoch": 0.05426448035861744,
"grad_norm": 1.5234375,
"learning_rate": 7.999583416177985e-05,
"loss": 2.4006,
"num_input_tokens_seen": 542638080,
"step": 460
},
{
"epoch": 0.0543824466202666,
"grad_norm": 1.953125,
"learning_rate": 7.999560594253506e-05,
"loss": 2.3288,
"num_input_tokens_seen": 543817728,
"step": 461
},
{
"epoch": 0.054500412881915775,
"grad_norm": 1.2421875,
"learning_rate": 7.999537163789534e-05,
"loss": 2.3453,
"num_input_tokens_seen": 544997376,
"step": 462
},
{
"epoch": 0.05461837914356494,
"grad_norm": 1.015625,
"learning_rate": 7.999513124789631e-05,
"loss": 2.3976,
"num_input_tokens_seen": 546177024,
"step": 463
},
{
"epoch": 0.05473634540521411,
"grad_norm": 1.0078125,
"learning_rate": 7.999488477257453e-05,
"loss": 2.3466,
"num_input_tokens_seen": 547356672,
"step": 464
},
{
"epoch": 0.054854311666863276,
"grad_norm": 1.1171875,
"learning_rate": 7.999463221196753e-05,
"loss": 2.3788,
"num_input_tokens_seen": 548536320,
"step": 465
},
{
"epoch": 0.05497227792851245,
"grad_norm": 1.1171875,
"learning_rate": 7.999437356611375e-05,
"loss": 2.2683,
"num_input_tokens_seen": 549715968,
"step": 466
},
{
"epoch": 0.05509024419016161,
"grad_norm": 1.109375,
"learning_rate": 7.99941088350525e-05,
"loss": 2.2787,
"num_input_tokens_seen": 550895616,
"step": 467
},
{
"epoch": 0.055208210451810784,
"grad_norm": 1.5,
"learning_rate": 7.999383801882408e-05,
"loss": 2.3686,
"num_input_tokens_seen": 552075264,
"step": 468
},
{
"epoch": 0.05532617671345995,
"grad_norm": 1.8203125,
"learning_rate": 7.99935611174697e-05,
"loss": 2.283,
"num_input_tokens_seen": 553254912,
"step": 469
},
{
"epoch": 0.05544414297510912,
"grad_norm": 1.2890625,
"learning_rate": 7.999327813103149e-05,
"loss": 2.3282,
"num_input_tokens_seen": 554434560,
"step": 470
},
{
"epoch": 0.055562109236758285,
"grad_norm": 1.390625,
"learning_rate": 7.999298905955251e-05,
"loss": 2.5573,
"num_input_tokens_seen": 555614208,
"step": 471
},
{
"epoch": 0.05568007549840746,
"grad_norm": 1.1640625,
"learning_rate": 7.999269390307672e-05,
"loss": 2.3396,
"num_input_tokens_seen": 556793856,
"step": 472
},
{
"epoch": 0.05579804176005662,
"grad_norm": 1.484375,
"learning_rate": 7.999239266164906e-05,
"loss": 2.424,
"num_input_tokens_seen": 557973504,
"step": 473
},
{
"epoch": 0.05591600802170579,
"grad_norm": 1.40625,
"learning_rate": 7.999208533531534e-05,
"loss": 2.2834,
"num_input_tokens_seen": 559153152,
"step": 474
},
{
"epoch": 0.05603397428335496,
"grad_norm": 1.1953125,
"learning_rate": 7.999177192412233e-05,
"loss": 2.3158,
"num_input_tokens_seen": 560332800,
"step": 475
},
{
"epoch": 0.05615194054500413,
"grad_norm": 1.5703125,
"learning_rate": 7.999145242811773e-05,
"loss": 2.3454,
"num_input_tokens_seen": 561512448,
"step": 476
},
{
"epoch": 0.056269906806653294,
"grad_norm": 1.796875,
"learning_rate": 7.999112684735013e-05,
"loss": 2.3211,
"num_input_tokens_seen": 562692096,
"step": 477
},
{
"epoch": 0.056387873068302466,
"grad_norm": 1.046875,
"learning_rate": 7.999079518186908e-05,
"loss": 2.2571,
"num_input_tokens_seen": 563871744,
"step": 478
},
{
"epoch": 0.05650583932995163,
"grad_norm": 1.4140625,
"learning_rate": 7.999045743172504e-05,
"loss": 2.2938,
"num_input_tokens_seen": 565051392,
"step": 479
},
{
"epoch": 0.0566238055916008,
"grad_norm": 1.6328125,
"learning_rate": 7.99901135969694e-05,
"loss": 2.4742,
"num_input_tokens_seen": 566231040,
"step": 480
},
{
"epoch": 0.05674177185324997,
"grad_norm": 1.484375,
"learning_rate": 7.998976367765446e-05,
"loss": 2.3188,
"num_input_tokens_seen": 567410688,
"step": 481
},
{
"epoch": 0.05685973811489914,
"grad_norm": 1.7109375,
"learning_rate": 7.998940767383348e-05,
"loss": 2.2454,
"num_input_tokens_seen": 568590336,
"step": 482
},
{
"epoch": 0.05697770437654831,
"grad_norm": 1.1328125,
"learning_rate": 7.998904558556062e-05,
"loss": 2.3923,
"num_input_tokens_seen": 569769984,
"step": 483
},
{
"epoch": 0.057095670638197475,
"grad_norm": 1.421875,
"learning_rate": 7.998867741289097e-05,
"loss": 2.3465,
"num_input_tokens_seen": 570949632,
"step": 484
},
{
"epoch": 0.05721363689984665,
"grad_norm": 1.0078125,
"learning_rate": 7.998830315588056e-05,
"loss": 2.3016,
"num_input_tokens_seen": 572129280,
"step": 485
},
{
"epoch": 0.05733160316149581,
"grad_norm": 1.734375,
"learning_rate": 7.998792281458631e-05,
"loss": 2.2886,
"num_input_tokens_seen": 573308928,
"step": 486
},
{
"epoch": 0.05744956942314498,
"grad_norm": 1.140625,
"learning_rate": 7.998753638906613e-05,
"loss": 2.3398,
"num_input_tokens_seen": 574488576,
"step": 487
},
{
"epoch": 0.05756753568479415,
"grad_norm": 2.0625,
"learning_rate": 7.998714387937875e-05,
"loss": 2.3558,
"num_input_tokens_seen": 575668224,
"step": 488
},
{
"epoch": 0.05768550194644332,
"grad_norm": 1.7578125,
"learning_rate": 7.998674528558397e-05,
"loss": 2.3123,
"num_input_tokens_seen": 576847872,
"step": 489
},
{
"epoch": 0.057803468208092484,
"grad_norm": 1.109375,
"learning_rate": 7.998634060774239e-05,
"loss": 2.4113,
"num_input_tokens_seen": 578027520,
"step": 490
},
{
"epoch": 0.057921434469741656,
"grad_norm": 2.125,
"learning_rate": 7.998592984591557e-05,
"loss": 2.3259,
"num_input_tokens_seen": 579207168,
"step": 491
},
{
"epoch": 0.05803940073139082,
"grad_norm": 1.390625,
"learning_rate": 7.998551300016603e-05,
"loss": 2.39,
"num_input_tokens_seen": 580386816,
"step": 492
},
{
"epoch": 0.05815736699303999,
"grad_norm": 1.078125,
"learning_rate": 7.99850900705572e-05,
"loss": 2.3108,
"num_input_tokens_seen": 581566464,
"step": 493
},
{
"epoch": 0.05827533325468916,
"grad_norm": 1.0625,
"learning_rate": 7.998466105715342e-05,
"loss": 2.2948,
"num_input_tokens_seen": 582746112,
"step": 494
},
{
"epoch": 0.05839329951633833,
"grad_norm": 1.0390625,
"learning_rate": 7.998422596001997e-05,
"loss": 2.3144,
"num_input_tokens_seen": 583925760,
"step": 495
},
{
"epoch": 0.058511265777987494,
"grad_norm": 1.2734375,
"learning_rate": 7.998378477922303e-05,
"loss": 2.3439,
"num_input_tokens_seen": 585105408,
"step": 496
},
{
"epoch": 0.058629232039636665,
"grad_norm": 1.0546875,
"learning_rate": 7.998333751482976e-05,
"loss": 2.3061,
"num_input_tokens_seen": 586285056,
"step": 497
},
{
"epoch": 0.05874719830128583,
"grad_norm": 1.0703125,
"learning_rate": 7.99828841669082e-05,
"loss": 2.2375,
"num_input_tokens_seen": 587464704,
"step": 498
},
{
"epoch": 0.058865164562935,
"grad_norm": 1.078125,
"learning_rate": 7.998242473552732e-05,
"loss": 2.2959,
"num_input_tokens_seen": 588644352,
"step": 499
},
{
"epoch": 0.058983130824584167,
"grad_norm": 1.546875,
"learning_rate": 7.9981959220757e-05,
"loss": 2.3324,
"num_input_tokens_seen": 589824000,
"step": 500
},
{
"epoch": 0.05910109708623334,
"grad_norm": 1.7734375,
"learning_rate": 7.998148762266812e-05,
"loss": 2.267,
"num_input_tokens_seen": 591003648,
"step": 501
},
{
"epoch": 0.0592190633478825,
"grad_norm": 1.0859375,
"learning_rate": 7.99810099413324e-05,
"loss": 2.376,
"num_input_tokens_seen": 592183296,
"step": 502
},
{
"epoch": 0.059337029609531675,
"grad_norm": 0.91015625,
"learning_rate": 7.998052617682253e-05,
"loss": 2.2746,
"num_input_tokens_seen": 593362944,
"step": 503
},
{
"epoch": 0.05945499587118084,
"grad_norm": 1.1015625,
"learning_rate": 7.998003632921212e-05,
"loss": 2.3335,
"num_input_tokens_seen": 594542592,
"step": 504
},
{
"epoch": 0.05957296213283001,
"grad_norm": 1.4921875,
"learning_rate": 7.997954039857569e-05,
"loss": 2.2334,
"num_input_tokens_seen": 595722240,
"step": 505
},
{
"epoch": 0.059690928394479176,
"grad_norm": 1.578125,
"learning_rate": 7.99790383849887e-05,
"loss": 2.3015,
"num_input_tokens_seen": 596901888,
"step": 506
},
{
"epoch": 0.05980889465612835,
"grad_norm": 1.390625,
"learning_rate": 7.997853028852755e-05,
"loss": 2.2773,
"num_input_tokens_seen": 598081536,
"step": 507
},
{
"epoch": 0.05992686091777751,
"grad_norm": 1.0,
"learning_rate": 7.997801610926952e-05,
"loss": 2.2407,
"num_input_tokens_seen": 599261184,
"step": 508
},
{
"epoch": 0.060044827179426684,
"grad_norm": 1.0625,
"learning_rate": 7.997749584729286e-05,
"loss": 2.4037,
"num_input_tokens_seen": 600440832,
"step": 509
},
{
"epoch": 0.060162793441075856,
"grad_norm": 1.3125,
"learning_rate": 7.997696950267672e-05,
"loss": 2.3178,
"num_input_tokens_seen": 601620480,
"step": 510
},
{
"epoch": 0.06028075970272502,
"grad_norm": 1.53125,
"learning_rate": 7.997643707550117e-05,
"loss": 2.2891,
"num_input_tokens_seen": 602800128,
"step": 511
},
{
"epoch": 0.06039872596437419,
"grad_norm": 1.4765625,
"learning_rate": 7.997589856584725e-05,
"loss": 2.3047,
"num_input_tokens_seen": 603979776,
"step": 512
},
{
"epoch": 0.06051669222602336,
"grad_norm": 1.0234375,
"learning_rate": 7.99753539737969e-05,
"loss": 2.2871,
"num_input_tokens_seen": 605159424,
"step": 513
},
{
"epoch": 0.06063465848767253,
"grad_norm": 1.796875,
"learning_rate": 7.997480329943294e-05,
"loss": 2.2966,
"num_input_tokens_seen": 606339072,
"step": 514
},
{
"epoch": 0.06075262474932169,
"grad_norm": 0.95703125,
"learning_rate": 7.99742465428392e-05,
"loss": 2.1938,
"num_input_tokens_seen": 607518720,
"step": 515
},
{
"epoch": 0.060870591010970865,
"grad_norm": 0.87890625,
"learning_rate": 7.997368370410035e-05,
"loss": 2.2878,
"num_input_tokens_seen": 608698368,
"step": 516
},
{
"epoch": 0.06098855727262003,
"grad_norm": 1.046875,
"learning_rate": 7.997311478330206e-05,
"loss": 2.2645,
"num_input_tokens_seen": 609878016,
"step": 517
},
{
"epoch": 0.0611065235342692,
"grad_norm": 0.99609375,
"learning_rate": 7.997253978053087e-05,
"loss": 2.2939,
"num_input_tokens_seen": 611057664,
"step": 518
},
{
"epoch": 0.061224489795918366,
"grad_norm": 0.99609375,
"learning_rate": 7.997195869587428e-05,
"loss": 2.2763,
"num_input_tokens_seen": 612237312,
"step": 519
},
{
"epoch": 0.06134245605756754,
"grad_norm": 1.671875,
"learning_rate": 7.997137152942071e-05,
"loss": 2.3172,
"num_input_tokens_seen": 613416960,
"step": 520
},
{
"epoch": 0.0614604223192167,
"grad_norm": 1.8203125,
"learning_rate": 7.997077828125948e-05,
"loss": 2.2249,
"num_input_tokens_seen": 614596608,
"step": 521
},
{
"epoch": 0.061578388580865874,
"grad_norm": 0.78515625,
"learning_rate": 7.997017895148088e-05,
"loss": 2.3344,
"num_input_tokens_seen": 615776256,
"step": 522
},
{
"epoch": 0.06169635484251504,
"grad_norm": 1.90625,
"learning_rate": 7.996957354017605e-05,
"loss": 2.2967,
"num_input_tokens_seen": 616955904,
"step": 523
},
{
"epoch": 0.06181432110416421,
"grad_norm": 1.4765625,
"learning_rate": 7.996896204743716e-05,
"loss": 2.2287,
"num_input_tokens_seen": 618135552,
"step": 524
},
{
"epoch": 0.061932287365813375,
"grad_norm": 0.8671875,
"learning_rate": 7.996834447335722e-05,
"loss": 2.2897,
"num_input_tokens_seen": 619315200,
"step": 525
},
{
"epoch": 0.06205025362746255,
"grad_norm": 1.2578125,
"learning_rate": 7.99677208180302e-05,
"loss": 2.3208,
"num_input_tokens_seen": 620494848,
"step": 526
},
{
"epoch": 0.06216821988911171,
"grad_norm": 2.078125,
"learning_rate": 7.996709108155098e-05,
"loss": 2.2778,
"num_input_tokens_seen": 621674496,
"step": 527
},
{
"epoch": 0.06228618615076088,
"grad_norm": 0.91015625,
"learning_rate": 7.99664552640154e-05,
"loss": 2.1945,
"num_input_tokens_seen": 622854144,
"step": 528
},
{
"epoch": 0.06240415241241005,
"grad_norm": 2.53125,
"learning_rate": 7.996581336552018e-05,
"loss": 2.3752,
"num_input_tokens_seen": 624033792,
"step": 529
},
{
"epoch": 0.06252211867405921,
"grad_norm": 1.140625,
"learning_rate": 7.996516538616299e-05,
"loss": 2.3217,
"num_input_tokens_seen": 625213440,
"step": 530
},
{
"epoch": 0.06264008493570838,
"grad_norm": 1.890625,
"learning_rate": 7.996451132604242e-05,
"loss": 2.3374,
"num_input_tokens_seen": 626393088,
"step": 531
},
{
"epoch": 0.06275805119735756,
"grad_norm": 1.328125,
"learning_rate": 7.9963851185258e-05,
"loss": 2.2926,
"num_input_tokens_seen": 627572736,
"step": 532
},
{
"epoch": 0.06287601745900673,
"grad_norm": 1.1328125,
"learning_rate": 7.996318496391015e-05,
"loss": 2.2946,
"num_input_tokens_seen": 628752384,
"step": 533
},
{
"epoch": 0.06299398372065589,
"grad_norm": 1.4375,
"learning_rate": 7.996251266210025e-05,
"loss": 2.2977,
"num_input_tokens_seen": 629932032,
"step": 534
},
{
"epoch": 0.06311194998230506,
"grad_norm": 1.2109375,
"learning_rate": 7.996183427993058e-05,
"loss": 2.2898,
"num_input_tokens_seen": 631111680,
"step": 535
},
{
"epoch": 0.06322991624395423,
"grad_norm": 1.203125,
"learning_rate": 7.996114981750438e-05,
"loss": 2.2557,
"num_input_tokens_seen": 632291328,
"step": 536
},
{
"epoch": 0.0633478825056034,
"grad_norm": 1.0546875,
"learning_rate": 7.996045927492578e-05,
"loss": 2.3759,
"num_input_tokens_seen": 633470976,
"step": 537
},
{
"epoch": 0.06346584876725257,
"grad_norm": 1.1015625,
"learning_rate": 7.995976265229984e-05,
"loss": 2.2283,
"num_input_tokens_seen": 634650624,
"step": 538
},
{
"epoch": 0.06358381502890173,
"grad_norm": 1.1875,
"learning_rate": 7.995905994973257e-05,
"loss": 2.4155,
"num_input_tokens_seen": 635830272,
"step": 539
},
{
"epoch": 0.0637017812905509,
"grad_norm": 1.2890625,
"learning_rate": 7.995835116733086e-05,
"loss": 2.3187,
"num_input_tokens_seen": 637009920,
"step": 540
},
{
"epoch": 0.06381974755220007,
"grad_norm": 1.671875,
"learning_rate": 7.995763630520257e-05,
"loss": 2.2263,
"num_input_tokens_seen": 638189568,
"step": 541
},
{
"epoch": 0.06393771381384925,
"grad_norm": 1.0546875,
"learning_rate": 7.995691536345647e-05,
"loss": 2.2788,
"num_input_tokens_seen": 639369216,
"step": 542
},
{
"epoch": 0.0640556800754984,
"grad_norm": 1.0703125,
"learning_rate": 7.995618834220223e-05,
"loss": 2.2173,
"num_input_tokens_seen": 640548864,
"step": 543
},
{
"epoch": 0.06417364633714757,
"grad_norm": 1.1953125,
"learning_rate": 7.99554552415505e-05,
"loss": 2.2742,
"num_input_tokens_seen": 641728512,
"step": 544
},
{
"epoch": 0.06429161259879675,
"grad_norm": 1.453125,
"learning_rate": 7.99547160616128e-05,
"loss": 2.2516,
"num_input_tokens_seen": 642908160,
"step": 545
},
{
"epoch": 0.06440957886044592,
"grad_norm": 1.0078125,
"learning_rate": 7.995397080250162e-05,
"loss": 2.33,
"num_input_tokens_seen": 644087808,
"step": 546
},
{
"epoch": 0.06452754512209508,
"grad_norm": 1.0703125,
"learning_rate": 7.995321946433033e-05,
"loss": 2.2447,
"num_input_tokens_seen": 645267456,
"step": 547
},
{
"epoch": 0.06464551138374425,
"grad_norm": 1.7421875,
"learning_rate": 7.995246204721324e-05,
"loss": 2.369,
"num_input_tokens_seen": 646447104,
"step": 548
},
{
"epoch": 0.06476347764539342,
"grad_norm": 1.25,
"learning_rate": 7.995169855126561e-05,
"loss": 2.2753,
"num_input_tokens_seen": 647626752,
"step": 549
},
{
"epoch": 0.06488144390704259,
"grad_norm": 1.6171875,
"learning_rate": 7.995092897660363e-05,
"loss": 2.2828,
"num_input_tokens_seen": 648806400,
"step": 550
},
{
"epoch": 0.06499941016869175,
"grad_norm": 1.015625,
"learning_rate": 7.995015332334433e-05,
"loss": 2.244,
"num_input_tokens_seen": 649986048,
"step": 551
},
{
"epoch": 0.06511737643034092,
"grad_norm": 1.5859375,
"learning_rate": 7.994937159160578e-05,
"loss": 2.2447,
"num_input_tokens_seen": 651165696,
"step": 552
},
{
"epoch": 0.06523534269199009,
"grad_norm": 1.0234375,
"learning_rate": 7.99485837815069e-05,
"loss": 2.2295,
"num_input_tokens_seen": 652345344,
"step": 553
},
{
"epoch": 0.06535330895363926,
"grad_norm": 1.1875,
"learning_rate": 7.994778989316757e-05,
"loss": 2.2049,
"num_input_tokens_seen": 653524992,
"step": 554
},
{
"epoch": 0.06547127521528842,
"grad_norm": 0.9609375,
"learning_rate": 7.994698992670855e-05,
"loss": 2.257,
"num_input_tokens_seen": 654704640,
"step": 555
},
{
"epoch": 0.0655892414769376,
"grad_norm": 1.0625,
"learning_rate": 7.99461838822516e-05,
"loss": 2.1423,
"num_input_tokens_seen": 655884288,
"step": 556
},
{
"epoch": 0.06570720773858676,
"grad_norm": 1.109375,
"learning_rate": 7.994537175991934e-05,
"loss": 2.2261,
"num_input_tokens_seen": 657063936,
"step": 557
},
{
"epoch": 0.06582517400023594,
"grad_norm": 1.5859375,
"learning_rate": 7.994455355983532e-05,
"loss": 2.2693,
"num_input_tokens_seen": 658243584,
"step": 558
},
{
"epoch": 0.0659431402618851,
"grad_norm": 1.125,
"learning_rate": 7.994372928212406e-05,
"loss": 2.3283,
"num_input_tokens_seen": 659423232,
"step": 559
},
{
"epoch": 0.06606110652353427,
"grad_norm": 0.9609375,
"learning_rate": 7.994289892691094e-05,
"loss": 2.2631,
"num_input_tokens_seen": 660602880,
"step": 560
},
{
"epoch": 0.06617907278518344,
"grad_norm": 0.93359375,
"learning_rate": 7.994206249432234e-05,
"loss": 2.1609,
"num_input_tokens_seen": 661782528,
"step": 561
},
{
"epoch": 0.06629703904683261,
"grad_norm": 1.4375,
"learning_rate": 7.994121998448549e-05,
"loss": 2.1385,
"num_input_tokens_seen": 662962176,
"step": 562
},
{
"epoch": 0.06641500530848178,
"grad_norm": 1.625,
"learning_rate": 7.99403713975286e-05,
"loss": 2.22,
"num_input_tokens_seen": 664141824,
"step": 563
},
{
"epoch": 0.06653297157013094,
"grad_norm": 0.8125,
"learning_rate": 7.993951673358078e-05,
"loss": 2.2546,
"num_input_tokens_seen": 665321472,
"step": 564
},
{
"epoch": 0.06665093783178011,
"grad_norm": 1.2734375,
"learning_rate": 7.993865599277206e-05,
"loss": 2.217,
"num_input_tokens_seen": 666501120,
"step": 565
},
{
"epoch": 0.06676890409342928,
"grad_norm": 1.6484375,
"learning_rate": 7.993778917523343e-05,
"loss": 2.1732,
"num_input_tokens_seen": 667680768,
"step": 566
},
{
"epoch": 0.06688687035507845,
"grad_norm": 0.96484375,
"learning_rate": 7.993691628109673e-05,
"loss": 2.2929,
"num_input_tokens_seen": 668860416,
"step": 567
},
{
"epoch": 0.06700483661672761,
"grad_norm": 1.203125,
"learning_rate": 7.993603731049481e-05,
"loss": 2.3019,
"num_input_tokens_seen": 670040064,
"step": 568
},
{
"epoch": 0.06712280287837678,
"grad_norm": 1.25,
"learning_rate": 7.99351522635614e-05,
"loss": 2.2712,
"num_input_tokens_seen": 671219712,
"step": 569
},
{
"epoch": 0.06724076914002595,
"grad_norm": 1.078125,
"learning_rate": 7.993426114043115e-05,
"loss": 2.1995,
"num_input_tokens_seen": 672399360,
"step": 570
},
{
"epoch": 0.06735873540167513,
"grad_norm": 0.99609375,
"learning_rate": 7.993336394123965e-05,
"loss": 2.189,
"num_input_tokens_seen": 673579008,
"step": 571
},
{
"epoch": 0.06747670166332428,
"grad_norm": 1.0390625,
"learning_rate": 7.993246066612343e-05,
"loss": 2.2997,
"num_input_tokens_seen": 674758656,
"step": 572
},
{
"epoch": 0.06759466792497346,
"grad_norm": 0.95703125,
"learning_rate": 7.993155131521991e-05,
"loss": 2.2692,
"num_input_tokens_seen": 675938304,
"step": 573
},
{
"epoch": 0.06771263418662263,
"grad_norm": 1.125,
"learning_rate": 7.993063588866742e-05,
"loss": 2.243,
"num_input_tokens_seen": 677117952,
"step": 574
},
{
"epoch": 0.0678306004482718,
"grad_norm": 1.09375,
"learning_rate": 7.992971438660529e-05,
"loss": 2.184,
"num_input_tokens_seen": 678297600,
"step": 575
},
{
"epoch": 0.06794856670992096,
"grad_norm": 0.94921875,
"learning_rate": 7.99287868091737e-05,
"loss": 2.1793,
"num_input_tokens_seen": 679477248,
"step": 576
},
{
"epoch": 0.06806653297157013,
"grad_norm": 1.171875,
"learning_rate": 7.99278531565138e-05,
"loss": 2.3607,
"num_input_tokens_seen": 680656896,
"step": 577
},
{
"epoch": 0.0681844992332193,
"grad_norm": 0.96484375,
"learning_rate": 7.992691342876765e-05,
"loss": 2.2439,
"num_input_tokens_seen": 681836544,
"step": 578
},
{
"epoch": 0.06830246549486847,
"grad_norm": 1.09375,
"learning_rate": 7.99259676260782e-05,
"loss": 2.2214,
"num_input_tokens_seen": 683016192,
"step": 579
},
{
"epoch": 0.06842043175651763,
"grad_norm": 1.1015625,
"learning_rate": 7.992501574858937e-05,
"loss": 2.1635,
"num_input_tokens_seen": 684195840,
"step": 580
},
{
"epoch": 0.0685383980181668,
"grad_norm": 1.25,
"learning_rate": 7.9924057796446e-05,
"loss": 2.1955,
"num_input_tokens_seen": 685375488,
"step": 581
},
{
"epoch": 0.06865636427981597,
"grad_norm": 0.8515625,
"learning_rate": 7.992309376979385e-05,
"loss": 2.1876,
"num_input_tokens_seen": 686555136,
"step": 582
},
{
"epoch": 0.06877433054146515,
"grad_norm": 0.9296875,
"learning_rate": 7.992212366877959e-05,
"loss": 2.1243,
"num_input_tokens_seen": 687734784,
"step": 583
},
{
"epoch": 0.0688922968031143,
"grad_norm": 1.1875,
"learning_rate": 7.992114749355079e-05,
"loss": 2.2922,
"num_input_tokens_seen": 688914432,
"step": 584
},
{
"epoch": 0.06901026306476347,
"grad_norm": 1.1796875,
"learning_rate": 7.992016524425603e-05,
"loss": 2.1665,
"num_input_tokens_seen": 690094080,
"step": 585
},
{
"epoch": 0.06912822932641265,
"grad_norm": 0.84375,
"learning_rate": 7.991917692104473e-05,
"loss": 2.207,
"num_input_tokens_seen": 691273728,
"step": 586
},
{
"epoch": 0.06924619558806182,
"grad_norm": 0.8671875,
"learning_rate": 7.991818252406726e-05,
"loss": 2.1341,
"num_input_tokens_seen": 692453376,
"step": 587
},
{
"epoch": 0.06936416184971098,
"grad_norm": 1.109375,
"learning_rate": 7.991718205347494e-05,
"loss": 2.1968,
"num_input_tokens_seen": 693633024,
"step": 588
},
{
"epoch": 0.06948212811136015,
"grad_norm": 1.078125,
"learning_rate": 7.991617550941998e-05,
"loss": 2.1095,
"num_input_tokens_seen": 694812672,
"step": 589
},
{
"epoch": 0.06960009437300932,
"grad_norm": 0.89453125,
"learning_rate": 7.991516289205554e-05,
"loss": 2.2366,
"num_input_tokens_seen": 695992320,
"step": 590
},
{
"epoch": 0.06971806063465849,
"grad_norm": 1.0703125,
"learning_rate": 7.991414420153569e-05,
"loss": 2.2333,
"num_input_tokens_seen": 697171968,
"step": 591
},
{
"epoch": 0.06983602689630766,
"grad_norm": 0.8125,
"learning_rate": 7.991311943801539e-05,
"loss": 2.1065,
"num_input_tokens_seen": 698351616,
"step": 592
},
{
"epoch": 0.06995399315795682,
"grad_norm": 1.0703125,
"learning_rate": 7.99120886016506e-05,
"loss": 2.1005,
"num_input_tokens_seen": 699531264,
"step": 593
},
{
"epoch": 0.07007195941960599,
"grad_norm": 1.046875,
"learning_rate": 7.991105169259815e-05,
"loss": 2.2168,
"num_input_tokens_seen": 700710912,
"step": 594
},
{
"epoch": 0.07018992568125516,
"grad_norm": 1.0078125,
"learning_rate": 7.991000871101581e-05,
"loss": 2.3617,
"num_input_tokens_seen": 701890560,
"step": 595
},
{
"epoch": 0.07030789194290434,
"grad_norm": 1.703125,
"learning_rate": 7.990895965706227e-05,
"loss": 2.2199,
"num_input_tokens_seen": 703070208,
"step": 596
},
{
"epoch": 0.0704258582045535,
"grad_norm": 1.0078125,
"learning_rate": 7.990790453089714e-05,
"loss": 2.1908,
"num_input_tokens_seen": 704249856,
"step": 597
},
{
"epoch": 0.07054382446620266,
"grad_norm": 0.87109375,
"learning_rate": 7.990684333268097e-05,
"loss": 2.1718,
"num_input_tokens_seen": 705429504,
"step": 598
},
{
"epoch": 0.07066179072785184,
"grad_norm": 0.80078125,
"learning_rate": 7.99057760625752e-05,
"loss": 2.1296,
"num_input_tokens_seen": 706609152,
"step": 599
},
{
"epoch": 0.07077975698950101,
"grad_norm": 0.84375,
"learning_rate": 7.990470272074225e-05,
"loss": 2.1698,
"num_input_tokens_seen": 707788800,
"step": 600
},
{
"epoch": 0.07077975698950101,
"eval_wikipedia_loss": 2.343088150024414,
"eval_wikipedia_runtime": 167.6638,
"eval_wikipedia_samples_per_second": 4.187,
"eval_wikipedia_steps_per_second": 0.179,
"num_input_tokens_seen": 707788800,
"step": 600
},
{
"epoch": 0.07077975698950101,
"eval_toxicity_loss": 4.090839862823486,
"eval_toxicity_runtime": 1.2034,
"eval_toxicity_samples_per_second": 1.662,
"eval_toxicity_steps_per_second": 0.831,
"num_input_tokens_seen": 707788800,
"step": 600
},
{
"epoch": 0.00011796626164916834,
"grad_norm": 1.2421875,
"learning_rate": 7.99036233073454e-05,
"loss": 2.1019,
"num_input_tokens_seen": 708968448,
"step": 601
},
{
"epoch": 0.0002359325232983367,
"grad_norm": 1.328125,
"learning_rate": 7.99025378225489e-05,
"loss": 2.1446,
"num_input_tokens_seen": 710148096,
"step": 602
},
{
"epoch": 0.000353898784947505,
"grad_norm": 1.203125,
"learning_rate": 7.990144626651791e-05,
"loss": 2.1271,
"num_input_tokens_seen": 711327744,
"step": 603
},
{
"epoch": 0.0004718650465966734,
"grad_norm": 1.3359375,
"learning_rate": 7.990034863941851e-05,
"loss": 2.1123,
"num_input_tokens_seen": 712507392,
"step": 604
},
{
"epoch": 0.0005898313082458417,
"grad_norm": 0.8203125,
"learning_rate": 7.989924494141771e-05,
"loss": 2.1567,
"num_input_tokens_seen": 713687040,
"step": 605
},
{
"epoch": 0.00070779756989501,
"grad_norm": 0.91796875,
"learning_rate": 7.989813517268343e-05,
"loss": 2.1786,
"num_input_tokens_seen": 714866688,
"step": 606
},
{
"epoch": 0.0008257638315441783,
"grad_norm": 1.078125,
"learning_rate": 7.989701933338453e-05,
"loss": 2.2267,
"num_input_tokens_seen": 716046336,
"step": 607
},
{
"epoch": 0.0009437300931933467,
"grad_norm": 1.578125,
"learning_rate": 7.989589742369077e-05,
"loss": 2.1507,
"num_input_tokens_seen": 717225984,
"step": 608
},
{
"epoch": 0.001061696354842515,
"grad_norm": 1.0546875,
"learning_rate": 7.989476944377286e-05,
"loss": 2.1274,
"num_input_tokens_seen": 718405632,
"step": 609
},
{
"epoch": 0.0011796626164916834,
"grad_norm": 0.984375,
"learning_rate": 7.989363539380245e-05,
"loss": 2.3054,
"num_input_tokens_seen": 719585280,
"step": 610
},
{
"epoch": 0.0012976288781408518,
"grad_norm": 1.0,
"learning_rate": 7.989249527395205e-05,
"loss": 2.1765,
"num_input_tokens_seen": 720764928,
"step": 611
},
{
"epoch": 0.00141559513979002,
"grad_norm": 1.078125,
"learning_rate": 7.989134908439515e-05,
"loss": 2.1669,
"num_input_tokens_seen": 721944576,
"step": 612
},
{
"epoch": 0.0015335614014391884,
"grad_norm": 0.97265625,
"learning_rate": 7.989019682530614e-05,
"loss": 2.1091,
"num_input_tokens_seen": 723124224,
"step": 613
},
{
"epoch": 0.0016515276630883566,
"grad_norm": 1.265625,
"learning_rate": 7.988903849686033e-05,
"loss": 2.0888,
"num_input_tokens_seen": 724303872,
"step": 614
},
{
"epoch": 0.001769493924737525,
"grad_norm": 0.94921875,
"learning_rate": 7.988787409923398e-05,
"loss": 2.1604,
"num_input_tokens_seen": 725483520,
"step": 615
},
{
"epoch": 0.0018874601863866935,
"grad_norm": 0.96875,
"learning_rate": 7.988670363260425e-05,
"loss": 2.1344,
"num_input_tokens_seen": 726663168,
"step": 616
},
{
"epoch": 0.0020054264480358617,
"grad_norm": 0.85546875,
"learning_rate": 7.988552709714921e-05,
"loss": 2.2063,
"num_input_tokens_seen": 727842816,
"step": 617
},
{
"epoch": 0.00212339270968503,
"grad_norm": 1.21875,
"learning_rate": 7.98843444930479e-05,
"loss": 2.2359,
"num_input_tokens_seen": 729022464,
"step": 618
},
{
"epoch": 0.0022413589713341986,
"grad_norm": 1.4609375,
"learning_rate": 7.988315582048022e-05,
"loss": 2.1383,
"num_input_tokens_seen": 730202112,
"step": 619
},
{
"epoch": 0.0023593252329833668,
"grad_norm": 0.99609375,
"learning_rate": 7.988196107962707e-05,
"loss": 2.2032,
"num_input_tokens_seen": 731381760,
"step": 620
},
{
"epoch": 0.002477291494632535,
"grad_norm": 0.828125,
"learning_rate": 7.98807602706702e-05,
"loss": 2.2634,
"num_input_tokens_seen": 732561408,
"step": 621
},
{
"epoch": 0.0025952577562817036,
"grad_norm": 0.90234375,
"learning_rate": 7.987955339379234e-05,
"loss": 2.177,
"num_input_tokens_seen": 733741056,
"step": 622
},
{
"epoch": 0.002713224017930872,
"grad_norm": 1.4375,
"learning_rate": 7.987834044917709e-05,
"loss": 2.1918,
"num_input_tokens_seen": 734920704,
"step": 623
},
{
"epoch": 0.00283119027958004,
"grad_norm": 0.98828125,
"learning_rate": 7.9877121437009e-05,
"loss": 2.2417,
"num_input_tokens_seen": 736100352,
"step": 624
},
{
"epoch": 0.0029491565412292082,
"grad_norm": 0.8203125,
"learning_rate": 7.987589635747359e-05,
"loss": 2.1719,
"num_input_tokens_seen": 737280000,
"step": 625
},
{
"epoch": 0.003067122802878377,
"grad_norm": 1.0,
"learning_rate": 7.987466521075722e-05,
"loss": 2.2465,
"num_input_tokens_seen": 738459648,
"step": 626
},
{
"epoch": 0.003185089064527545,
"grad_norm": 1.0,
"learning_rate": 7.987342799704721e-05,
"loss": 2.1183,
"num_input_tokens_seen": 739639296,
"step": 627
},
{
"epoch": 0.0033030553261767133,
"grad_norm": 1.1796875,
"learning_rate": 7.987218471653181e-05,
"loss": 2.0837,
"num_input_tokens_seen": 740818944,
"step": 628
},
{
"epoch": 0.003421021587825882,
"grad_norm": 1.2734375,
"learning_rate": 7.987093536940019e-05,
"loss": 2.1209,
"num_input_tokens_seen": 741998592,
"step": 629
},
{
"epoch": 0.00353898784947505,
"grad_norm": 0.71875,
"learning_rate": 7.986967995584245e-05,
"loss": 2.137,
"num_input_tokens_seen": 743178240,
"step": 630
},
{
"epoch": 0.0036569541111242184,
"grad_norm": 0.91015625,
"learning_rate": 7.986841847604958e-05,
"loss": 2.1377,
"num_input_tokens_seen": 744357888,
"step": 631
},
{
"epoch": 0.003774920372773387,
"grad_norm": 1.5625,
"learning_rate": 7.986715093021353e-05,
"loss": 2.2592,
"num_input_tokens_seen": 745537536,
"step": 632
},
{
"epoch": 0.003892886634422555,
"grad_norm": 0.99609375,
"learning_rate": 7.986587731852717e-05,
"loss": 2.1295,
"num_input_tokens_seen": 746717184,
"step": 633
},
{
"epoch": 0.004010852896071723,
"grad_norm": 0.85546875,
"learning_rate": 7.986459764118427e-05,
"loss": 2.1378,
"num_input_tokens_seen": 747896832,
"step": 634
},
{
"epoch": 0.004128819157720892,
"grad_norm": 1.53125,
"learning_rate": 7.986331189837952e-05,
"loss": 2.1956,
"num_input_tokens_seen": 749076480,
"step": 635
},
{
"epoch": 0.00424678541937006,
"grad_norm": 0.9296875,
"learning_rate": 7.986202009030858e-05,
"loss": 2.2073,
"num_input_tokens_seen": 750256128,
"step": 636
},
{
"epoch": 0.004364751681019229,
"grad_norm": 1.0703125,
"learning_rate": 7.986072221716798e-05,
"loss": 2.1139,
"num_input_tokens_seen": 751435776,
"step": 637
},
{
"epoch": 0.004482717942668397,
"grad_norm": 1.0390625,
"learning_rate": 7.985941827915519e-05,
"loss": 2.2407,
"num_input_tokens_seen": 752615424,
"step": 638
},
{
"epoch": 0.004600684204317565,
"grad_norm": 1.2578125,
"learning_rate": 7.985810827646862e-05,
"loss": 2.1899,
"num_input_tokens_seen": 753795072,
"step": 639
},
{
"epoch": 0.0047186504659667335,
"grad_norm": 1.0625,
"learning_rate": 7.985679220930758e-05,
"loss": 2.2486,
"num_input_tokens_seen": 754974720,
"step": 640
},
{
"epoch": 0.004836616727615902,
"grad_norm": 0.875,
"learning_rate": 7.985547007787231e-05,
"loss": 2.1191,
"num_input_tokens_seen": 756154368,
"step": 641
},
{
"epoch": 0.00495458298926507,
"grad_norm": 0.8515625,
"learning_rate": 7.985414188236398e-05,
"loss": 2.0786,
"num_input_tokens_seen": 757334016,
"step": 642
},
{
"epoch": 0.005072549250914238,
"grad_norm": 1.0078125,
"learning_rate": 7.985280762298468e-05,
"loss": 2.2075,
"num_input_tokens_seen": 758513664,
"step": 643
},
{
"epoch": 0.005190515512563407,
"grad_norm": 1.0546875,
"learning_rate": 7.985146729993741e-05,
"loss": 2.1767,
"num_input_tokens_seen": 759693312,
"step": 644
},
{
"epoch": 0.0053084817742125754,
"grad_norm": 0.9921875,
"learning_rate": 7.985012091342611e-05,
"loss": 2.1301,
"num_input_tokens_seen": 760872960,
"step": 645
},
{
"epoch": 0.005426448035861744,
"grad_norm": 0.96484375,
"learning_rate": 7.984876846365564e-05,
"loss": 2.1197,
"num_input_tokens_seen": 762052608,
"step": 646
},
{
"epoch": 0.005544414297510912,
"grad_norm": 1.25,
"learning_rate": 7.984740995083175e-05,
"loss": 2.4474,
"num_input_tokens_seen": 763232256,
"step": 647
},
{
"epoch": 0.00566238055916008,
"grad_norm": 1.1796875,
"learning_rate": 7.984604537516118e-05,
"loss": 2.1303,
"num_input_tokens_seen": 764411904,
"step": 648
},
{
"epoch": 0.005780346820809248,
"grad_norm": 1.09375,
"learning_rate": 7.984467473685153e-05,
"loss": 2.2679,
"num_input_tokens_seen": 765591552,
"step": 649
},
{
"epoch": 0.0058983130824584165,
"grad_norm": 0.87109375,
"learning_rate": 7.984329803611133e-05,
"loss": 2.1687,
"num_input_tokens_seen": 766771200,
"step": 650
},
{
"epoch": 0.0060162793441075856,
"grad_norm": 1.1015625,
"learning_rate": 7.984191527315009e-05,
"loss": 2.1592,
"num_input_tokens_seen": 767950848,
"step": 651
},
{
"epoch": 0.006134245605756754,
"grad_norm": 0.9140625,
"learning_rate": 7.984052644817815e-05,
"loss": 2.1808,
"num_input_tokens_seen": 769130496,
"step": 652
},
{
"epoch": 0.006252211867405922,
"grad_norm": 0.80078125,
"learning_rate": 7.983913156140685e-05,
"loss": 2.209,
"num_input_tokens_seen": 770310144,
"step": 653
},
{
"epoch": 0.00637017812905509,
"grad_norm": 1.1640625,
"learning_rate": 7.983773061304843e-05,
"loss": 2.1247,
"num_input_tokens_seen": 771489792,
"step": 654
},
{
"epoch": 0.006488144390704258,
"grad_norm": 1.0078125,
"learning_rate": 7.983632360331603e-05,
"loss": 2.1624,
"num_input_tokens_seen": 772669440,
"step": 655
},
{
"epoch": 0.006606110652353427,
"grad_norm": 1.28125,
"learning_rate": 7.983491053242373e-05,
"loss": 2.1374,
"num_input_tokens_seen": 773849088,
"step": 656
},
{
"epoch": 0.006724076914002596,
"grad_norm": 1.078125,
"learning_rate": 7.983349140058654e-05,
"loss": 2.1465,
"num_input_tokens_seen": 775028736,
"step": 657
},
{
"epoch": 0.006842043175651764,
"grad_norm": 1.109375,
"learning_rate": 7.983206620802038e-05,
"loss": 2.1039,
"num_input_tokens_seen": 776208384,
"step": 658
},
{
"epoch": 0.006960009437300932,
"grad_norm": 0.90234375,
"learning_rate": 7.98306349549421e-05,
"loss": 2.2616,
"num_input_tokens_seen": 777388032,
"step": 659
},
{
"epoch": 0.0070779756989501,
"grad_norm": 1.1484375,
"learning_rate": 7.982919764156945e-05,
"loss": 2.1215,
"num_input_tokens_seen": 778567680,
"step": 660
},
{
"epoch": 0.0071959419605992685,
"grad_norm": 1.0078125,
"learning_rate": 7.982775426812114e-05,
"loss": 2.1266,
"num_input_tokens_seen": 779747328,
"step": 661
},
{
"epoch": 0.007313908222248437,
"grad_norm": 1.2890625,
"learning_rate": 7.982630483481678e-05,
"loss": 2.1429,
"num_input_tokens_seen": 780926976,
"step": 662
},
{
"epoch": 0.007431874483897605,
"grad_norm": 0.90234375,
"learning_rate": 7.982484934187687e-05,
"loss": 2.0725,
"num_input_tokens_seen": 782106624,
"step": 663
},
{
"epoch": 0.007549840745546774,
"grad_norm": 1.015625,
"learning_rate": 7.982338778952292e-05,
"loss": 2.2623,
"num_input_tokens_seen": 783286272,
"step": 664
},
{
"epoch": 0.007667807007195942,
"grad_norm": 1.515625,
"learning_rate": 7.982192017797727e-05,
"loss": 2.1474,
"num_input_tokens_seen": 784465920,
"step": 665
},
{
"epoch": 0.00778577326884511,
"grad_norm": 1.0390625,
"learning_rate": 7.982044650746321e-05,
"loss": 2.213,
"num_input_tokens_seen": 785645568,
"step": 666
},
{
"epoch": 0.007903739530494279,
"grad_norm": 0.94921875,
"learning_rate": 7.9818966778205e-05,
"loss": 2.1183,
"num_input_tokens_seen": 786825216,
"step": 667
},
{
"epoch": 0.008021705792143447,
"grad_norm": 1.453125,
"learning_rate": 7.981748099042777e-05,
"loss": 2.203,
"num_input_tokens_seen": 788004864,
"step": 668
},
{
"epoch": 0.008139672053792615,
"grad_norm": 1.1640625,
"learning_rate": 7.981598914435756e-05,
"loss": 2.247,
"num_input_tokens_seen": 789184512,
"step": 669
},
{
"epoch": 0.008257638315441783,
"grad_norm": 0.78515625,
"learning_rate": 7.98144912402214e-05,
"loss": 2.1766,
"num_input_tokens_seen": 790364160,
"step": 670
},
{
"epoch": 0.008375604577090951,
"grad_norm": 1.46875,
"learning_rate": 7.981298727824715e-05,
"loss": 2.1032,
"num_input_tokens_seen": 791543808,
"step": 671
},
{
"epoch": 0.00849357083874012,
"grad_norm": 1.0078125,
"learning_rate": 7.981147725866367e-05,
"loss": 2.0328,
"num_input_tokens_seen": 792723456,
"step": 672
},
{
"epoch": 0.008611537100389288,
"grad_norm": 0.98828125,
"learning_rate": 7.980996118170071e-05,
"loss": 2.156,
"num_input_tokens_seen": 793903104,
"step": 673
},
{
"epoch": 0.008729503362038458,
"grad_norm": 1.515625,
"learning_rate": 7.980843904758894e-05,
"loss": 2.2632,
"num_input_tokens_seen": 795082752,
"step": 674
},
{
"epoch": 0.008847469623687626,
"grad_norm": 0.98828125,
"learning_rate": 7.980691085655995e-05,
"loss": 2.1247,
"num_input_tokens_seen": 796262400,
"step": 675
},
{
"epoch": 0.008965435885336794,
"grad_norm": 0.85546875,
"learning_rate": 7.980537660884625e-05,
"loss": 2.1808,
"num_input_tokens_seen": 797442048,
"step": 676
},
{
"epoch": 0.009083402146985962,
"grad_norm": 1.375,
"learning_rate": 7.980383630468132e-05,
"loss": 2.1755,
"num_input_tokens_seen": 798621696,
"step": 677
},
{
"epoch": 0.00920136840863513,
"grad_norm": 1.0546875,
"learning_rate": 7.980228994429947e-05,
"loss": 2.1093,
"num_input_tokens_seen": 799801344,
"step": 678
},
{
"epoch": 0.009319334670284299,
"grad_norm": 0.78515625,
"learning_rate": 7.9800737527936e-05,
"loss": 2.0614,
"num_input_tokens_seen": 800980992,
"step": 679
},
{
"epoch": 0.009437300931933467,
"grad_norm": 0.78125,
"learning_rate": 7.979917905582712e-05,
"loss": 2.0862,
"num_input_tokens_seen": 802160640,
"step": 680
},
{
"epoch": 0.009555267193582635,
"grad_norm": 1.140625,
"learning_rate": 7.979761452820993e-05,
"loss": 2.1008,
"num_input_tokens_seen": 803340288,
"step": 681
},
{
"epoch": 0.009673233455231803,
"grad_norm": 1.0859375,
"learning_rate": 7.979604394532251e-05,
"loss": 2.1894,
"num_input_tokens_seen": 804519936,
"step": 682
},
{
"epoch": 0.009791199716880972,
"grad_norm": 0.921875,
"learning_rate": 7.979446730740381e-05,
"loss": 2.1344,
"num_input_tokens_seen": 805699584,
"step": 683
},
{
"epoch": 0.00990916597853014,
"grad_norm": 1.1171875,
"learning_rate": 7.979288461469371e-05,
"loss": 2.1419,
"num_input_tokens_seen": 806879232,
"step": 684
},
{
"epoch": 0.010027132240179308,
"grad_norm": 0.8046875,
"learning_rate": 7.9791295867433e-05,
"loss": 2.1314,
"num_input_tokens_seen": 808058880,
"step": 685
},
{
"epoch": 0.010145098501828476,
"grad_norm": 0.89453125,
"learning_rate": 7.978970106586347e-05,
"loss": 2.1138,
"num_input_tokens_seen": 809238528,
"step": 686
},
{
"epoch": 0.010263064763477646,
"grad_norm": 0.78125,
"learning_rate": 7.978810021022773e-05,
"loss": 2.1573,
"num_input_tokens_seen": 810418176,
"step": 687
},
{
"epoch": 0.010381031025126814,
"grad_norm": 0.96875,
"learning_rate": 7.978649330076936e-05,
"loss": 2.239,
"num_input_tokens_seen": 811597824,
"step": 688
},
{
"epoch": 0.010498997286775983,
"grad_norm": 1.109375,
"learning_rate": 7.978488033773285e-05,
"loss": 2.0574,
"num_input_tokens_seen": 812777472,
"step": 689
},
{
"epoch": 0.010616963548425151,
"grad_norm": 1.046875,
"learning_rate": 7.978326132136364e-05,
"loss": 2.0513,
"num_input_tokens_seen": 813957120,
"step": 690
},
{
"epoch": 0.010734929810074319,
"grad_norm": 1.0390625,
"learning_rate": 7.978163625190803e-05,
"loss": 2.1201,
"num_input_tokens_seen": 815136768,
"step": 691
},
{
"epoch": 0.010852896071723487,
"grad_norm": 0.93359375,
"learning_rate": 7.978000512961329e-05,
"loss": 2.0681,
"num_input_tokens_seen": 816316416,
"step": 692
},
{
"epoch": 0.010970862333372655,
"grad_norm": 1.1015625,
"learning_rate": 7.977836795472761e-05,
"loss": 2.0907,
"num_input_tokens_seen": 817496064,
"step": 693
},
{
"epoch": 0.011088828595021824,
"grad_norm": 0.94140625,
"learning_rate": 7.977672472750006e-05,
"loss": 2.1739,
"num_input_tokens_seen": 818675712,
"step": 694
},
{
"epoch": 0.011206794856670992,
"grad_norm": 0.84765625,
"learning_rate": 7.977507544818069e-05,
"loss": 2.1536,
"num_input_tokens_seen": 819855360,
"step": 695
},
{
"epoch": 0.01132476111832016,
"grad_norm": 1.4453125,
"learning_rate": 7.977342011702043e-05,
"loss": 2.0564,
"num_input_tokens_seen": 821035008,
"step": 696
},
{
"epoch": 0.011442727379969328,
"grad_norm": 0.90234375,
"learning_rate": 7.977175873427114e-05,
"loss": 2.0971,
"num_input_tokens_seen": 822214656,
"step": 697
},
{
"epoch": 0.011560693641618497,
"grad_norm": 0.81640625,
"learning_rate": 7.977009130018561e-05,
"loss": 2.1439,
"num_input_tokens_seen": 823394304,
"step": 698
},
{
"epoch": 0.011678659903267665,
"grad_norm": 0.91796875,
"learning_rate": 7.976841781501751e-05,
"loss": 2.0159,
"num_input_tokens_seen": 824573952,
"step": 699
},
{
"epoch": 0.011796626164916833,
"grad_norm": 0.8515625,
"learning_rate": 7.97667382790215e-05,
"loss": 2.2684,
"num_input_tokens_seen": 825753600,
"step": 700
},
{
"epoch": 0.011914592426566003,
"grad_norm": 0.79296875,
"learning_rate": 7.976505269245314e-05,
"loss": 2.0875,
"num_input_tokens_seen": 826933248,
"step": 701
},
{
"epoch": 0.012032558688215171,
"grad_norm": 0.76171875,
"learning_rate": 7.976336105556884e-05,
"loss": 2.0368,
"num_input_tokens_seen": 828112896,
"step": 702
},
{
"epoch": 0.01215052494986434,
"grad_norm": 0.83203125,
"learning_rate": 7.976166336862602e-05,
"loss": 2.0681,
"num_input_tokens_seen": 829292544,
"step": 703
},
{
"epoch": 0.012268491211513508,
"grad_norm": 1.1953125,
"learning_rate": 7.975995963188297e-05,
"loss": 2.1092,
"num_input_tokens_seen": 830472192,
"step": 704
},
{
"epoch": 0.012386457473162676,
"grad_norm": 1.8046875,
"learning_rate": 7.975824984559893e-05,
"loss": 2.1025,
"num_input_tokens_seen": 831651840,
"step": 705
},
{
"epoch": 0.012504423734811844,
"grad_norm": 0.77734375,
"learning_rate": 7.975653401003404e-05,
"loss": 2.1325,
"num_input_tokens_seen": 832831488,
"step": 706
},
{
"epoch": 0.012622389996461012,
"grad_norm": 1.4921875,
"learning_rate": 7.975481212544938e-05,
"loss": 2.1961,
"num_input_tokens_seen": 834011136,
"step": 707
},
{
"epoch": 0.01274035625811018,
"grad_norm": 1.4296875,
"learning_rate": 7.97530841921069e-05,
"loss": 2.0702,
"num_input_tokens_seen": 835190784,
"step": 708
},
{
"epoch": 0.012858322519759349,
"grad_norm": 0.90234375,
"learning_rate": 7.975135021026956e-05,
"loss": 2.165,
"num_input_tokens_seen": 836370432,
"step": 709
},
{
"epoch": 0.012976288781408517,
"grad_norm": 1.984375,
"learning_rate": 7.974961018020115e-05,
"loss": 2.1599,
"num_input_tokens_seen": 837550080,
"step": 710
},
{
"epoch": 0.013094255043057685,
"grad_norm": 0.99609375,
"learning_rate": 7.974786410216643e-05,
"loss": 2.0392,
"num_input_tokens_seen": 838729728,
"step": 711
},
{
"epoch": 0.013212221304706853,
"grad_norm": 2.28125,
"learning_rate": 7.974611197643108e-05,
"loss": 2.2548,
"num_input_tokens_seen": 839909376,
"step": 712
},
{
"epoch": 0.013330187566356021,
"grad_norm": 1.2890625,
"learning_rate": 7.974435380326166e-05,
"loss": 2.0817,
"num_input_tokens_seen": 841089024,
"step": 713
},
{
"epoch": 0.013448153828005191,
"grad_norm": 2.46875,
"learning_rate": 7.97425895829257e-05,
"loss": 2.1358,
"num_input_tokens_seen": 842268672,
"step": 714
},
{
"epoch": 0.01356612008965436,
"grad_norm": 1.8828125,
"learning_rate": 7.974081931569163e-05,
"loss": 2.207,
"num_input_tokens_seen": 843448320,
"step": 715
},
{
"epoch": 0.013684086351303528,
"grad_norm": 2.296875,
"learning_rate": 7.97390430018288e-05,
"loss": 2.106,
"num_input_tokens_seen": 844627968,
"step": 716
},
{
"epoch": 0.013802052612952696,
"grad_norm": 1.8125,
"learning_rate": 7.973726064160746e-05,
"loss": 2.0721,
"num_input_tokens_seen": 845807616,
"step": 717
},
{
"epoch": 0.013920018874601864,
"grad_norm": 1.984375,
"learning_rate": 7.973547223529882e-05,
"loss": 2.1795,
"num_input_tokens_seen": 846987264,
"step": 718
},
{
"epoch": 0.014037985136251032,
"grad_norm": 1.703125,
"learning_rate": 7.973367778317497e-05,
"loss": 2.1144,
"num_input_tokens_seen": 848166912,
"step": 719
},
{
"epoch": 0.0141559513979002,
"grad_norm": 2.046875,
"learning_rate": 7.973187728550897e-05,
"loss": 2.0617,
"num_input_tokens_seen": 849346560,
"step": 720
},
{
"epoch": 0.014273917659549369,
"grad_norm": 1.6875,
"learning_rate": 7.973007074257472e-05,
"loss": 1.9967,
"num_input_tokens_seen": 850526208,
"step": 721
},
{
"epoch": 0.014391883921198537,
"grad_norm": 1.6328125,
"learning_rate": 7.972825815464713e-05,
"loss": 2.0598,
"num_input_tokens_seen": 851705856,
"step": 722
},
{
"epoch": 0.014509850182847705,
"grad_norm": 1.6875,
"learning_rate": 7.972643952200198e-05,
"loss": 2.0309,
"num_input_tokens_seen": 852885504,
"step": 723
},
{
"epoch": 0.014627816444496873,
"grad_norm": 1.3828125,
"learning_rate": 7.972461484491597e-05,
"loss": 2.0394,
"num_input_tokens_seen": 854065152,
"step": 724
},
{
"epoch": 0.014745782706146042,
"grad_norm": 1.4765625,
"learning_rate": 7.972278412366672e-05,
"loss": 2.0426,
"num_input_tokens_seen": 855244800,
"step": 725
},
{
"epoch": 0.01486374896779521,
"grad_norm": 1.421875,
"learning_rate": 7.97209473585328e-05,
"loss": 2.0396,
"num_input_tokens_seen": 856424448,
"step": 726
},
{
"epoch": 0.014981715229444378,
"grad_norm": 1.09375,
"learning_rate": 7.971910454979367e-05,
"loss": 2.0892,
"num_input_tokens_seen": 857604096,
"step": 727
},
{
"epoch": 0.015099681491093548,
"grad_norm": 1.8515625,
"learning_rate": 7.971725569772968e-05,
"loss": 2.1033,
"num_input_tokens_seen": 858783744,
"step": 728
},
{
"epoch": 0.015217647752742716,
"grad_norm": 1.5234375,
"learning_rate": 7.97154008026222e-05,
"loss": 1.975,
"num_input_tokens_seen": 859963392,
"step": 729
},
{
"epoch": 0.015335614014391884,
"grad_norm": 1.703125,
"learning_rate": 7.97135398647534e-05,
"loss": 2.0152,
"num_input_tokens_seen": 861143040,
"step": 730
},
{
"epoch": 0.015453580276041053,
"grad_norm": 1.4453125,
"learning_rate": 7.971167288440646e-05,
"loss": 2.0447,
"num_input_tokens_seen": 862322688,
"step": 731
},
{
"epoch": 0.01557154653769022,
"grad_norm": 1.4921875,
"learning_rate": 7.970979986186541e-05,
"loss": 2.0918,
"num_input_tokens_seen": 863502336,
"step": 732
},
{
"epoch": 0.01568951279933939,
"grad_norm": 1.2734375,
"learning_rate": 7.970792079741527e-05,
"loss": 2.0034,
"num_input_tokens_seen": 864681984,
"step": 733
},
{
"epoch": 0.015807479060988557,
"grad_norm": 1.28125,
"learning_rate": 7.970603569134192e-05,
"loss": 2.0668,
"num_input_tokens_seen": 865861632,
"step": 734
},
{
"epoch": 0.015925445322637725,
"grad_norm": 1.171875,
"learning_rate": 7.970414454393218e-05,
"loss": 1.9964,
"num_input_tokens_seen": 867041280,
"step": 735
},
{
"epoch": 0.016043411584286894,
"grad_norm": 1.703125,
"learning_rate": 7.970224735547382e-05,
"loss": 1.9982,
"num_input_tokens_seen": 868220928,
"step": 736
},
{
"epoch": 0.016161377845936062,
"grad_norm": 1.4921875,
"learning_rate": 7.970034412625547e-05,
"loss": 2.1421,
"num_input_tokens_seen": 869400576,
"step": 737
},
{
"epoch": 0.01627934410758523,
"grad_norm": 1.5078125,
"learning_rate": 7.96984348565667e-05,
"loss": 2.0866,
"num_input_tokens_seen": 870580224,
"step": 738
},
{
"epoch": 0.0163973103692344,
"grad_norm": 1.296875,
"learning_rate": 7.969651954669805e-05,
"loss": 2.1172,
"num_input_tokens_seen": 871759872,
"step": 739
},
{
"epoch": 0.016515276630883566,
"grad_norm": 1.515625,
"learning_rate": 7.96945981969409e-05,
"loss": 2.1618,
"num_input_tokens_seen": 872939520,
"step": 740
},
{
"epoch": 0.016633242892532735,
"grad_norm": 1.1640625,
"learning_rate": 7.96926708075876e-05,
"loss": 2.0966,
"num_input_tokens_seen": 874119168,
"step": 741
},
{
"epoch": 0.016751209154181903,
"grad_norm": 1.609375,
"learning_rate": 7.969073737893142e-05,
"loss": 1.9877,
"num_input_tokens_seen": 875298816,
"step": 742
},
{
"epoch": 0.01686917541583107,
"grad_norm": 1.2421875,
"learning_rate": 7.968879791126652e-05,
"loss": 2.09,
"num_input_tokens_seen": 876478464,
"step": 743
},
{
"epoch": 0.01698714167748024,
"grad_norm": 1.5703125,
"learning_rate": 7.968685240488798e-05,
"loss": 2.0698,
"num_input_tokens_seen": 877658112,
"step": 744
},
{
"epoch": 0.017105107939129408,
"grad_norm": 1.296875,
"learning_rate": 7.968490086009184e-05,
"loss": 2.1101,
"num_input_tokens_seen": 878837760,
"step": 745
},
{
"epoch": 0.017223074200778576,
"grad_norm": 1.1640625,
"learning_rate": 7.9682943277175e-05,
"loss": 1.96,
"num_input_tokens_seen": 880017408,
"step": 746
},
{
"epoch": 0.017341040462427744,
"grad_norm": 1.0546875,
"learning_rate": 7.968097965643533e-05,
"loss": 1.9999,
"num_input_tokens_seen": 881197056,
"step": 747
},
{
"epoch": 0.017459006724076916,
"grad_norm": 1.140625,
"learning_rate": 7.96790099981716e-05,
"loss": 2.0833,
"num_input_tokens_seen": 882376704,
"step": 748
},
{
"epoch": 0.017576972985726084,
"grad_norm": 0.87890625,
"learning_rate": 7.967703430268349e-05,
"loss": 1.9634,
"num_input_tokens_seen": 883556352,
"step": 749
},
{
"epoch": 0.017694939247375252,
"grad_norm": 1.359375,
"learning_rate": 7.967505257027158e-05,
"loss": 2.1086,
"num_input_tokens_seen": 884736000,
"step": 750
},
{
"epoch": 0.01781290550902442,
"grad_norm": 0.875,
"learning_rate": 7.967306480123745e-05,
"loss": 2.0866,
"num_input_tokens_seen": 885915648,
"step": 751
},
{
"epoch": 0.01793087177067359,
"grad_norm": 1.2734375,
"learning_rate": 7.967107099588349e-05,
"loss": 2.1024,
"num_input_tokens_seen": 887095296,
"step": 752
},
{
"epoch": 0.018048838032322757,
"grad_norm": 0.91015625,
"learning_rate": 7.966907115451311e-05,
"loss": 2.1854,
"num_input_tokens_seen": 888274944,
"step": 753
},
{
"epoch": 0.018166804293971925,
"grad_norm": 1.5625,
"learning_rate": 7.966706527743052e-05,
"loss": 2.0433,
"num_input_tokens_seen": 889454592,
"step": 754
},
{
"epoch": 0.018284770555621093,
"grad_norm": 1.1640625,
"learning_rate": 7.966505336494098e-05,
"loss": 2.0656,
"num_input_tokens_seen": 890634240,
"step": 755
},
{
"epoch": 0.01840273681727026,
"grad_norm": 1.546875,
"learning_rate": 7.96630354173506e-05,
"loss": 1.9556,
"num_input_tokens_seen": 891813888,
"step": 756
},
{
"epoch": 0.01852070307891943,
"grad_norm": 1.421875,
"learning_rate": 7.966101143496637e-05,
"loss": 2.0212,
"num_input_tokens_seen": 892993536,
"step": 757
},
{
"epoch": 0.018638669340568598,
"grad_norm": 1.4765625,
"learning_rate": 7.965898141809629e-05,
"loss": 2.1419,
"num_input_tokens_seen": 894173184,
"step": 758
},
{
"epoch": 0.018756635602217766,
"grad_norm": 1.2265625,
"learning_rate": 7.96569453670492e-05,
"loss": 2.0694,
"num_input_tokens_seen": 895352832,
"step": 759
},
{
"epoch": 0.018874601863866934,
"grad_norm": 1.2890625,
"learning_rate": 7.96549032821349e-05,
"loss": 2.0899,
"num_input_tokens_seen": 896532480,
"step": 760
},
{
"epoch": 0.018992568125516102,
"grad_norm": 0.9375,
"learning_rate": 7.96528551636641e-05,
"loss": 2.0442,
"num_input_tokens_seen": 897712128,
"step": 761
},
{
"epoch": 0.01911053438716527,
"grad_norm": 1.171875,
"learning_rate": 7.96508010119484e-05,
"loss": 2.0397,
"num_input_tokens_seen": 898891776,
"step": 762
},
{
"epoch": 0.01922850064881444,
"grad_norm": 0.86328125,
"learning_rate": 7.964874082730039e-05,
"loss": 2.0424,
"num_input_tokens_seen": 900071424,
"step": 763
},
{
"epoch": 0.019346466910463607,
"grad_norm": 0.9140625,
"learning_rate": 7.964667461003347e-05,
"loss": 2.1146,
"num_input_tokens_seen": 901251072,
"step": 764
},
{
"epoch": 0.019464433172112775,
"grad_norm": 0.80078125,
"learning_rate": 7.964460236046209e-05,
"loss": 2.0401,
"num_input_tokens_seen": 902430720,
"step": 765
},
{
"epoch": 0.019582399433761943,
"grad_norm": 1.0078125,
"learning_rate": 7.964252407890147e-05,
"loss": 1.9694,
"num_input_tokens_seen": 903610368,
"step": 766
},
{
"epoch": 0.01970036569541111,
"grad_norm": 1.0234375,
"learning_rate": 7.964043976566787e-05,
"loss": 1.9582,
"num_input_tokens_seen": 904790016,
"step": 767
},
{
"epoch": 0.01981833195706028,
"grad_norm": 1.3515625,
"learning_rate": 7.963834942107843e-05,
"loss": 2.0266,
"num_input_tokens_seen": 905969664,
"step": 768
},
{
"epoch": 0.019936298218709448,
"grad_norm": 0.8203125,
"learning_rate": 7.963625304545115e-05,
"loss": 2.1604,
"num_input_tokens_seen": 907149312,
"step": 769
},
{
"epoch": 0.020054264480358616,
"grad_norm": 0.90234375,
"learning_rate": 7.963415063910505e-05,
"loss": 2.1845,
"num_input_tokens_seen": 908328960,
"step": 770
},
{
"epoch": 0.020172230742007784,
"grad_norm": 0.828125,
"learning_rate": 7.963204220236e-05,
"loss": 2.0049,
"num_input_tokens_seen": 909508608,
"step": 771
},
{
"epoch": 0.020290197003656953,
"grad_norm": 1.3359375,
"learning_rate": 7.962992773553678e-05,
"loss": 2.0349,
"num_input_tokens_seen": 910688256,
"step": 772
},
{
"epoch": 0.02040816326530612,
"grad_norm": 0.99609375,
"learning_rate": 7.962780723895712e-05,
"loss": 1.9531,
"num_input_tokens_seen": 911867904,
"step": 773
},
{
"epoch": 0.020526129526955292,
"grad_norm": 0.9609375,
"learning_rate": 7.962568071294368e-05,
"loss": 2.0456,
"num_input_tokens_seen": 913047552,
"step": 774
},
{
"epoch": 0.02064409578860446,
"grad_norm": 1.1796875,
"learning_rate": 7.962354815781999e-05,
"loss": 2.1398,
"num_input_tokens_seen": 914227200,
"step": 775
},
{
"epoch": 0.02076206205025363,
"grad_norm": 1.0546875,
"learning_rate": 7.96214095739105e-05,
"loss": 2.1565,
"num_input_tokens_seen": 915406848,
"step": 776
},
{
"epoch": 0.020880028311902797,
"grad_norm": 1.4296875,
"learning_rate": 7.961926496154066e-05,
"loss": 2.0268,
"num_input_tokens_seen": 916586496,
"step": 777
},
{
"epoch": 0.020997994573551965,
"grad_norm": 1.015625,
"learning_rate": 7.961711432103672e-05,
"loss": 2.0407,
"num_input_tokens_seen": 917766144,
"step": 778
},
{
"epoch": 0.021115960835201134,
"grad_norm": 1.2890625,
"learning_rate": 7.961495765272593e-05,
"loss": 1.9779,
"num_input_tokens_seen": 918945792,
"step": 779
},
{
"epoch": 0.021233927096850302,
"grad_norm": 1.0234375,
"learning_rate": 7.961279495693644e-05,
"loss": 2.0524,
"num_input_tokens_seen": 920125440,
"step": 780
},
{
"epoch": 0.02135189335849947,
"grad_norm": 0.90234375,
"learning_rate": 7.961062623399728e-05,
"loss": 2.012,
"num_input_tokens_seen": 921305088,
"step": 781
},
{
"epoch": 0.021469859620148638,
"grad_norm": 0.8125,
"learning_rate": 7.960845148423844e-05,
"loss": 1.9574,
"num_input_tokens_seen": 922484736,
"step": 782
},
{
"epoch": 0.021587825881797806,
"grad_norm": 0.78515625,
"learning_rate": 7.960627070799081e-05,
"loss": 2.0983,
"num_input_tokens_seen": 923664384,
"step": 783
},
{
"epoch": 0.021705792143446975,
"grad_norm": 1.1640625,
"learning_rate": 7.96040839055862e-05,
"loss": 2.0531,
"num_input_tokens_seen": 924844032,
"step": 784
},
{
"epoch": 0.021823758405096143,
"grad_norm": 1.0703125,
"learning_rate": 7.960189107735734e-05,
"loss": 2.0866,
"num_input_tokens_seen": 926023680,
"step": 785
},
{
"epoch": 0.02194172466674531,
"grad_norm": 0.88671875,
"learning_rate": 7.959969222363786e-05,
"loss": 2.012,
"num_input_tokens_seen": 927203328,
"step": 786
},
{
"epoch": 0.02205969092839448,
"grad_norm": 0.77734375,
"learning_rate": 7.959748734476231e-05,
"loss": 2.0116,
"num_input_tokens_seen": 928382976,
"step": 787
},
{
"epoch": 0.022177657190043647,
"grad_norm": 0.96875,
"learning_rate": 7.95952764410662e-05,
"loss": 2.0609,
"num_input_tokens_seen": 929562624,
"step": 788
},
{
"epoch": 0.022295623451692816,
"grad_norm": 0.86328125,
"learning_rate": 7.95930595128859e-05,
"loss": 2.0261,
"num_input_tokens_seen": 930742272,
"step": 789
},
{
"epoch": 0.022413589713341984,
"grad_norm": 0.984375,
"learning_rate": 7.959083656055872e-05,
"loss": 1.9831,
"num_input_tokens_seen": 931921920,
"step": 790
},
{
"epoch": 0.022531555974991152,
"grad_norm": 1.5078125,
"learning_rate": 7.958860758442289e-05,
"loss": 2.0684,
"num_input_tokens_seen": 933101568,
"step": 791
},
{
"epoch": 0.02264952223664032,
"grad_norm": 0.85546875,
"learning_rate": 7.958637258481755e-05,
"loss": 2.114,
"num_input_tokens_seen": 934281216,
"step": 792
},
{
"epoch": 0.02276748849828949,
"grad_norm": 0.80859375,
"learning_rate": 7.958413156208275e-05,
"loss": 2.0545,
"num_input_tokens_seen": 935460864,
"step": 793
},
{
"epoch": 0.022885454759938657,
"grad_norm": 0.8046875,
"learning_rate": 7.958188451655949e-05,
"loss": 2.1056,
"num_input_tokens_seen": 936640512,
"step": 794
},
{
"epoch": 0.023003421021587825,
"grad_norm": 0.8359375,
"learning_rate": 7.957963144858964e-05,
"loss": 1.9934,
"num_input_tokens_seen": 937820160,
"step": 795
},
{
"epoch": 0.023121387283236993,
"grad_norm": 0.82421875,
"learning_rate": 7.957737235851602e-05,
"loss": 2.0787,
"num_input_tokens_seen": 938999808,
"step": 796
},
{
"epoch": 0.02323935354488616,
"grad_norm": 0.76171875,
"learning_rate": 7.957510724668234e-05,
"loss": 2.0125,
"num_input_tokens_seen": 940179456,
"step": 797
},
{
"epoch": 0.02335731980653533,
"grad_norm": 0.7734375,
"learning_rate": 7.957283611343325e-05,
"loss": 1.9983,
"num_input_tokens_seen": 941359104,
"step": 798
},
{
"epoch": 0.023475286068184498,
"grad_norm": 0.73046875,
"learning_rate": 7.95705589591143e-05,
"loss": 2.0102,
"num_input_tokens_seen": 942538752,
"step": 799
},
{
"epoch": 0.023593252329833666,
"grad_norm": 0.921875,
"learning_rate": 7.956827578407198e-05,
"loss": 2.0244,
"num_input_tokens_seen": 943718400,
"step": 800
},
{
"epoch": 0.023593252329833666,
"eval_wikipedia_loss": 2.2881267070770264,
"eval_wikipedia_runtime": 163.2839,
"eval_wikipedia_samples_per_second": 4.299,
"eval_wikipedia_steps_per_second": 0.184,
"num_input_tokens_seen": 943718400,
"step": 800
},
{
"epoch": 0.023593252329833666,
"eval_toxicity_loss": 4.0411834716796875,
"eval_toxicity_runtime": 0.9512,
"eval_toxicity_samples_per_second": 2.103,
"eval_toxicity_steps_per_second": 1.051,
"num_input_tokens_seen": 943718400,
"step": 800
},
{
"epoch": 0.023711218591482838,
"grad_norm": 0.80859375,
"learning_rate": 7.956598658865366e-05,
"loss": 2.0013,
"num_input_tokens_seen": 944898048,
"step": 801
},
{
"epoch": 0.023829184853132006,
"grad_norm": 0.8203125,
"learning_rate": 7.956369137320765e-05,
"loss": 2.1992,
"num_input_tokens_seen": 946077696,
"step": 802
},
{
"epoch": 0.023947151114781174,
"grad_norm": 0.95703125,
"learning_rate": 7.956139013808317e-05,
"loss": 1.9885,
"num_input_tokens_seen": 947257344,
"step": 803
},
{
"epoch": 0.024065117376430342,
"grad_norm": 1.515625,
"learning_rate": 7.955908288363037e-05,
"loss": 2.0842,
"num_input_tokens_seen": 948436992,
"step": 804
},
{
"epoch": 0.02418308363807951,
"grad_norm": 1.0859375,
"learning_rate": 7.955676961020027e-05,
"loss": 2.056,
"num_input_tokens_seen": 949616640,
"step": 805
},
{
"epoch": 0.02430104989972868,
"grad_norm": 0.7578125,
"learning_rate": 7.955445031814487e-05,
"loss": 1.9946,
"num_input_tokens_seen": 950796288,
"step": 806
},
{
"epoch": 0.024419016161377847,
"grad_norm": 1.078125,
"learning_rate": 7.955212500781705e-05,
"loss": 2.0638,
"num_input_tokens_seen": 951975936,
"step": 807
},
{
"epoch": 0.024536982423027015,
"grad_norm": 1.3515625,
"learning_rate": 7.954979367957059e-05,
"loss": 2.0086,
"num_input_tokens_seen": 953155584,
"step": 808
},
{
"epoch": 0.024654948684676183,
"grad_norm": 0.90625,
"learning_rate": 7.954745633376022e-05,
"loss": 1.9927,
"num_input_tokens_seen": 954335232,
"step": 809
},
{
"epoch": 0.02477291494632535,
"grad_norm": 1.0390625,
"learning_rate": 7.954511297074158e-05,
"loss": 2.074,
"num_input_tokens_seen": 955514880,
"step": 810
},
{
"epoch": 0.02489088120797452,
"grad_norm": 1.15625,
"learning_rate": 7.95427635908712e-05,
"loss": 2.0084,
"num_input_tokens_seen": 956694528,
"step": 811
},
{
"epoch": 0.025008847469623688,
"grad_norm": 0.8984375,
"learning_rate": 7.954040819450654e-05,
"loss": 2.0414,
"num_input_tokens_seen": 957874176,
"step": 812
},
{
"epoch": 0.025126813731272856,
"grad_norm": 1.0078125,
"learning_rate": 7.953804678200599e-05,
"loss": 2.0282,
"num_input_tokens_seen": 959053824,
"step": 813
},
{
"epoch": 0.025244779992922024,
"grad_norm": 1.078125,
"learning_rate": 7.953567935372883e-05,
"loss": 2.0436,
"num_input_tokens_seen": 960233472,
"step": 814
},
{
"epoch": 0.025362746254571193,
"grad_norm": 0.75,
"learning_rate": 7.953330591003528e-05,
"loss": 2.0458,
"num_input_tokens_seen": 961413120,
"step": 815
},
{
"epoch": 0.02548071251622036,
"grad_norm": 0.94140625,
"learning_rate": 7.953092645128646e-05,
"loss": 1.9329,
"num_input_tokens_seen": 962592768,
"step": 816
},
{
"epoch": 0.02559867877786953,
"grad_norm": 1.0703125,
"learning_rate": 7.95285409778444e-05,
"loss": 1.9266,
"num_input_tokens_seen": 963772416,
"step": 817
},
{
"epoch": 0.025716645039518697,
"grad_norm": 0.96484375,
"learning_rate": 7.952614949007208e-05,
"loss": 1.9903,
"num_input_tokens_seen": 964952064,
"step": 818
},
{
"epoch": 0.025834611301167865,
"grad_norm": 0.765625,
"learning_rate": 7.952375198833333e-05,
"loss": 2.1298,
"num_input_tokens_seen": 966131712,
"step": 819
},
{
"epoch": 0.025952577562817034,
"grad_norm": 0.8671875,
"learning_rate": 7.952134847299296e-05,
"loss": 2.0372,
"num_input_tokens_seen": 967311360,
"step": 820
},
{
"epoch": 0.026070543824466202,
"grad_norm": 0.90625,
"learning_rate": 7.951893894441665e-05,
"loss": 1.9484,
"num_input_tokens_seen": 968491008,
"step": 821
},
{
"epoch": 0.02618851008611537,
"grad_norm": 0.87109375,
"learning_rate": 7.951652340297105e-05,
"loss": 1.9851,
"num_input_tokens_seen": 969670656,
"step": 822
},
{
"epoch": 0.026306476347764538,
"grad_norm": 0.8046875,
"learning_rate": 7.951410184902364e-05,
"loss": 2.0184,
"num_input_tokens_seen": 970850304,
"step": 823
},
{
"epoch": 0.026424442609413706,
"grad_norm": 0.7578125,
"learning_rate": 7.951167428294288e-05,
"loss": 1.9319,
"num_input_tokens_seen": 972029952,
"step": 824
},
{
"epoch": 0.026542408871062875,
"grad_norm": 0.71875,
"learning_rate": 7.950924070509816e-05,
"loss": 2.0554,
"num_input_tokens_seen": 973209600,
"step": 825
},
{
"epoch": 0.026660375132712043,
"grad_norm": 0.76953125,
"learning_rate": 7.950680111585972e-05,
"loss": 2.0083,
"num_input_tokens_seen": 974389248,
"step": 826
},
{
"epoch": 0.02677834139436121,
"grad_norm": 0.74609375,
"learning_rate": 7.950435551559875e-05,
"loss": 2.1506,
"num_input_tokens_seen": 975568896,
"step": 827
},
{
"epoch": 0.026896307656010383,
"grad_norm": 0.796875,
"learning_rate": 7.950190390468734e-05,
"loss": 2.0305,
"num_input_tokens_seen": 976748544,
"step": 828
},
{
"epoch": 0.02701427391765955,
"grad_norm": 0.8515625,
"learning_rate": 7.949944628349855e-05,
"loss": 2.0154,
"num_input_tokens_seen": 977928192,
"step": 829
},
{
"epoch": 0.02713224017930872,
"grad_norm": 0.796875,
"learning_rate": 7.949698265240626e-05,
"loss": 1.918,
"num_input_tokens_seen": 979107840,
"step": 830
},
{
"epoch": 0.027250206440957887,
"grad_norm": 0.875,
"learning_rate": 7.949451301178536e-05,
"loss": 1.9816,
"num_input_tokens_seen": 980287488,
"step": 831
},
{
"epoch": 0.027368172702607056,
"grad_norm": 0.80859375,
"learning_rate": 7.949203736201159e-05,
"loss": 2.0066,
"num_input_tokens_seen": 981467136,
"step": 832
},
{
"epoch": 0.027486138964256224,
"grad_norm": 0.91015625,
"learning_rate": 7.948955570346161e-05,
"loss": 1.9281,
"num_input_tokens_seen": 982646784,
"step": 833
},
{
"epoch": 0.027604105225905392,
"grad_norm": 0.8125,
"learning_rate": 7.948706803651302e-05,
"loss": 1.9352,
"num_input_tokens_seen": 983826432,
"step": 834
},
{
"epoch": 0.02772207148755456,
"grad_norm": 1.0078125,
"learning_rate": 7.948457436154434e-05,
"loss": 1.9934,
"num_input_tokens_seen": 985006080,
"step": 835
},
{
"epoch": 0.02784003774920373,
"grad_norm": 0.8359375,
"learning_rate": 7.948207467893496e-05,
"loss": 2.0736,
"num_input_tokens_seen": 986185728,
"step": 836
},
{
"epoch": 0.027958004010852897,
"grad_norm": 0.859375,
"learning_rate": 7.947956898906523e-05,
"loss": 2.0934,
"num_input_tokens_seen": 987365376,
"step": 837
},
{
"epoch": 0.028075970272502065,
"grad_norm": 0.875,
"learning_rate": 7.947705729231637e-05,
"loss": 2.0025,
"num_input_tokens_seen": 988545024,
"step": 838
},
{
"epoch": 0.028193936534151233,
"grad_norm": 0.7734375,
"learning_rate": 7.947453958907057e-05,
"loss": 1.964,
"num_input_tokens_seen": 989724672,
"step": 839
},
{
"epoch": 0.0283119027958004,
"grad_norm": 0.7734375,
"learning_rate": 7.947201587971088e-05,
"loss": 2.0628,
"num_input_tokens_seen": 990904320,
"step": 840
},
{
"epoch": 0.02842986905744957,
"grad_norm": 0.88671875,
"learning_rate": 7.946948616462129e-05,
"loss": 1.9837,
"num_input_tokens_seen": 992083968,
"step": 841
},
{
"epoch": 0.028547835319098738,
"grad_norm": 1.09375,
"learning_rate": 7.94669504441867e-05,
"loss": 1.942,
"num_input_tokens_seen": 993263616,
"step": 842
},
{
"epoch": 0.028665801580747906,
"grad_norm": 1.1640625,
"learning_rate": 7.946440871879295e-05,
"loss": 1.9283,
"num_input_tokens_seen": 994443264,
"step": 843
},
{
"epoch": 0.028783767842397074,
"grad_norm": 0.8515625,
"learning_rate": 7.946186098882672e-05,
"loss": 1.969,
"num_input_tokens_seen": 995622912,
"step": 844
},
{
"epoch": 0.028901734104046242,
"grad_norm": 0.7578125,
"learning_rate": 7.94593072546757e-05,
"loss": 1.9535,
"num_input_tokens_seen": 996802560,
"step": 845
},
{
"epoch": 0.02901970036569541,
"grad_norm": 0.83984375,
"learning_rate": 7.94567475167284e-05,
"loss": 2.0129,
"num_input_tokens_seen": 997982208,
"step": 846
},
{
"epoch": 0.02913766662734458,
"grad_norm": 0.94921875,
"learning_rate": 7.945418177537432e-05,
"loss": 1.9015,
"num_input_tokens_seen": 999161856,
"step": 847
},
{
"epoch": 0.029255632888993747,
"grad_norm": 1.4921875,
"learning_rate": 7.945161003100383e-05,
"loss": 1.9187,
"num_input_tokens_seen": 1000341504,
"step": 848
},
{
"epoch": 0.029373599150642915,
"grad_norm": 1.0,
"learning_rate": 7.944903228400822e-05,
"loss": 1.9865,
"num_input_tokens_seen": 1001521152,
"step": 849
},
{
"epoch": 0.029491565412292083,
"grad_norm": 0.703125,
"learning_rate": 7.944644853477972e-05,
"loss": 2.0223,
"num_input_tokens_seen": 1002700800,
"step": 850
},
{
"epoch": 0.02960953167394125,
"grad_norm": 0.84375,
"learning_rate": 7.944385878371143e-05,
"loss": 1.9971,
"num_input_tokens_seen": 1003880448,
"step": 851
},
{
"epoch": 0.02972749793559042,
"grad_norm": 0.8046875,
"learning_rate": 7.944126303119738e-05,
"loss": 1.9669,
"num_input_tokens_seen": 1005060096,
"step": 852
},
{
"epoch": 0.029845464197239588,
"grad_norm": 0.7421875,
"learning_rate": 7.943866127763253e-05,
"loss": 1.9804,
"num_input_tokens_seen": 1006239744,
"step": 853
},
{
"epoch": 0.029963430458888756,
"grad_norm": 0.98046875,
"learning_rate": 7.943605352341274e-05,
"loss": 1.8757,
"num_input_tokens_seen": 1007419392,
"step": 854
},
{
"epoch": 0.030081396720537928,
"grad_norm": 1.046875,
"learning_rate": 7.943343976893479e-05,
"loss": 1.9259,
"num_input_tokens_seen": 1008599040,
"step": 855
},
{
"epoch": 0.030199362982187096,
"grad_norm": 0.90234375,
"learning_rate": 7.943082001459636e-05,
"loss": 2.0436,
"num_input_tokens_seen": 1009778688,
"step": 856
},
{
"epoch": 0.030317329243836264,
"grad_norm": 0.8125,
"learning_rate": 7.942819426079605e-05,
"loss": 1.987,
"num_input_tokens_seen": 1010958336,
"step": 857
},
{
"epoch": 0.030435295505485432,
"grad_norm": 0.75390625,
"learning_rate": 7.942556250793337e-05,
"loss": 2.1196,
"num_input_tokens_seen": 1012137984,
"step": 858
},
{
"epoch": 0.0305532617671346,
"grad_norm": 0.76953125,
"learning_rate": 7.942292475640875e-05,
"loss": 1.9059,
"num_input_tokens_seen": 1013317632,
"step": 859
},
{
"epoch": 0.03067122802878377,
"grad_norm": 0.82421875,
"learning_rate": 7.942028100662351e-05,
"loss": 2.0198,
"num_input_tokens_seen": 1014497280,
"step": 860
},
{
"epoch": 0.030789194290432937,
"grad_norm": 0.83984375,
"learning_rate": 7.941763125897995e-05,
"loss": 1.9257,
"num_input_tokens_seen": 1015676928,
"step": 861
},
{
"epoch": 0.030907160552082105,
"grad_norm": 0.9375,
"learning_rate": 7.941497551388116e-05,
"loss": 2.0592,
"num_input_tokens_seen": 1016856576,
"step": 862
},
{
"epoch": 0.031025126813731273,
"grad_norm": 1.1796875,
"learning_rate": 7.941231377173129e-05,
"loss": 1.9876,
"num_input_tokens_seen": 1018036224,
"step": 863
},
{
"epoch": 0.03114309307538044,
"grad_norm": 1.2734375,
"learning_rate": 7.940964603293528e-05,
"loss": 2.0618,
"num_input_tokens_seen": 1019215872,
"step": 864
},
{
"epoch": 0.031261059337029606,
"grad_norm": 0.9609375,
"learning_rate": 7.940697229789903e-05,
"loss": 1.9937,
"num_input_tokens_seen": 1020395520,
"step": 865
},
{
"epoch": 0.03137902559867878,
"grad_norm": 0.75,
"learning_rate": 7.940429256702939e-05,
"loss": 1.9126,
"num_input_tokens_seen": 1021575168,
"step": 866
},
{
"epoch": 0.03149699186032794,
"grad_norm": 0.92578125,
"learning_rate": 7.940160684073406e-05,
"loss": 1.97,
"num_input_tokens_seen": 1022754816,
"step": 867
},
{
"epoch": 0.031614958121977114,
"grad_norm": 1.1640625,
"learning_rate": 7.939891511942168e-05,
"loss": 1.9342,
"num_input_tokens_seen": 1023934464,
"step": 868
},
{
"epoch": 0.031732924383626286,
"grad_norm": 0.94921875,
"learning_rate": 7.939621740350179e-05,
"loss": 1.9973,
"num_input_tokens_seen": 1025114112,
"step": 869
},
{
"epoch": 0.03185089064527545,
"grad_norm": 0.91015625,
"learning_rate": 7.939351369338487e-05,
"loss": 1.9705,
"num_input_tokens_seen": 1026293760,
"step": 870
},
{
"epoch": 0.03196885690692462,
"grad_norm": 0.75,
"learning_rate": 7.939080398948229e-05,
"loss": 1.9954,
"num_input_tokens_seen": 1027473408,
"step": 871
},
{
"epoch": 0.03208682316857379,
"grad_norm": 0.96484375,
"learning_rate": 7.938808829220632e-05,
"loss": 2.0211,
"num_input_tokens_seen": 1028653056,
"step": 872
},
{
"epoch": 0.03220478943022296,
"grad_norm": 1.2265625,
"learning_rate": 7.938536660197018e-05,
"loss": 1.9562,
"num_input_tokens_seen": 1029832704,
"step": 873
},
{
"epoch": 0.032322755691872124,
"grad_norm": 1.1328125,
"learning_rate": 7.938263891918796e-05,
"loss": 1.927,
"num_input_tokens_seen": 1031012352,
"step": 874
},
{
"epoch": 0.032440721953521295,
"grad_norm": 0.83203125,
"learning_rate": 7.93799052442747e-05,
"loss": 2.042,
"num_input_tokens_seen": 1032192000,
"step": 875
},
{
"epoch": 0.03255868821517046,
"grad_norm": 1.1328125,
"learning_rate": 7.937716557764632e-05,
"loss": 1.9994,
"num_input_tokens_seen": 1033371648,
"step": 876
},
{
"epoch": 0.03267665447681963,
"grad_norm": 1.0703125,
"learning_rate": 7.937441991971966e-05,
"loss": 1.9857,
"num_input_tokens_seen": 1034551296,
"step": 877
},
{
"epoch": 0.0327946207384688,
"grad_norm": 0.84375,
"learning_rate": 7.93716682709125e-05,
"loss": 1.9931,
"num_input_tokens_seen": 1035730944,
"step": 878
},
{
"epoch": 0.03291258700011797,
"grad_norm": 0.8046875,
"learning_rate": 7.936891063164346e-05,
"loss": 1.8531,
"num_input_tokens_seen": 1036910592,
"step": 879
},
{
"epoch": 0.03303055326176713,
"grad_norm": 0.875,
"learning_rate": 7.936614700233218e-05,
"loss": 1.8975,
"num_input_tokens_seen": 1038090240,
"step": 880
},
{
"epoch": 0.033148519523416305,
"grad_norm": 0.72265625,
"learning_rate": 7.936337738339911e-05,
"loss": 1.9755,
"num_input_tokens_seen": 1039269888,
"step": 881
},
{
"epoch": 0.03326648578506547,
"grad_norm": 0.79296875,
"learning_rate": 7.936060177526567e-05,
"loss": 1.9892,
"num_input_tokens_seen": 1040449536,
"step": 882
},
{
"epoch": 0.03338445204671464,
"grad_norm": 0.91796875,
"learning_rate": 7.935782017835417e-05,
"loss": 1.8774,
"num_input_tokens_seen": 1041629184,
"step": 883
},
{
"epoch": 0.033502418308363806,
"grad_norm": 0.98828125,
"learning_rate": 7.935503259308782e-05,
"loss": 1.8458,
"num_input_tokens_seen": 1042808832,
"step": 884
},
{
"epoch": 0.03362038457001298,
"grad_norm": 0.89453125,
"learning_rate": 7.935223901989078e-05,
"loss": 1.9332,
"num_input_tokens_seen": 1043988480,
"step": 885
},
{
"epoch": 0.03373835083166214,
"grad_norm": 0.87109375,
"learning_rate": 7.934943945918806e-05,
"loss": 1.9231,
"num_input_tokens_seen": 1045168128,
"step": 886
},
{
"epoch": 0.033856317093311314,
"grad_norm": 1.421875,
"learning_rate": 7.934663391140566e-05,
"loss": 1.9749,
"num_input_tokens_seen": 1046347776,
"step": 887
},
{
"epoch": 0.03397428335496048,
"grad_norm": 0.8046875,
"learning_rate": 7.934382237697043e-05,
"loss": 1.9468,
"num_input_tokens_seen": 1047527424,
"step": 888
},
{
"epoch": 0.03409224961660965,
"grad_norm": 0.90234375,
"learning_rate": 7.934100485631014e-05,
"loss": 1.9363,
"num_input_tokens_seen": 1048707072,
"step": 889
},
{
"epoch": 0.034210215878258815,
"grad_norm": 0.85546875,
"learning_rate": 7.933818134985349e-05,
"loss": 1.9687,
"num_input_tokens_seen": 1049886720,
"step": 890
},
{
"epoch": 0.03432818213990799,
"grad_norm": 0.8125,
"learning_rate": 7.933535185803008e-05,
"loss": 2.0288,
"num_input_tokens_seen": 1051066368,
"step": 891
},
{
"epoch": 0.03444614840155715,
"grad_norm": 0.8125,
"learning_rate": 7.933251638127042e-05,
"loss": 1.9486,
"num_input_tokens_seen": 1052246016,
"step": 892
},
{
"epoch": 0.03456411466320632,
"grad_norm": 0.84375,
"learning_rate": 7.932967492000594e-05,
"loss": 1.8646,
"num_input_tokens_seen": 1053425664,
"step": 893
},
{
"epoch": 0.03468208092485549,
"grad_norm": 0.8515625,
"learning_rate": 7.932682747466896e-05,
"loss": 1.9055,
"num_input_tokens_seen": 1054605312,
"step": 894
},
{
"epoch": 0.03480004718650466,
"grad_norm": 0.8203125,
"learning_rate": 7.932397404569274e-05,
"loss": 1.9025,
"num_input_tokens_seen": 1055784960,
"step": 895
},
{
"epoch": 0.03491801344815383,
"grad_norm": 0.75390625,
"learning_rate": 7.932111463351142e-05,
"loss": 1.9739,
"num_input_tokens_seen": 1056964608,
"step": 896
},
{
"epoch": 0.035035979709802996,
"grad_norm": 0.73828125,
"learning_rate": 7.931824923856006e-05,
"loss": 2.0355,
"num_input_tokens_seen": 1058144256,
"step": 897
},
{
"epoch": 0.03515394597145217,
"grad_norm": 0.94921875,
"learning_rate": 7.931537786127464e-05,
"loss": 2.0128,
"num_input_tokens_seen": 1059323904,
"step": 898
},
{
"epoch": 0.03527191223310133,
"grad_norm": 0.7578125,
"learning_rate": 7.931250050209206e-05,
"loss": 1.9691,
"num_input_tokens_seen": 1060503552,
"step": 899
},
{
"epoch": 0.035389878494750504,
"grad_norm": 0.74609375,
"learning_rate": 7.93096171614501e-05,
"loss": 2.0682,
"num_input_tokens_seen": 1061683200,
"step": 900
},
{
"epoch": 0.03550784475639967,
"grad_norm": 0.83203125,
"learning_rate": 7.930672783978745e-05,
"loss": 1.9375,
"num_input_tokens_seen": 1062862848,
"step": 901
},
{
"epoch": 0.03562581101804884,
"grad_norm": 1.1015625,
"learning_rate": 7.930383253754376e-05,
"loss": 1.945,
"num_input_tokens_seen": 1064042496,
"step": 902
},
{
"epoch": 0.035743777279698005,
"grad_norm": 1.140625,
"learning_rate": 7.930093125515954e-05,
"loss": 1.9439,
"num_input_tokens_seen": 1065222144,
"step": 903
},
{
"epoch": 0.03586174354134718,
"grad_norm": 0.984375,
"learning_rate": 7.92980239930762e-05,
"loss": 1.8896,
"num_input_tokens_seen": 1066401792,
"step": 904
},
{
"epoch": 0.03597970980299634,
"grad_norm": 1.03125,
"learning_rate": 7.929511075173612e-05,
"loss": 1.9784,
"num_input_tokens_seen": 1067581440,
"step": 905
},
{
"epoch": 0.03609767606464551,
"grad_norm": 0.84765625,
"learning_rate": 7.929219153158253e-05,
"loss": 2.0275,
"num_input_tokens_seen": 1068761088,
"step": 906
},
{
"epoch": 0.03621564232629468,
"grad_norm": 0.95703125,
"learning_rate": 7.928926633305962e-05,
"loss": 1.822,
"num_input_tokens_seen": 1069940736,
"step": 907
},
{
"epoch": 0.03633360858794385,
"grad_norm": 1.015625,
"learning_rate": 7.928633515661242e-05,
"loss": 1.9427,
"num_input_tokens_seen": 1071120384,
"step": 908
},
{
"epoch": 0.036451574849593014,
"grad_norm": 1.0859375,
"learning_rate": 7.928339800268697e-05,
"loss": 1.9247,
"num_input_tokens_seen": 1072300032,
"step": 909
},
{
"epoch": 0.036569541111242186,
"grad_norm": 0.9921875,
"learning_rate": 7.928045487173013e-05,
"loss": 1.8571,
"num_input_tokens_seen": 1073479680,
"step": 910
},
{
"epoch": 0.03668750737289135,
"grad_norm": 0.875,
"learning_rate": 7.927750576418968e-05,
"loss": 1.987,
"num_input_tokens_seen": 1074659328,
"step": 911
},
{
"epoch": 0.03680547363454052,
"grad_norm": 0.9296875,
"learning_rate": 7.927455068051437e-05,
"loss": 2.0172,
"num_input_tokens_seen": 1075838976,
"step": 912
},
{
"epoch": 0.03692343989618969,
"grad_norm": 0.78125,
"learning_rate": 7.927158962115382e-05,
"loss": 1.9209,
"num_input_tokens_seen": 1077018624,
"step": 913
},
{
"epoch": 0.03704140615783886,
"grad_norm": 0.9140625,
"learning_rate": 7.926862258655853e-05,
"loss": 1.8831,
"num_input_tokens_seen": 1078198272,
"step": 914
},
{
"epoch": 0.037159372419488024,
"grad_norm": 0.91015625,
"learning_rate": 7.926564957717998e-05,
"loss": 1.8794,
"num_input_tokens_seen": 1079377920,
"step": 915
},
{
"epoch": 0.037277338681137195,
"grad_norm": 0.8828125,
"learning_rate": 7.926267059347046e-05,
"loss": 1.8865,
"num_input_tokens_seen": 1080557568,
"step": 916
},
{
"epoch": 0.03739530494278636,
"grad_norm": 0.9140625,
"learning_rate": 7.925968563588328e-05,
"loss": 1.83,
"num_input_tokens_seen": 1081737216,
"step": 917
},
{
"epoch": 0.03751327120443553,
"grad_norm": 0.92578125,
"learning_rate": 7.925669470487258e-05,
"loss": 1.9207,
"num_input_tokens_seen": 1082916864,
"step": 918
},
{
"epoch": 0.0376312374660847,
"grad_norm": 1.0078125,
"learning_rate": 7.925369780089345e-05,
"loss": 1.9191,
"num_input_tokens_seen": 1084096512,
"step": 919
},
{
"epoch": 0.03774920372773387,
"grad_norm": 0.79296875,
"learning_rate": 7.925069492440188e-05,
"loss": 1.8717,
"num_input_tokens_seen": 1085276160,
"step": 920
},
{
"epoch": 0.03786716998938304,
"grad_norm": 0.83203125,
"learning_rate": 7.924768607585472e-05,
"loss": 1.9212,
"num_input_tokens_seen": 1086455808,
"step": 921
},
{
"epoch": 0.037985136251032205,
"grad_norm": 0.765625,
"learning_rate": 7.924467125570982e-05,
"loss": 1.9191,
"num_input_tokens_seen": 1087635456,
"step": 922
},
{
"epoch": 0.038103102512681376,
"grad_norm": 0.91796875,
"learning_rate": 7.924165046442586e-05,
"loss": 1.9994,
"num_input_tokens_seen": 1088815104,
"step": 923
},
{
"epoch": 0.03822106877433054,
"grad_norm": 0.75390625,
"learning_rate": 7.923862370246247e-05,
"loss": 1.9119,
"num_input_tokens_seen": 1089994752,
"step": 924
},
{
"epoch": 0.03833903503597971,
"grad_norm": 0.78125,
"learning_rate": 7.923559097028017e-05,
"loss": 1.9034,
"num_input_tokens_seen": 1091174400,
"step": 925
},
{
"epoch": 0.03845700129762888,
"grad_norm": 0.90234375,
"learning_rate": 7.92325522683404e-05,
"loss": 1.9324,
"num_input_tokens_seen": 1092354048,
"step": 926
},
{
"epoch": 0.03857496755927805,
"grad_norm": 0.765625,
"learning_rate": 7.92295075971055e-05,
"loss": 1.9267,
"num_input_tokens_seen": 1093533696,
"step": 927
},
{
"epoch": 0.038692933820927214,
"grad_norm": 0.828125,
"learning_rate": 7.922645695703873e-05,
"loss": 1.9072,
"num_input_tokens_seen": 1094713344,
"step": 928
},
{
"epoch": 0.038810900082576386,
"grad_norm": 0.9609375,
"learning_rate": 7.922340034860424e-05,
"loss": 1.9171,
"num_input_tokens_seen": 1095892992,
"step": 929
},
{
"epoch": 0.03892886634422555,
"grad_norm": 0.9296875,
"learning_rate": 7.922033777226711e-05,
"loss": 1.9862,
"num_input_tokens_seen": 1097072640,
"step": 930
},
{
"epoch": 0.03904683260587472,
"grad_norm": 0.95703125,
"learning_rate": 7.92172692284933e-05,
"loss": 1.9236,
"num_input_tokens_seen": 1098252288,
"step": 931
},
{
"epoch": 0.03916479886752389,
"grad_norm": 0.7890625,
"learning_rate": 7.92141947177497e-05,
"loss": 1.9795,
"num_input_tokens_seen": 1099431936,
"step": 932
},
{
"epoch": 0.03928276512917306,
"grad_norm": 0.875,
"learning_rate": 7.92111142405041e-05,
"loss": 1.9381,
"num_input_tokens_seen": 1100611584,
"step": 933
},
{
"epoch": 0.03940073139082222,
"grad_norm": 0.8046875,
"learning_rate": 7.92080277972252e-05,
"loss": 1.8348,
"num_input_tokens_seen": 1101791232,
"step": 934
},
{
"epoch": 0.039518697652471395,
"grad_norm": 0.8203125,
"learning_rate": 7.920493538838262e-05,
"loss": 1.8595,
"num_input_tokens_seen": 1102970880,
"step": 935
},
{
"epoch": 0.03963666391412056,
"grad_norm": 0.921875,
"learning_rate": 7.920183701444686e-05,
"loss": 1.8674,
"num_input_tokens_seen": 1104150528,
"step": 936
},
{
"epoch": 0.03975463017576973,
"grad_norm": 0.88671875,
"learning_rate": 7.919873267588936e-05,
"loss": 1.8276,
"num_input_tokens_seen": 1105330176,
"step": 937
},
{
"epoch": 0.039872596437418896,
"grad_norm": 0.78515625,
"learning_rate": 7.919562237318243e-05,
"loss": 1.8478,
"num_input_tokens_seen": 1106509824,
"step": 938
},
{
"epoch": 0.03999056269906807,
"grad_norm": 0.765625,
"learning_rate": 7.91925061067993e-05,
"loss": 1.9069,
"num_input_tokens_seen": 1107689472,
"step": 939
},
{
"epoch": 0.04010852896071723,
"grad_norm": 0.79296875,
"learning_rate": 7.918938387721413e-05,
"loss": 1.8889,
"num_input_tokens_seen": 1108869120,
"step": 940
},
{
"epoch": 0.040226495222366404,
"grad_norm": 0.828125,
"learning_rate": 7.918625568490199e-05,
"loss": 1.9663,
"num_input_tokens_seen": 1110048768,
"step": 941
},
{
"epoch": 0.04034446148401557,
"grad_norm": 0.953125,
"learning_rate": 7.918312153033882e-05,
"loss": 1.9711,
"num_input_tokens_seen": 1111228416,
"step": 942
},
{
"epoch": 0.04046242774566474,
"grad_norm": 1.046875,
"learning_rate": 7.917998141400147e-05,
"loss": 1.9236,
"num_input_tokens_seen": 1112408064,
"step": 943
},
{
"epoch": 0.040580394007313905,
"grad_norm": 1.0703125,
"learning_rate": 7.917683533636773e-05,
"loss": 1.9647,
"num_input_tokens_seen": 1113587712,
"step": 944
},
{
"epoch": 0.04069836026896308,
"grad_norm": 0.97265625,
"learning_rate": 7.91736832979163e-05,
"loss": 1.9309,
"num_input_tokens_seen": 1114767360,
"step": 945
},
{
"epoch": 0.04081632653061224,
"grad_norm": 0.81640625,
"learning_rate": 7.917052529912675e-05,
"loss": 2.0155,
"num_input_tokens_seen": 1115947008,
"step": 946
},
{
"epoch": 0.04093429279226141,
"grad_norm": 0.8359375,
"learning_rate": 7.916736134047956e-05,
"loss": 1.8469,
"num_input_tokens_seen": 1117126656,
"step": 947
},
{
"epoch": 0.041052259053910585,
"grad_norm": 0.890625,
"learning_rate": 7.916419142245615e-05,
"loss": 1.9435,
"num_input_tokens_seen": 1118306304,
"step": 948
},
{
"epoch": 0.04117022531555975,
"grad_norm": 1.15625,
"learning_rate": 7.916101554553882e-05,
"loss": 1.8697,
"num_input_tokens_seen": 1119485952,
"step": 949
},
{
"epoch": 0.04128819157720892,
"grad_norm": 0.95703125,
"learning_rate": 7.91578337102108e-05,
"loss": 1.979,
"num_input_tokens_seen": 1120665600,
"step": 950
},
{
"epoch": 0.041406157838858086,
"grad_norm": 0.71875,
"learning_rate": 7.91546459169562e-05,
"loss": 1.9996,
"num_input_tokens_seen": 1121845248,
"step": 951
},
{
"epoch": 0.04152412410050726,
"grad_norm": 0.734375,
"learning_rate": 7.915145216626004e-05,
"loss": 1.9028,
"num_input_tokens_seen": 1123024896,
"step": 952
},
{
"epoch": 0.04164209036215642,
"grad_norm": 0.77734375,
"learning_rate": 7.914825245860827e-05,
"loss": 1.9244,
"num_input_tokens_seen": 1124204544,
"step": 953
},
{
"epoch": 0.041760056623805594,
"grad_norm": 0.79296875,
"learning_rate": 7.914504679448771e-05,
"loss": 2.0091,
"num_input_tokens_seen": 1125384192,
"step": 954
},
{
"epoch": 0.04187802288545476,
"grad_norm": 0.7109375,
"learning_rate": 7.914183517438612e-05,
"loss": 1.881,
"num_input_tokens_seen": 1126563840,
"step": 955
},
{
"epoch": 0.04199598914710393,
"grad_norm": 1.03125,
"learning_rate": 7.913861759879215e-05,
"loss": 1.9797,
"num_input_tokens_seen": 1127743488,
"step": 956
},
{
"epoch": 0.042113955408753095,
"grad_norm": 1.0078125,
"learning_rate": 7.913539406819537e-05,
"loss": 1.8792,
"num_input_tokens_seen": 1128923136,
"step": 957
},
{
"epoch": 0.04223192167040227,
"grad_norm": 0.921875,
"learning_rate": 7.913216458308624e-05,
"loss": 1.8002,
"num_input_tokens_seen": 1130102784,
"step": 958
},
{
"epoch": 0.04234988793205143,
"grad_norm": 1.0703125,
"learning_rate": 7.912892914395612e-05,
"loss": 1.9734,
"num_input_tokens_seen": 1131282432,
"step": 959
},
{
"epoch": 0.042467854193700603,
"grad_norm": 1.0078125,
"learning_rate": 7.91256877512973e-05,
"loss": 1.8323,
"num_input_tokens_seen": 1132462080,
"step": 960
},
{
"epoch": 0.04258582045534977,
"grad_norm": 0.87109375,
"learning_rate": 7.912244040560296e-05,
"loss": 1.8318,
"num_input_tokens_seen": 1133641728,
"step": 961
},
{
"epoch": 0.04270378671699894,
"grad_norm": 0.87890625,
"learning_rate": 7.911918710736718e-05,
"loss": 1.8872,
"num_input_tokens_seen": 1134821376,
"step": 962
},
{
"epoch": 0.042821752978648105,
"grad_norm": 1.0078125,
"learning_rate": 7.911592785708496e-05,
"loss": 1.9225,
"num_input_tokens_seen": 1136001024,
"step": 963
},
{
"epoch": 0.042939719240297276,
"grad_norm": 1.1015625,
"learning_rate": 7.911266265525221e-05,
"loss": 1.7839,
"num_input_tokens_seen": 1137180672,
"step": 964
},
{
"epoch": 0.04305768550194644,
"grad_norm": 0.88671875,
"learning_rate": 7.910939150236573e-05,
"loss": 1.8849,
"num_input_tokens_seen": 1138360320,
"step": 965
},
{
"epoch": 0.04317565176359561,
"grad_norm": 1.3125,
"learning_rate": 7.910611439892322e-05,
"loss": 1.8538,
"num_input_tokens_seen": 1139539968,
"step": 966
},
{
"epoch": 0.04329361802524478,
"grad_norm": 1.1171875,
"learning_rate": 7.91028313454233e-05,
"loss": 1.8658,
"num_input_tokens_seen": 1140719616,
"step": 967
},
{
"epoch": 0.04341158428689395,
"grad_norm": 0.89453125,
"learning_rate": 7.909954234236551e-05,
"loss": 1.859,
"num_input_tokens_seen": 1141899264,
"step": 968
},
{
"epoch": 0.043529550548543114,
"grad_norm": 0.890625,
"learning_rate": 7.909624739025026e-05,
"loss": 1.8266,
"num_input_tokens_seen": 1143078912,
"step": 969
},
{
"epoch": 0.043647516810192286,
"grad_norm": 0.87109375,
"learning_rate": 7.909294648957889e-05,
"loss": 1.819,
"num_input_tokens_seen": 1144258560,
"step": 970
},
{
"epoch": 0.04376548307184145,
"grad_norm": 1.1875,
"learning_rate": 7.908963964085362e-05,
"loss": 1.8053,
"num_input_tokens_seen": 1145438208,
"step": 971
},
{
"epoch": 0.04388344933349062,
"grad_norm": 0.92578125,
"learning_rate": 7.908632684457762e-05,
"loss": 1.8585,
"num_input_tokens_seen": 1146617856,
"step": 972
},
{
"epoch": 0.04400141559513979,
"grad_norm": 0.80859375,
"learning_rate": 7.908300810125491e-05,
"loss": 1.8612,
"num_input_tokens_seen": 1147797504,
"step": 973
},
{
"epoch": 0.04411938185678896,
"grad_norm": 0.953125,
"learning_rate": 7.907968341139046e-05,
"loss": 1.7937,
"num_input_tokens_seen": 1148977152,
"step": 974
},
{
"epoch": 0.04423734811843813,
"grad_norm": 0.85546875,
"learning_rate": 7.907635277549013e-05,
"loss": 1.9481,
"num_input_tokens_seen": 1150156800,
"step": 975
},
{
"epoch": 0.044355314380087295,
"grad_norm": 0.85546875,
"learning_rate": 7.907301619406067e-05,
"loss": 2.0073,
"num_input_tokens_seen": 1151336448,
"step": 976
},
{
"epoch": 0.044473280641736467,
"grad_norm": 0.9375,
"learning_rate": 7.906967366760974e-05,
"loss": 1.8315,
"num_input_tokens_seen": 1152516096,
"step": 977
},
{
"epoch": 0.04459124690338563,
"grad_norm": 0.9453125,
"learning_rate": 7.906632519664592e-05,
"loss": 1.9134,
"num_input_tokens_seen": 1153695744,
"step": 978
},
{
"epoch": 0.0447092131650348,
"grad_norm": 1.15625,
"learning_rate": 7.90629707816787e-05,
"loss": 1.8331,
"num_input_tokens_seen": 1154875392,
"step": 979
},
{
"epoch": 0.04482717942668397,
"grad_norm": 1.1484375,
"learning_rate": 7.905961042321842e-05,
"loss": 1.8492,
"num_input_tokens_seen": 1156055040,
"step": 980
},
{
"epoch": 0.04494514568833314,
"grad_norm": 0.890625,
"learning_rate": 7.905624412177642e-05,
"loss": 1.9344,
"num_input_tokens_seen": 1157234688,
"step": 981
},
{
"epoch": 0.045063111949982304,
"grad_norm": 0.96484375,
"learning_rate": 7.905287187786483e-05,
"loss": 1.8134,
"num_input_tokens_seen": 1158414336,
"step": 982
},
{
"epoch": 0.045181078211631476,
"grad_norm": 1.03125,
"learning_rate": 7.904949369199678e-05,
"loss": 1.9497,
"num_input_tokens_seen": 1159593984,
"step": 983
},
{
"epoch": 0.04529904447328064,
"grad_norm": 0.953125,
"learning_rate": 7.904610956468626e-05,
"loss": 1.7974,
"num_input_tokens_seen": 1160773632,
"step": 984
},
{
"epoch": 0.04541701073492981,
"grad_norm": 0.8359375,
"learning_rate": 7.904271949644816e-05,
"loss": 1.8029,
"num_input_tokens_seen": 1161953280,
"step": 985
},
{
"epoch": 0.04553497699657898,
"grad_norm": 0.8046875,
"learning_rate": 7.903932348779829e-05,
"loss": 1.9024,
"num_input_tokens_seen": 1163132928,
"step": 986
},
{
"epoch": 0.04565294325822815,
"grad_norm": 1.1875,
"learning_rate": 7.903592153925336e-05,
"loss": 1.8734,
"num_input_tokens_seen": 1164312576,
"step": 987
},
{
"epoch": 0.04577090951987731,
"grad_norm": 0.78125,
"learning_rate": 7.903251365133098e-05,
"loss": 1.8746,
"num_input_tokens_seen": 1165492224,
"step": 988
},
{
"epoch": 0.045888875781526485,
"grad_norm": 0.79296875,
"learning_rate": 7.902909982454966e-05,
"loss": 1.9698,
"num_input_tokens_seen": 1166671872,
"step": 989
},
{
"epoch": 0.04600684204317565,
"grad_norm": 0.8359375,
"learning_rate": 7.902568005942882e-05,
"loss": 1.8322,
"num_input_tokens_seen": 1167851520,
"step": 990
},
{
"epoch": 0.04612480830482482,
"grad_norm": 0.88671875,
"learning_rate": 7.902225435648881e-05,
"loss": 1.8545,
"num_input_tokens_seen": 1169031168,
"step": 991
},
{
"epoch": 0.046242774566473986,
"grad_norm": 0.79296875,
"learning_rate": 7.901882271625082e-05,
"loss": 1.8321,
"num_input_tokens_seen": 1170210816,
"step": 992
},
{
"epoch": 0.04636074082812316,
"grad_norm": 0.83984375,
"learning_rate": 7.901538513923699e-05,
"loss": 1.8958,
"num_input_tokens_seen": 1171390464,
"step": 993
},
{
"epoch": 0.04647870708977232,
"grad_norm": 1.0078125,
"learning_rate": 7.901194162597036e-05,
"loss": 1.8725,
"num_input_tokens_seen": 1172570112,
"step": 994
},
{
"epoch": 0.046596673351421494,
"grad_norm": 0.85546875,
"learning_rate": 7.900849217697486e-05,
"loss": 1.9313,
"num_input_tokens_seen": 1173749760,
"step": 995
},
{
"epoch": 0.04671463961307066,
"grad_norm": 1.0078125,
"learning_rate": 7.900503679277534e-05,
"loss": 1.9648,
"num_input_tokens_seen": 1174929408,
"step": 996
},
{
"epoch": 0.04683260587471983,
"grad_norm": 0.7265625,
"learning_rate": 7.900157547389752e-05,
"loss": 2.073,
"num_input_tokens_seen": 1176109056,
"step": 997
},
{
"epoch": 0.046950572136368995,
"grad_norm": 0.98828125,
"learning_rate": 7.899810822086806e-05,
"loss": 1.8845,
"num_input_tokens_seen": 1177288704,
"step": 998
},
{
"epoch": 0.04706853839801817,
"grad_norm": 0.921875,
"learning_rate": 7.899463503421451e-05,
"loss": 1.8769,
"num_input_tokens_seen": 1178468352,
"step": 999
},
{
"epoch": 0.04718650465966733,
"grad_norm": 0.98046875,
"learning_rate": 7.899115591446533e-05,
"loss": 1.9261,
"num_input_tokens_seen": 1179648000,
"step": 1000
},
{
"epoch": 0.04718650465966733,
"eval_wikipedia_loss": 2.2636232376098633,
"eval_wikipedia_runtime": 162.9449,
"eval_wikipedia_samples_per_second": 4.308,
"eval_wikipedia_steps_per_second": 0.184,
"num_input_tokens_seen": 1179648000,
"step": 1000
},
{
"epoch": 0.04718650465966733,
"eval_toxicity_loss": 3.9912185668945312,
"eval_toxicity_runtime": 0.9818,
"eval_toxicity_samples_per_second": 2.037,
"eval_toxicity_steps_per_second": 1.018,
"num_input_tokens_seen": 1179648000,
"step": 1000
},
{
"epoch": 0.00011796626164916834,
"grad_norm": 1.125,
"learning_rate": 7.898767086214984e-05,
"loss": 1.693,
"num_input_tokens_seen": 1180827648,
"step": 1001
},
{
"epoch": 0.0002359325232983367,
"grad_norm": 1.15625,
"learning_rate": 7.898417987779834e-05,
"loss": 1.7842,
"num_input_tokens_seen": 1182007296,
"step": 1002
},
{
"epoch": 0.000353898784947505,
"grad_norm": 1.1953125,
"learning_rate": 7.898068296194194e-05,
"loss": 1.6674,
"num_input_tokens_seen": 1183186944,
"step": 1003
},
{
"epoch": 0.0004718650465966734,
"grad_norm": 0.9453125,
"learning_rate": 7.897718011511275e-05,
"loss": 1.6872,
"num_input_tokens_seen": 1184366592,
"step": 1004
},
{
"epoch": 0.0005898313082458417,
"grad_norm": 1.0234375,
"learning_rate": 7.89736713378437e-05,
"loss": 1.7766,
"num_input_tokens_seen": 1185546240,
"step": 1005
},
{
"epoch": 0.00070779756989501,
"grad_norm": 1.1015625,
"learning_rate": 7.897015663066865e-05,
"loss": 1.7041,
"num_input_tokens_seen": 1186725888,
"step": 1006
},
{
"epoch": 0.0008257638315441783,
"grad_norm": 1.1484375,
"learning_rate": 7.896663599412241e-05,
"loss": 1.8234,
"num_input_tokens_seen": 1187905536,
"step": 1007
},
{
"epoch": 0.0009437300931933467,
"grad_norm": 1.3515625,
"learning_rate": 7.896310942874061e-05,
"loss": 1.6623,
"num_input_tokens_seen": 1189085184,
"step": 1008
},
{
"epoch": 0.001061696354842515,
"grad_norm": 1.0625,
"learning_rate": 7.895957693505987e-05,
"loss": 1.7256,
"num_input_tokens_seen": 1190264832,
"step": 1009
},
{
"epoch": 0.0011796626164916834,
"grad_norm": 1.3203125,
"learning_rate": 7.89560385136176e-05,
"loss": 1.832,
"num_input_tokens_seen": 1191444480,
"step": 1010
},
{
"epoch": 0.0012976288781408518,
"grad_norm": 1.1953125,
"learning_rate": 7.895249416495223e-05,
"loss": 1.7132,
"num_input_tokens_seen": 1192624128,
"step": 1011
},
{
"epoch": 0.00141559513979002,
"grad_norm": 0.98046875,
"learning_rate": 7.8948943889603e-05,
"loss": 1.7064,
"num_input_tokens_seen": 1193803776,
"step": 1012
},
{
"epoch": 0.0015335614014391884,
"grad_norm": 1.1171875,
"learning_rate": 7.894538768811011e-05,
"loss": 1.5843,
"num_input_tokens_seen": 1194983424,
"step": 1013
},
{
"epoch": 0.0016515276630883566,
"grad_norm": 0.99609375,
"learning_rate": 7.894182556101465e-05,
"loss": 1.6346,
"num_input_tokens_seen": 1196163072,
"step": 1014
},
{
"epoch": 0.001769493924737525,
"grad_norm": 0.8984375,
"learning_rate": 7.893825750885859e-05,
"loss": 1.6726,
"num_input_tokens_seen": 1197342720,
"step": 1015
},
{
"epoch": 0.0018874601863866935,
"grad_norm": 0.9921875,
"learning_rate": 7.89346835321848e-05,
"loss": 1.6584,
"num_input_tokens_seen": 1198522368,
"step": 1016
},
{
"epoch": 0.0020054264480358617,
"grad_norm": 1.2421875,
"learning_rate": 7.89311036315371e-05,
"loss": 1.7154,
"num_input_tokens_seen": 1199702016,
"step": 1017
},
{
"epoch": 0.00212339270968503,
"grad_norm": 0.8828125,
"learning_rate": 7.892751780746016e-05,
"loss": 1.8305,
"num_input_tokens_seen": 1200881664,
"step": 1018
},
{
"epoch": 0.0022413589713341986,
"grad_norm": 0.85546875,
"learning_rate": 7.892392606049957e-05,
"loss": 1.7607,
"num_input_tokens_seen": 1202061312,
"step": 1019
},
{
"epoch": 0.0023593252329833668,
"grad_norm": 1.109375,
"learning_rate": 7.892032839120181e-05,
"loss": 1.7711,
"num_input_tokens_seen": 1203240960,
"step": 1020
},
{
"epoch": 0.002477291494632535,
"grad_norm": 1.21875,
"learning_rate": 7.89167248001143e-05,
"loss": 1.7846,
"num_input_tokens_seen": 1204420608,
"step": 1021
},
{
"epoch": 0.0025952577562817036,
"grad_norm": 0.9453125,
"learning_rate": 7.89131152877853e-05,
"loss": 1.7185,
"num_input_tokens_seen": 1205600256,
"step": 1022
},
{
"epoch": 0.002713224017930872,
"grad_norm": 1.3203125,
"learning_rate": 7.890949985476401e-05,
"loss": 1.6928,
"num_input_tokens_seen": 1206779904,
"step": 1023
},
{
"epoch": 0.00283119027958004,
"grad_norm": 0.9453125,
"learning_rate": 7.890587850160054e-05,
"loss": 1.844,
"num_input_tokens_seen": 1207959552,
"step": 1024
},
{
"epoch": 0.0029491565412292082,
"grad_norm": 1.09375,
"learning_rate": 7.890225122884587e-05,
"loss": 1.7122,
"num_input_tokens_seen": 1209139200,
"step": 1025
},
{
"epoch": 0.003067122802878377,
"grad_norm": 1.296875,
"learning_rate": 7.88986180370519e-05,
"loss": 1.7787,
"num_input_tokens_seen": 1210318848,
"step": 1026
},
{
"epoch": 0.003185089064527545,
"grad_norm": 0.97265625,
"learning_rate": 7.889497892677144e-05,
"loss": 1.6745,
"num_input_tokens_seen": 1211498496,
"step": 1027
},
{
"epoch": 0.0033030553261767133,
"grad_norm": 1.2421875,
"learning_rate": 7.889133389855816e-05,
"loss": 1.6877,
"num_input_tokens_seen": 1212678144,
"step": 1028
},
{
"epoch": 0.003421021587825882,
"grad_norm": 0.96875,
"learning_rate": 7.888768295296667e-05,
"loss": 1.6812,
"num_input_tokens_seen": 1213857792,
"step": 1029
},
{
"epoch": 0.00353898784947505,
"grad_norm": 1.40625,
"learning_rate": 7.888402609055247e-05,
"loss": 1.6829,
"num_input_tokens_seen": 1215037440,
"step": 1030
},
{
"epoch": 0.0036569541111242184,
"grad_norm": 0.9609375,
"learning_rate": 7.888036331187195e-05,
"loss": 1.698,
"num_input_tokens_seen": 1216217088,
"step": 1031
},
{
"epoch": 0.003774920372773387,
"grad_norm": 1.5234375,
"learning_rate": 7.887669461748241e-05,
"loss": 1.8116,
"num_input_tokens_seen": 1217396736,
"step": 1032
},
{
"epoch": 0.003892886634422555,
"grad_norm": 0.94140625,
"learning_rate": 7.887302000794204e-05,
"loss": 1.7049,
"num_input_tokens_seen": 1218576384,
"step": 1033
},
{
"epoch": 0.004010852896071723,
"grad_norm": 1.0703125,
"learning_rate": 7.886933948380997e-05,
"loss": 1.6622,
"num_input_tokens_seen": 1219756032,
"step": 1034
},
{
"epoch": 0.004128819157720892,
"grad_norm": 1.078125,
"learning_rate": 7.886565304564616e-05,
"loss": 1.7226,
"num_input_tokens_seen": 1220935680,
"step": 1035
},
{
"epoch": 0.00424678541937006,
"grad_norm": 0.7890625,
"learning_rate": 7.886196069401153e-05,
"loss": 1.8099,
"num_input_tokens_seen": 1222115328,
"step": 1036
},
{
"epoch": 0.004364751681019229,
"grad_norm": 1.140625,
"learning_rate": 7.885826242946786e-05,
"loss": 1.6033,
"num_input_tokens_seen": 1223294976,
"step": 1037
},
{
"epoch": 0.004482717942668397,
"grad_norm": 0.87109375,
"learning_rate": 7.885455825257785e-05,
"loss": 1.8159,
"num_input_tokens_seen": 1224474624,
"step": 1038
},
{
"epoch": 0.004600684204317565,
"grad_norm": 1.0078125,
"learning_rate": 7.885084816390511e-05,
"loss": 1.7535,
"num_input_tokens_seen": 1225654272,
"step": 1039
},
{
"epoch": 0.0047186504659667335,
"grad_norm": 0.9140625,
"learning_rate": 7.884713216401412e-05,
"loss": 1.8334,
"num_input_tokens_seen": 1226833920,
"step": 1040
},
{
"epoch": 0.004836616727615902,
"grad_norm": 0.9609375,
"learning_rate": 7.88434102534703e-05,
"loss": 1.6799,
"num_input_tokens_seen": 1228013568,
"step": 1041
},
{
"epoch": 0.00495458298926507,
"grad_norm": 0.99609375,
"learning_rate": 7.883968243283992e-05,
"loss": 1.6823,
"num_input_tokens_seen": 1229193216,
"step": 1042
},
{
"epoch": 0.005072549250914238,
"grad_norm": 1.125,
"learning_rate": 7.883594870269019e-05,
"loss": 1.7959,
"num_input_tokens_seen": 1230372864,
"step": 1043
},
{
"epoch": 0.005190515512563407,
"grad_norm": 1.078125,
"learning_rate": 7.883220906358917e-05,
"loss": 1.7708,
"num_input_tokens_seen": 1231552512,
"step": 1044
},
{
"epoch": 0.0053084817742125754,
"grad_norm": 0.89453125,
"learning_rate": 7.88284635161059e-05,
"loss": 1.7716,
"num_input_tokens_seen": 1232732160,
"step": 1045
},
{
"epoch": 0.005426448035861744,
"grad_norm": 0.97265625,
"learning_rate": 7.882471206081023e-05,
"loss": 1.7232,
"num_input_tokens_seen": 1233911808,
"step": 1046
},
{
"epoch": 0.005544414297510912,
"grad_norm": 1.0234375,
"learning_rate": 7.882095469827299e-05,
"loss": 2.0954,
"num_input_tokens_seen": 1235091456,
"step": 1047
},
{
"epoch": 0.00566238055916008,
"grad_norm": 0.94140625,
"learning_rate": 7.881719142906584e-05,
"loss": 1.7849,
"num_input_tokens_seen": 1236271104,
"step": 1048
},
{
"epoch": 0.005780346820809248,
"grad_norm": 0.94921875,
"learning_rate": 7.881342225376137e-05,
"loss": 1.9185,
"num_input_tokens_seen": 1237450752,
"step": 1049
},
{
"epoch": 0.0058983130824584165,
"grad_norm": 1.0234375,
"learning_rate": 7.880964717293307e-05,
"loss": 1.7487,
"num_input_tokens_seen": 1238630400,
"step": 1050
},
{
"epoch": 0.0060162793441075856,
"grad_norm": 0.921875,
"learning_rate": 7.880586618715532e-05,
"loss": 1.7824,
"num_input_tokens_seen": 1239810048,
"step": 1051
},
{
"epoch": 0.006134245605756754,
"grad_norm": 1.1171875,
"learning_rate": 7.88020792970034e-05,
"loss": 1.8,
"num_input_tokens_seen": 1240989696,
"step": 1052
},
{
"epoch": 0.006252211867405922,
"grad_norm": 0.9453125,
"learning_rate": 7.87982865030535e-05,
"loss": 1.8677,
"num_input_tokens_seen": 1242169344,
"step": 1053
},
{
"epoch": 0.00637017812905509,
"grad_norm": 1.0625,
"learning_rate": 7.879448780588271e-05,
"loss": 1.7398,
"num_input_tokens_seen": 1243348992,
"step": 1054
},
{
"epoch": 0.006488144390704258,
"grad_norm": 1.390625,
"learning_rate": 7.879068320606899e-05,
"loss": 1.7627,
"num_input_tokens_seen": 1244528640,
"step": 1055
},
{
"epoch": 0.006606110652353427,
"grad_norm": 1.0859375,
"learning_rate": 7.878687270419122e-05,
"loss": 1.854,
"num_input_tokens_seen": 1245708288,
"step": 1056
},
{
"epoch": 0.006724076914002596,
"grad_norm": 1.109375,
"learning_rate": 7.878305630082919e-05,
"loss": 1.7986,
"num_input_tokens_seen": 1246887936,
"step": 1057
},
{
"epoch": 0.006842043175651764,
"grad_norm": 1.078125,
"learning_rate": 7.877923399656354e-05,
"loss": 1.785,
"num_input_tokens_seen": 1248067584,
"step": 1058
},
{
"epoch": 0.006960009437300932,
"grad_norm": 0.94921875,
"learning_rate": 7.877540579197586e-05,
"loss": 1.9014,
"num_input_tokens_seen": 1249247232,
"step": 1059
},
{
"epoch": 0.0070779756989501,
"grad_norm": 1.109375,
"learning_rate": 7.877157168764863e-05,
"loss": 1.8105,
"num_input_tokens_seen": 1250426880,
"step": 1060
},
{
"epoch": 0.0071959419605992685,
"grad_norm": 0.99609375,
"learning_rate": 7.876773168416519e-05,
"loss": 1.801,
"num_input_tokens_seen": 1251606528,
"step": 1061
},
{
"epoch": 0.007313908222248437,
"grad_norm": 0.94140625,
"learning_rate": 7.87638857821098e-05,
"loss": 1.8477,
"num_input_tokens_seen": 1252786176,
"step": 1062
},
{
"epoch": 0.007431874483897605,
"grad_norm": 1.046875,
"learning_rate": 7.876003398206765e-05,
"loss": 1.7174,
"num_input_tokens_seen": 1253965824,
"step": 1063
},
{
"epoch": 0.007549840745546774,
"grad_norm": 0.9375,
"learning_rate": 7.875617628462477e-05,
"loss": 1.9367,
"num_input_tokens_seen": 1255145472,
"step": 1064
},
{
"epoch": 0.007667807007195942,
"grad_norm": 1.2265625,
"learning_rate": 7.875231269036813e-05,
"loss": 1.8491,
"num_input_tokens_seen": 1256325120,
"step": 1065
},
{
"epoch": 0.00778577326884511,
"grad_norm": 1.0625,
"learning_rate": 7.874844319988556e-05,
"loss": 1.8657,
"num_input_tokens_seen": 1257504768,
"step": 1066
},
{
"epoch": 0.007903739530494279,
"grad_norm": 0.94140625,
"learning_rate": 7.874456781376584e-05,
"loss": 1.8064,
"num_input_tokens_seen": 1258684416,
"step": 1067
},
{
"epoch": 0.008021705792143447,
"grad_norm": 1.09375,
"learning_rate": 7.874068653259861e-05,
"loss": 1.93,
"num_input_tokens_seen": 1259864064,
"step": 1068
},
{
"epoch": 0.008139672053792615,
"grad_norm": 0.96875,
"learning_rate": 7.87367993569744e-05,
"loss": 1.9201,
"num_input_tokens_seen": 1261043712,
"step": 1069
},
{
"epoch": 0.008257638315441783,
"grad_norm": 0.984375,
"learning_rate": 7.873290628748465e-05,
"loss": 1.826,
"num_input_tokens_seen": 1262223360,
"step": 1070
},
{
"epoch": 0.008375604577090951,
"grad_norm": 0.98828125,
"learning_rate": 7.87290073247217e-05,
"loss": 1.7763,
"num_input_tokens_seen": 1263403008,
"step": 1071
},
{
"epoch": 0.00849357083874012,
"grad_norm": 0.859375,
"learning_rate": 7.872510246927879e-05,
"loss": 1.6945,
"num_input_tokens_seen": 1264582656,
"step": 1072
},
{
"epoch": 0.008611537100389288,
"grad_norm": 1.046875,
"learning_rate": 7.872119172175006e-05,
"loss": 1.8033,
"num_input_tokens_seen": 1265762304,
"step": 1073
},
{
"epoch": 0.008729503362038458,
"grad_norm": 0.9453125,
"learning_rate": 7.871727508273049e-05,
"loss": 1.9424,
"num_input_tokens_seen": 1266941952,
"step": 1074
},
{
"epoch": 0.008847469623687626,
"grad_norm": 0.90625,
"learning_rate": 7.871335255281606e-05,
"loss": 1.7735,
"num_input_tokens_seen": 1268121600,
"step": 1075
},
{
"epoch": 0.008965435885336794,
"grad_norm": 0.9453125,
"learning_rate": 7.870942413260354e-05,
"loss": 1.8556,
"num_input_tokens_seen": 1269301248,
"step": 1076
},
{
"epoch": 0.009083402146985962,
"grad_norm": 0.92578125,
"learning_rate": 7.87054898226907e-05,
"loss": 1.8768,
"num_input_tokens_seen": 1270480896,
"step": 1077
},
{
"epoch": 0.00920136840863513,
"grad_norm": 1.0859375,
"learning_rate": 7.87015496236761e-05,
"loss": 1.7556,
"num_input_tokens_seen": 1271660544,
"step": 1078
},
{
"epoch": 0.009319334670284299,
"grad_norm": 0.9453125,
"learning_rate": 7.869760353615926e-05,
"loss": 1.7153,
"num_input_tokens_seen": 1272840192,
"step": 1079
},
{
"epoch": 0.009437300931933467,
"grad_norm": 0.9921875,
"learning_rate": 7.869365156074061e-05,
"loss": 1.7451,
"num_input_tokens_seen": 1274019840,
"step": 1080
},
{
"epoch": 0.009555267193582635,
"grad_norm": 0.99609375,
"learning_rate": 7.868969369802142e-05,
"loss": 1.732,
"num_input_tokens_seen": 1275199488,
"step": 1081
},
{
"epoch": 0.009673233455231803,
"grad_norm": 1.015625,
"learning_rate": 7.86857299486039e-05,
"loss": 1.8516,
"num_input_tokens_seen": 1276379136,
"step": 1082
},
{
"epoch": 0.009791199716880972,
"grad_norm": 1.15625,
"learning_rate": 7.868176031309115e-05,
"loss": 1.7259,
"num_input_tokens_seen": 1277558784,
"step": 1083
},
{
"epoch": 0.00990916597853014,
"grad_norm": 1.03125,
"learning_rate": 7.867778479208712e-05,
"loss": 1.8024,
"num_input_tokens_seen": 1278738432,
"step": 1084
},
{
"epoch": 0.010027132240179308,
"grad_norm": 1.0703125,
"learning_rate": 7.867380338619672e-05,
"loss": 1.7921,
"num_input_tokens_seen": 1279918080,
"step": 1085
},
{
"epoch": 0.010145098501828476,
"grad_norm": 1.0234375,
"learning_rate": 7.866981609602572e-05,
"loss": 1.7405,
"num_input_tokens_seen": 1281097728,
"step": 1086
},
{
"epoch": 0.010263064763477646,
"grad_norm": 0.92578125,
"learning_rate": 7.86658229221808e-05,
"loss": 1.8575,
"num_input_tokens_seen": 1282277376,
"step": 1087
},
{
"epoch": 0.010381031025126814,
"grad_norm": 1.046875,
"learning_rate": 7.866182386526954e-05,
"loss": 1.9094,
"num_input_tokens_seen": 1283457024,
"step": 1088
},
{
"epoch": 0.010498997286775983,
"grad_norm": 1.25,
"learning_rate": 7.865781892590036e-05,
"loss": 1.7406,
"num_input_tokens_seen": 1284636672,
"step": 1089
},
{
"epoch": 0.010616963548425151,
"grad_norm": 1.3125,
"learning_rate": 7.865380810468265e-05,
"loss": 1.6847,
"num_input_tokens_seen": 1285816320,
"step": 1090
},
{
"epoch": 0.010734929810074319,
"grad_norm": 1.0,
"learning_rate": 7.864979140222666e-05,
"loss": 1.7862,
"num_input_tokens_seen": 1286995968,
"step": 1091
},
{
"epoch": 0.010852896071723487,
"grad_norm": 0.98046875,
"learning_rate": 7.864576881914354e-05,
"loss": 1.7013,
"num_input_tokens_seen": 1288175616,
"step": 1092
},
{
"epoch": 0.010970862333372655,
"grad_norm": 0.91015625,
"learning_rate": 7.86417403560453e-05,
"loss": 1.735,
"num_input_tokens_seen": 1289355264,
"step": 1093
},
{
"epoch": 0.011088828595021824,
"grad_norm": 0.9609375,
"learning_rate": 7.863770601354492e-05,
"loss": 1.8639,
"num_input_tokens_seen": 1290534912,
"step": 1094
},
{
"epoch": 0.011206794856670992,
"grad_norm": 1.046875,
"learning_rate": 7.863366579225622e-05,
"loss": 1.8047,
"num_input_tokens_seen": 1291714560,
"step": 1095
},
{
"epoch": 0.01132476111832016,
"grad_norm": 0.8046875,
"learning_rate": 7.862961969279391e-05,
"loss": 1.7503,
"num_input_tokens_seen": 1292894208,
"step": 1096
},
{
"epoch": 0.011442727379969328,
"grad_norm": 1.1171875,
"learning_rate": 7.862556771577363e-05,
"loss": 1.7217,
"num_input_tokens_seen": 1294073856,
"step": 1097
},
{
"epoch": 0.011560693641618497,
"grad_norm": 1.078125,
"learning_rate": 7.862150986181187e-05,
"loss": 1.7873,
"num_input_tokens_seen": 1295253504,
"step": 1098
},
{
"epoch": 0.011678659903267665,
"grad_norm": 0.98828125,
"learning_rate": 7.861744613152609e-05,
"loss": 1.6499,
"num_input_tokens_seen": 1296433152,
"step": 1099
},
{
"epoch": 0.011796626164916833,
"grad_norm": 0.9375,
"learning_rate": 7.861337652553452e-05,
"loss": 1.9022,
"num_input_tokens_seen": 1297612800,
"step": 1100
},
{
"epoch": 0.011914592426566003,
"grad_norm": 1.1015625,
"learning_rate": 7.86093010444564e-05,
"loss": 1.7538,
"num_input_tokens_seen": 1298792448,
"step": 1101
},
{
"epoch": 0.012032558688215171,
"grad_norm": 0.9296875,
"learning_rate": 7.860521968891183e-05,
"loss": 1.6857,
"num_input_tokens_seen": 1299972096,
"step": 1102
},
{
"epoch": 0.01215052494986434,
"grad_norm": 0.95703125,
"learning_rate": 7.860113245952179e-05,
"loss": 1.7463,
"num_input_tokens_seen": 1301151744,
"step": 1103
},
{
"epoch": 0.012268491211513508,
"grad_norm": 0.96875,
"learning_rate": 7.859703935690812e-05,
"loss": 1.7936,
"num_input_tokens_seen": 1302331392,
"step": 1104
},
{
"epoch": 0.012386457473162676,
"grad_norm": 0.93359375,
"learning_rate": 7.859294038169364e-05,
"loss": 1.8009,
"num_input_tokens_seen": 1303511040,
"step": 1105
},
{
"epoch": 0.012504423734811844,
"grad_norm": 0.99609375,
"learning_rate": 7.858883553450199e-05,
"loss": 1.8159,
"num_input_tokens_seen": 1304690688,
"step": 1106
},
{
"epoch": 0.012622389996461012,
"grad_norm": 0.9765625,
"learning_rate": 7.858472481595775e-05,
"loss": 1.9189,
"num_input_tokens_seen": 1305870336,
"step": 1107
},
{
"epoch": 0.01274035625811018,
"grad_norm": 0.921875,
"learning_rate": 7.858060822668634e-05,
"loss": 1.7446,
"num_input_tokens_seen": 1307049984,
"step": 1108
},
{
"epoch": 0.012858322519759349,
"grad_norm": 0.953125,
"learning_rate": 7.857648576731412e-05,
"loss": 1.8123,
"num_input_tokens_seen": 1308229632,
"step": 1109
},
{
"epoch": 0.012976288781408517,
"grad_norm": 0.91796875,
"learning_rate": 7.857235743846834e-05,
"loss": 1.8646,
"num_input_tokens_seen": 1309409280,
"step": 1110
},
{
"epoch": 0.013094255043057685,
"grad_norm": 0.94140625,
"learning_rate": 7.856822324077713e-05,
"loss": 1.6761,
"num_input_tokens_seen": 1310588928,
"step": 1111
},
{
"epoch": 0.013212221304706853,
"grad_norm": 1.0703125,
"learning_rate": 7.856408317486951e-05,
"loss": 2.004,
"num_input_tokens_seen": 1311768576,
"step": 1112
},
{
"epoch": 0.013330187566356021,
"grad_norm": 1.0390625,
"learning_rate": 7.855993724137539e-05,
"loss": 1.7488,
"num_input_tokens_seen": 1312948224,
"step": 1113
},
{
"epoch": 0.013448153828005191,
"grad_norm": 1.078125,
"learning_rate": 7.855578544092559e-05,
"loss": 1.8885,
"num_input_tokens_seen": 1314127872,
"step": 1114
},
{
"epoch": 0.01356612008965436,
"grad_norm": 0.921875,
"learning_rate": 7.85516277741518e-05,
"loss": 1.9571,
"num_input_tokens_seen": 1315307520,
"step": 1115
},
{
"epoch": 0.013684086351303528,
"grad_norm": 1.046875,
"learning_rate": 7.854746424168664e-05,
"loss": 1.8484,
"num_input_tokens_seen": 1316487168,
"step": 1116
},
{
"epoch": 0.013802052612952696,
"grad_norm": 0.87109375,
"learning_rate": 7.854329484416358e-05,
"loss": 1.8106,
"num_input_tokens_seen": 1317666816,
"step": 1117
},
{
"epoch": 0.013920018874601864,
"grad_norm": 1.0390625,
"learning_rate": 7.853911958221699e-05,
"loss": 1.9163,
"num_input_tokens_seen": 1318846464,
"step": 1118
},
{
"epoch": 0.014037985136251032,
"grad_norm": 1.0390625,
"learning_rate": 7.853493845648218e-05,
"loss": 1.8216,
"num_input_tokens_seen": 1320026112,
"step": 1119
},
{
"epoch": 0.0141559513979002,
"grad_norm": 0.85546875,
"learning_rate": 7.853075146759527e-05,
"loss": 1.8019,
"num_input_tokens_seen": 1321205760,
"step": 1120
},
{
"epoch": 0.014273917659549369,
"grad_norm": 0.828125,
"learning_rate": 7.852655861619336e-05,
"loss": 1.7269,
"num_input_tokens_seen": 1322385408,
"step": 1121
},
{
"epoch": 0.014391883921198537,
"grad_norm": 0.9140625,
"learning_rate": 7.852235990291436e-05,
"loss": 1.7582,
"num_input_tokens_seen": 1323565056,
"step": 1122
},
{
"epoch": 0.014509850182847705,
"grad_norm": 0.92578125,
"learning_rate": 7.851815532839713e-05,
"loss": 1.7213,
"num_input_tokens_seen": 1324744704,
"step": 1123
},
{
"epoch": 0.014627816444496873,
"grad_norm": 0.90234375,
"learning_rate": 7.85139448932814e-05,
"loss": 1.703,
"num_input_tokens_seen": 1325924352,
"step": 1124
},
{
"epoch": 0.014745782706146042,
"grad_norm": 0.86328125,
"learning_rate": 7.85097285982078e-05,
"loss": 1.7028,
"num_input_tokens_seen": 1327104000,
"step": 1125
},
{
"epoch": 0.01486374896779521,
"grad_norm": 0.90625,
"learning_rate": 7.850550644381784e-05,
"loss": 1.7196,
"num_input_tokens_seen": 1328283648,
"step": 1126
},
{
"epoch": 0.014981715229444378,
"grad_norm": 1.0078125,
"learning_rate": 7.850127843075394e-05,
"loss": 1.6825,
"num_input_tokens_seen": 1329463296,
"step": 1127
},
{
"epoch": 0.015099681491093548,
"grad_norm": 1.4765625,
"learning_rate": 7.849704455965937e-05,
"loss": 1.7549,
"num_input_tokens_seen": 1330642944,
"step": 1128
},
{
"epoch": 0.015217647752742716,
"grad_norm": 1.078125,
"learning_rate": 7.849280483117834e-05,
"loss": 1.6149,
"num_input_tokens_seen": 1331822592,
"step": 1129
},
{
"epoch": 0.015335614014391884,
"grad_norm": 1.1484375,
"learning_rate": 7.848855924595594e-05,
"loss": 1.634,
"num_input_tokens_seen": 1333002240,
"step": 1130
},
{
"epoch": 0.015453580276041053,
"grad_norm": 1.0625,
"learning_rate": 7.848430780463814e-05,
"loss": 1.6845,
"num_input_tokens_seen": 1334181888,
"step": 1131
},
{
"epoch": 0.01557154653769022,
"grad_norm": 1.1640625,
"learning_rate": 7.848005050787178e-05,
"loss": 1.7053,
"num_input_tokens_seen": 1335361536,
"step": 1132
},
{
"epoch": 0.01568951279933939,
"grad_norm": 0.9921875,
"learning_rate": 7.847578735630464e-05,
"loss": 1.6666,
"num_input_tokens_seen": 1336541184,
"step": 1133
},
{
"epoch": 0.015807479060988557,
"grad_norm": 0.9921875,
"learning_rate": 7.847151835058534e-05,
"loss": 1.6833,
"num_input_tokens_seen": 1337720832,
"step": 1134
},
{
"epoch": 0.015925445322637725,
"grad_norm": 0.96875,
"learning_rate": 7.846724349136344e-05,
"loss": 1.587,
"num_input_tokens_seen": 1338900480,
"step": 1135
},
{
"epoch": 0.016043411584286894,
"grad_norm": 1.0703125,
"learning_rate": 7.846296277928937e-05,
"loss": 1.6853,
"num_input_tokens_seen": 1340080128,
"step": 1136
},
{
"epoch": 0.016161377845936062,
"grad_norm": 0.8515625,
"learning_rate": 7.845867621501442e-05,
"loss": 1.836,
"num_input_tokens_seen": 1341259776,
"step": 1137
},
{
"epoch": 0.01627934410758523,
"grad_norm": 1.0859375,
"learning_rate": 7.845438379919081e-05,
"loss": 1.7539,
"num_input_tokens_seen": 1342439424,
"step": 1138
},
{
"epoch": 0.0163973103692344,
"grad_norm": 0.98046875,
"learning_rate": 7.845008553247166e-05,
"loss": 1.749,
"num_input_tokens_seen": 1343619072,
"step": 1139
},
{
"epoch": 0.016515276630883566,
"grad_norm": 1.2734375,
"learning_rate": 7.844578141551092e-05,
"loss": 1.8204,
"num_input_tokens_seen": 1344798720,
"step": 1140
},
{
"epoch": 0.016633242892532735,
"grad_norm": 0.9921875,
"learning_rate": 7.844147144896349e-05,
"loss": 1.7323,
"num_input_tokens_seen": 1345978368,
"step": 1141
},
{
"epoch": 0.016751209154181903,
"grad_norm": 0.90234375,
"learning_rate": 7.843715563348515e-05,
"loss": 1.6755,
"num_input_tokens_seen": 1347158016,
"step": 1142
},
{
"epoch": 0.01686917541583107,
"grad_norm": 0.91796875,
"learning_rate": 7.843283396973253e-05,
"loss": 1.7822,
"num_input_tokens_seen": 1348337664,
"step": 1143
},
{
"epoch": 0.01698714167748024,
"grad_norm": 1.1015625,
"learning_rate": 7.842850645836319e-05,
"loss": 1.7529,
"num_input_tokens_seen": 1349517312,
"step": 1144
},
{
"epoch": 0.017105107939129408,
"grad_norm": 0.98046875,
"learning_rate": 7.842417310003557e-05,
"loss": 1.7434,
"num_input_tokens_seen": 1350696960,
"step": 1145
},
{
"epoch": 0.017223074200778576,
"grad_norm": 1.0390625,
"learning_rate": 7.8419833895409e-05,
"loss": 1.5889,
"num_input_tokens_seen": 1351876608,
"step": 1146
},
{
"epoch": 0.017341040462427744,
"grad_norm": 1.0,
"learning_rate": 7.84154888451437e-05,
"loss": 1.5818,
"num_input_tokens_seen": 1353056256,
"step": 1147
},
{
"epoch": 0.017459006724076916,
"grad_norm": 1.0,
"learning_rate": 7.841113794990076e-05,
"loss": 1.7138,
"num_input_tokens_seen": 1354235904,
"step": 1148
},
{
"epoch": 0.017576972985726084,
"grad_norm": 0.8984375,
"learning_rate": 7.840678121034219e-05,
"loss": 1.6106,
"num_input_tokens_seen": 1355415552,
"step": 1149
},
{
"epoch": 0.017694939247375252,
"grad_norm": 1.0546875,
"learning_rate": 7.840241862713087e-05,
"loss": 1.7624,
"num_input_tokens_seen": 1356595200,
"step": 1150
},
{
"epoch": 0.01781290550902442,
"grad_norm": 1.0625,
"learning_rate": 7.839805020093057e-05,
"loss": 1.6687,
"num_input_tokens_seen": 1357774848,
"step": 1151
},
{
"epoch": 0.01793087177067359,
"grad_norm": 1.109375,
"learning_rate": 7.839367593240596e-05,
"loss": 1.6976,
"num_input_tokens_seen": 1358954496,
"step": 1152
},
{
"epoch": 0.018048838032322757,
"grad_norm": 0.87890625,
"learning_rate": 7.838929582222259e-05,
"loss": 1.8581,
"num_input_tokens_seen": 1360134144,
"step": 1153
},
{
"epoch": 0.018166804293971925,
"grad_norm": 1.1640625,
"learning_rate": 7.83849098710469e-05,
"loss": 1.6766,
"num_input_tokens_seen": 1361313792,
"step": 1154
},
{
"epoch": 0.018284770555621093,
"grad_norm": 0.9140625,
"learning_rate": 7.838051807954621e-05,
"loss": 1.7232,
"num_input_tokens_seen": 1362493440,
"step": 1155
},
{
"epoch": 0.01840273681727026,
"grad_norm": 0.91796875,
"learning_rate": 7.837612044838877e-05,
"loss": 1.6636,
"num_input_tokens_seen": 1363673088,
"step": 1156
},
{
"epoch": 0.01852070307891943,
"grad_norm": 1.0078125,
"learning_rate": 7.837171697824363e-05,
"loss": 1.6333,
"num_input_tokens_seen": 1364852736,
"step": 1157
},
{
"epoch": 0.018638669340568598,
"grad_norm": 0.984375,
"learning_rate": 7.836730766978085e-05,
"loss": 1.8061,
"num_input_tokens_seen": 1366032384,
"step": 1158
},
{
"epoch": 0.018756635602217766,
"grad_norm": 1.03125,
"learning_rate": 7.836289252367125e-05,
"loss": 1.6617,
"num_input_tokens_seen": 1367212032,
"step": 1159
},
{
"epoch": 0.018874601863866934,
"grad_norm": 0.859375,
"learning_rate": 7.835847154058666e-05,
"loss": 1.8089,
"num_input_tokens_seen": 1368391680,
"step": 1160
},
{
"epoch": 0.018992568125516102,
"grad_norm": 0.92578125,
"learning_rate": 7.835404472119971e-05,
"loss": 1.697,
"num_input_tokens_seen": 1369571328,
"step": 1161
},
{
"epoch": 0.01911053438716527,
"grad_norm": 0.984375,
"learning_rate": 7.834961206618396e-05,
"loss": 1.6818,
"num_input_tokens_seen": 1370750976,
"step": 1162
},
{
"epoch": 0.01922850064881444,
"grad_norm": 0.86328125,
"learning_rate": 7.834517357621381e-05,
"loss": 1.6874,
"num_input_tokens_seen": 1371930624,
"step": 1163
},
{
"epoch": 0.019346466910463607,
"grad_norm": 0.9375,
"learning_rate": 7.834072925196463e-05,
"loss": 1.7766,
"num_input_tokens_seen": 1373110272,
"step": 1164
},
{
"epoch": 0.019464433172112775,
"grad_norm": 1.0,
"learning_rate": 7.83362790941126e-05,
"loss": 1.7272,
"num_input_tokens_seen": 1374289920,
"step": 1165
},
{
"epoch": 0.019582399433761943,
"grad_norm": 1.2734375,
"learning_rate": 7.833182310333483e-05,
"loss": 1.5872,
"num_input_tokens_seen": 1375469568,
"step": 1166
},
{
"epoch": 0.01970036569541111,
"grad_norm": 1.0234375,
"learning_rate": 7.83273612803093e-05,
"loss": 1.6183,
"num_input_tokens_seen": 1376649216,
"step": 1167
},
{
"epoch": 0.01981833195706028,
"grad_norm": 0.86328125,
"learning_rate": 7.832289362571489e-05,
"loss": 1.7423,
"num_input_tokens_seen": 1377828864,
"step": 1168
},
{
"epoch": 0.019936298218709448,
"grad_norm": 0.8515625,
"learning_rate": 7.831842014023137e-05,
"loss": 1.8817,
"num_input_tokens_seen": 1379008512,
"step": 1169
},
{
"epoch": 0.020054264480358616,
"grad_norm": 0.9765625,
"learning_rate": 7.831394082453938e-05,
"loss": 1.8874,
"num_input_tokens_seen": 1380188160,
"step": 1170
},
{
"epoch": 0.020172230742007784,
"grad_norm": 1.125,
"learning_rate": 7.830945567932043e-05,
"loss": 1.6719,
"num_input_tokens_seen": 1381367808,
"step": 1171
},
{
"epoch": 0.020290197003656953,
"grad_norm": 1.0859375,
"learning_rate": 7.830496470525699e-05,
"loss": 1.744,
"num_input_tokens_seen": 1382547456,
"step": 1172
},
{
"epoch": 0.02040816326530612,
"grad_norm": 0.98046875,
"learning_rate": 7.830046790303232e-05,
"loss": 1.6126,
"num_input_tokens_seen": 1383727104,
"step": 1173
},
{
"epoch": 0.020526129526955292,
"grad_norm": 1.1015625,
"learning_rate": 7.829596527333065e-05,
"loss": 1.7485,
"num_input_tokens_seen": 1384906752,
"step": 1174
},
{
"epoch": 0.02064409578860446,
"grad_norm": 1.0,
"learning_rate": 7.829145681683706e-05,
"loss": 1.8069,
"num_input_tokens_seen": 1386086400,
"step": 1175
},
{
"epoch": 0.02076206205025363,
"grad_norm": 1.1171875,
"learning_rate": 7.82869425342375e-05,
"loss": 1.837,
"num_input_tokens_seen": 1387266048,
"step": 1176
},
{
"epoch": 0.020880028311902797,
"grad_norm": 1.3984375,
"learning_rate": 7.828242242621884e-05,
"loss": 1.7099,
"num_input_tokens_seen": 1388445696,
"step": 1177
},
{
"epoch": 0.020997994573551965,
"grad_norm": 1.078125,
"learning_rate": 7.82778964934688e-05,
"loss": 1.7364,
"num_input_tokens_seen": 1389625344,
"step": 1178
},
{
"epoch": 0.021115960835201134,
"grad_norm": 1.46875,
"learning_rate": 7.827336473667604e-05,
"loss": 1.6631,
"num_input_tokens_seen": 1390804992,
"step": 1179
},
{
"epoch": 0.021233927096850302,
"grad_norm": 1.0703125,
"learning_rate": 7.826882715653005e-05,
"loss": 1.7466,
"num_input_tokens_seen": 1391984640,
"step": 1180
},
{
"epoch": 0.02135189335849947,
"grad_norm": 1.6953125,
"learning_rate": 7.826428375372125e-05,
"loss": 1.7107,
"num_input_tokens_seen": 1393164288,
"step": 1181
},
{
"epoch": 0.021469859620148638,
"grad_norm": 1.296875,
"learning_rate": 7.825973452894091e-05,
"loss": 1.6893,
"num_input_tokens_seen": 1394343936,
"step": 1182
},
{
"epoch": 0.021587825881797806,
"grad_norm": 1.2890625,
"learning_rate": 7.82551794828812e-05,
"loss": 1.8002,
"num_input_tokens_seen": 1395523584,
"step": 1183
},
{
"epoch": 0.021705792143446975,
"grad_norm": 1.0625,
"learning_rate": 7.82506186162352e-05,
"loss": 1.7513,
"num_input_tokens_seen": 1396703232,
"step": 1184
},
{
"epoch": 0.021823758405096143,
"grad_norm": 1.09375,
"learning_rate": 7.824605192969681e-05,
"loss": 1.8065,
"num_input_tokens_seen": 1397882880,
"step": 1185
},
{
"epoch": 0.02194172466674531,
"grad_norm": 1.078125,
"learning_rate": 7.824147942396091e-05,
"loss": 1.7148,
"num_input_tokens_seen": 1399062528,
"step": 1186
},
{
"epoch": 0.02205969092839448,
"grad_norm": 0.921875,
"learning_rate": 7.823690109972317e-05,
"loss": 1.7104,
"num_input_tokens_seen": 1400242176,
"step": 1187
},
{
"epoch": 0.022177657190043647,
"grad_norm": 1.0859375,
"learning_rate": 7.823231695768023e-05,
"loss": 1.7102,
"num_input_tokens_seen": 1401421824,
"step": 1188
},
{
"epoch": 0.022295623451692816,
"grad_norm": 0.83984375,
"learning_rate": 7.822772699852954e-05,
"loss": 1.7314,
"num_input_tokens_seen": 1402601472,
"step": 1189
},
{
"epoch": 0.022413589713341984,
"grad_norm": 0.99609375,
"learning_rate": 7.822313122296947e-05,
"loss": 1.6595,
"num_input_tokens_seen": 1403781120,
"step": 1190
},
{
"epoch": 0.022531555974991152,
"grad_norm": 1.046875,
"learning_rate": 7.821852963169931e-05,
"loss": 1.789,
"num_input_tokens_seen": 1404960768,
"step": 1191
},
{
"epoch": 0.02264952223664032,
"grad_norm": 0.98046875,
"learning_rate": 7.821392222541918e-05,
"loss": 1.8047,
"num_input_tokens_seen": 1406140416,
"step": 1192
},
{
"epoch": 0.02276748849828949,
"grad_norm": 0.875,
"learning_rate": 7.820930900483009e-05,
"loss": 1.7661,
"num_input_tokens_seen": 1407320064,
"step": 1193
},
{
"epoch": 0.022885454759938657,
"grad_norm": 0.921875,
"learning_rate": 7.820468997063396e-05,
"loss": 1.7989,
"num_input_tokens_seen": 1408499712,
"step": 1194
},
{
"epoch": 0.023003421021587825,
"grad_norm": 1.03125,
"learning_rate": 7.820006512353358e-05,
"loss": 1.6524,
"num_input_tokens_seen": 1409679360,
"step": 1195
},
{
"epoch": 0.023121387283236993,
"grad_norm": 0.8671875,
"learning_rate": 7.819543446423264e-05,
"loss": 1.7819,
"num_input_tokens_seen": 1410859008,
"step": 1196
},
{
"epoch": 0.02323935354488616,
"grad_norm": 0.88671875,
"learning_rate": 7.819079799343568e-05,
"loss": 1.723,
"num_input_tokens_seen": 1412038656,
"step": 1197
},
{
"epoch": 0.02335731980653533,
"grad_norm": 0.8203125,
"learning_rate": 7.818615571184816e-05,
"loss": 1.7317,
"num_input_tokens_seen": 1413218304,
"step": 1198
},
{
"epoch": 0.023475286068184498,
"grad_norm": 0.87109375,
"learning_rate": 7.818150762017642e-05,
"loss": 1.7398,
"num_input_tokens_seen": 1414397952,
"step": 1199
},
{
"epoch": 0.023593252329833666,
"grad_norm": 0.94921875,
"learning_rate": 7.817685371912765e-05,
"loss": 1.7017,
"num_input_tokens_seen": 1415577600,
"step": 1200
},
{
"epoch": 0.023593252329833666,
"eval_wikipedia_loss": 2.2694852352142334,
"eval_wikipedia_runtime": 161.0095,
"eval_wikipedia_samples_per_second": 4.36,
"eval_wikipedia_steps_per_second": 0.186,
"num_input_tokens_seen": 1415577600,
"step": 1200
},
{
"epoch": 0.023593252329833666,
"eval_toxicity_loss": 4.077282905578613,
"eval_toxicity_runtime": 1.0499,
"eval_toxicity_samples_per_second": 1.905,
"eval_toxicity_steps_per_second": 0.952,
"num_input_tokens_seen": 1415577600,
"step": 1200
},
{
"epoch": 0.023711218591482838,
"grad_norm": 0.86328125,
"learning_rate": 7.817219400940997e-05,
"loss": 1.7017,
"num_input_tokens_seen": 1416757248,
"step": 1201
},
{
"epoch": 0.023829184853132006,
"grad_norm": 0.88671875,
"learning_rate": 7.816752849173236e-05,
"loss": 1.9035,
"num_input_tokens_seen": 1417936896,
"step": 1202
},
{
"epoch": 0.023947151114781174,
"grad_norm": 0.92578125,
"learning_rate": 7.816285716680469e-05,
"loss": 1.6969,
"num_input_tokens_seen": 1419116544,
"step": 1203
},
{
"epoch": 0.024065117376430342,
"grad_norm": 0.9609375,
"learning_rate": 7.815818003533769e-05,
"loss": 1.857,
"num_input_tokens_seen": 1420296192,
"step": 1204
},
{
"epoch": 0.02418308363807951,
"grad_norm": 0.84765625,
"learning_rate": 7.815349709804301e-05,
"loss": 1.794,
"num_input_tokens_seen": 1421475840,
"step": 1205
},
{
"epoch": 0.02430104989972868,
"grad_norm": 0.953125,
"learning_rate": 7.814880835563315e-05,
"loss": 1.7131,
"num_input_tokens_seen": 1422655488,
"step": 1206
},
{
"epoch": 0.024419016161377847,
"grad_norm": 0.828125,
"learning_rate": 7.814411380882151e-05,
"loss": 1.7848,
"num_input_tokens_seen": 1423835136,
"step": 1207
},
{
"epoch": 0.024536982423027015,
"grad_norm": 0.97265625,
"learning_rate": 7.813941345832242e-05,
"loss": 1.7348,
"num_input_tokens_seen": 1425014784,
"step": 1208
},
{
"epoch": 0.024654948684676183,
"grad_norm": 1.1171875,
"learning_rate": 7.813470730485098e-05,
"loss": 1.6488,
"num_input_tokens_seen": 1426194432,
"step": 1209
},
{
"epoch": 0.02477291494632535,
"grad_norm": 0.93359375,
"learning_rate": 7.812999534912326e-05,
"loss": 1.7997,
"num_input_tokens_seen": 1427374080,
"step": 1210
},
{
"epoch": 0.02489088120797452,
"grad_norm": 0.796875,
"learning_rate": 7.812527759185621e-05,
"loss": 1.7712,
"num_input_tokens_seen": 1428553728,
"step": 1211
},
{
"epoch": 0.025008847469623688,
"grad_norm": 1.078125,
"learning_rate": 7.812055403376764e-05,
"loss": 1.7391,
"num_input_tokens_seen": 1429733376,
"step": 1212
},
{
"epoch": 0.025126813731272856,
"grad_norm": 0.8984375,
"learning_rate": 7.811582467557623e-05,
"loss": 1.737,
"num_input_tokens_seen": 1430913024,
"step": 1213
},
{
"epoch": 0.025244779992922024,
"grad_norm": 0.94140625,
"learning_rate": 7.811108951800157e-05,
"loss": 1.7396,
"num_input_tokens_seen": 1432092672,
"step": 1214
},
{
"epoch": 0.025362746254571193,
"grad_norm": 1.109375,
"learning_rate": 7.810634856176413e-05,
"loss": 1.7338,
"num_input_tokens_seen": 1433272320,
"step": 1215
},
{
"epoch": 0.02548071251622036,
"grad_norm": 0.9375,
"learning_rate": 7.810160180758523e-05,
"loss": 1.6398,
"num_input_tokens_seen": 1434451968,
"step": 1216
},
{
"epoch": 0.02559867877786953,
"grad_norm": 0.96484375,
"learning_rate": 7.809684925618712e-05,
"loss": 1.6268,
"num_input_tokens_seen": 1435631616,
"step": 1217
},
{
"epoch": 0.025716645039518697,
"grad_norm": 1.1328125,
"learning_rate": 7.80920909082929e-05,
"loss": 1.6998,
"num_input_tokens_seen": 1436811264,
"step": 1218
},
{
"epoch": 0.025834611301167865,
"grad_norm": 0.91015625,
"learning_rate": 7.808732676462655e-05,
"loss": 1.8721,
"num_input_tokens_seen": 1437990912,
"step": 1219
},
{
"epoch": 0.025952577562817034,
"grad_norm": 0.8671875,
"learning_rate": 7.808255682591295e-05,
"loss": 1.7911,
"num_input_tokens_seen": 1439170560,
"step": 1220
},
{
"epoch": 0.026070543824466202,
"grad_norm": 1.1328125,
"learning_rate": 7.807778109287787e-05,
"loss": 1.6226,
"num_input_tokens_seen": 1440350208,
"step": 1221
},
{
"epoch": 0.02618851008611537,
"grad_norm": 0.9140625,
"learning_rate": 7.807299956624792e-05,
"loss": 1.6958,
"num_input_tokens_seen": 1441529856,
"step": 1222
},
{
"epoch": 0.026306476347764538,
"grad_norm": 0.9765625,
"learning_rate": 7.806821224675062e-05,
"loss": 1.7473,
"num_input_tokens_seen": 1442709504,
"step": 1223
},
{
"epoch": 0.026424442609413706,
"grad_norm": 0.90234375,
"learning_rate": 7.80634191351144e-05,
"loss": 1.6486,
"num_input_tokens_seen": 1443889152,
"step": 1224
},
{
"epoch": 0.026542408871062875,
"grad_norm": 0.921875,
"learning_rate": 7.80586202320685e-05,
"loss": 1.7505,
"num_input_tokens_seen": 1445068800,
"step": 1225
},
{
"epoch": 0.026660375132712043,
"grad_norm": 0.94140625,
"learning_rate": 7.80538155383431e-05,
"loss": 1.7118,
"num_input_tokens_seen": 1446248448,
"step": 1226
},
{
"epoch": 0.02677834139436121,
"grad_norm": 0.91796875,
"learning_rate": 7.804900505466923e-05,
"loss": 1.8553,
"num_input_tokens_seen": 1447428096,
"step": 1227
},
{
"epoch": 0.026896307656010383,
"grad_norm": 0.9375,
"learning_rate": 7.804418878177882e-05,
"loss": 1.7392,
"num_input_tokens_seen": 1448607744,
"step": 1228
},
{
"epoch": 0.02701427391765955,
"grad_norm": 0.90625,
"learning_rate": 7.803936672040468e-05,
"loss": 1.7218,
"num_input_tokens_seen": 1449787392,
"step": 1229
},
{
"epoch": 0.02713224017930872,
"grad_norm": 1.046875,
"learning_rate": 7.803453887128049e-05,
"loss": 1.5923,
"num_input_tokens_seen": 1450967040,
"step": 1230
},
{
"epoch": 0.027250206440957887,
"grad_norm": 0.83984375,
"learning_rate": 7.802970523514081e-05,
"loss": 1.6976,
"num_input_tokens_seen": 1452146688,
"step": 1231
},
{
"epoch": 0.027368172702607056,
"grad_norm": 1.0078125,
"learning_rate": 7.802486581272109e-05,
"loss": 1.7327,
"num_input_tokens_seen": 1453326336,
"step": 1232
},
{
"epoch": 0.027486138964256224,
"grad_norm": 0.78515625,
"learning_rate": 7.802002060475765e-05,
"loss": 1.6797,
"num_input_tokens_seen": 1454505984,
"step": 1233
},
{
"epoch": 0.027604105225905392,
"grad_norm": 1.1015625,
"learning_rate": 7.801516961198771e-05,
"loss": 1.6193,
"num_input_tokens_seen": 1455685632,
"step": 1234
},
{
"epoch": 0.02772207148755456,
"grad_norm": 0.953125,
"learning_rate": 7.801031283514933e-05,
"loss": 1.6832,
"num_input_tokens_seen": 1456865280,
"step": 1235
},
{
"epoch": 0.02784003774920373,
"grad_norm": 1.1484375,
"learning_rate": 7.800545027498151e-05,
"loss": 1.8275,
"num_input_tokens_seen": 1458044928,
"step": 1236
},
{
"epoch": 0.027958004010852897,
"grad_norm": 0.984375,
"learning_rate": 7.800058193222407e-05,
"loss": 1.808,
"num_input_tokens_seen": 1459224576,
"step": 1237
},
{
"epoch": 0.028075970272502065,
"grad_norm": 1.2578125,
"learning_rate": 7.799570780761775e-05,
"loss": 1.6494,
"num_input_tokens_seen": 1460404224,
"step": 1238
},
{
"epoch": 0.028193936534151233,
"grad_norm": 0.96875,
"learning_rate": 7.799082790190414e-05,
"loss": 1.6716,
"num_input_tokens_seen": 1461583872,
"step": 1239
},
{
"epoch": 0.0283119027958004,
"grad_norm": 1.125,
"learning_rate": 7.798594221582575e-05,
"loss": 1.7444,
"num_input_tokens_seen": 1462763520,
"step": 1240
},
{
"epoch": 0.02842986905744957,
"grad_norm": 1.015625,
"learning_rate": 7.798105075012591e-05,
"loss": 1.6863,
"num_input_tokens_seen": 1463943168,
"step": 1241
},
{
"epoch": 0.028547835319098738,
"grad_norm": 0.87109375,
"learning_rate": 7.797615350554892e-05,
"loss": 1.6711,
"num_input_tokens_seen": 1465122816,
"step": 1242
},
{
"epoch": 0.028665801580747906,
"grad_norm": 1.0703125,
"learning_rate": 7.797125048283985e-05,
"loss": 1.612,
"num_input_tokens_seen": 1466302464,
"step": 1243
},
{
"epoch": 0.028783767842397074,
"grad_norm": 0.9609375,
"learning_rate": 7.796634168274473e-05,
"loss": 1.6763,
"num_input_tokens_seen": 1467482112,
"step": 1244
},
{
"epoch": 0.028901734104046242,
"grad_norm": 1.03125,
"learning_rate": 7.796142710601041e-05,
"loss": 1.6243,
"num_input_tokens_seen": 1468661760,
"step": 1245
},
{
"epoch": 0.02901970036569541,
"grad_norm": 1.0,
"learning_rate": 7.795650675338471e-05,
"loss": 1.6789,
"num_input_tokens_seen": 1469841408,
"step": 1246
},
{
"epoch": 0.02913766662734458,
"grad_norm": 1.125,
"learning_rate": 7.795158062561622e-05,
"loss": 1.5523,
"num_input_tokens_seen": 1471021056,
"step": 1247
},
{
"epoch": 0.029255632888993747,
"grad_norm": 1.2578125,
"learning_rate": 7.794664872345446e-05,
"loss": 1.5118,
"num_input_tokens_seen": 1472200704,
"step": 1248
},
{
"epoch": 0.029373599150642915,
"grad_norm": 1.0859375,
"learning_rate": 7.794171104764984e-05,
"loss": 1.6745,
"num_input_tokens_seen": 1473380352,
"step": 1249
},
{
"epoch": 0.029491565412292083,
"grad_norm": 1.0390625,
"learning_rate": 7.793676759895363e-05,
"loss": 1.7021,
"num_input_tokens_seen": 1474560000,
"step": 1250
},
{
"epoch": 0.02960953167394125,
"grad_norm": 0.921875,
"learning_rate": 7.793181837811801e-05,
"loss": 1.7004,
"num_input_tokens_seen": 1475739648,
"step": 1251
},
{
"epoch": 0.02972749793559042,
"grad_norm": 1.1640625,
"learning_rate": 7.792686338589598e-05,
"loss": 1.6231,
"num_input_tokens_seen": 1476919296,
"step": 1252
},
{
"epoch": 0.029845464197239588,
"grad_norm": 0.80859375,
"learning_rate": 7.792190262304146e-05,
"loss": 1.7307,
"num_input_tokens_seen": 1478098944,
"step": 1253
},
{
"epoch": 0.029963430458888756,
"grad_norm": 1.25,
"learning_rate": 7.791693609030922e-05,
"loss": 1.4952,
"num_input_tokens_seen": 1479278592,
"step": 1254
},
{
"epoch": 0.030081396720537928,
"grad_norm": 1.046875,
"learning_rate": 7.791196378845495e-05,
"loss": 1.6335,
"num_input_tokens_seen": 1480458240,
"step": 1255
},
{
"epoch": 0.030199362982187096,
"grad_norm": 1.140625,
"learning_rate": 7.79069857182352e-05,
"loss": 1.7717,
"num_input_tokens_seen": 1481637888,
"step": 1256
},
{
"epoch": 0.030317329243836264,
"grad_norm": 1.078125,
"learning_rate": 7.790200188040737e-05,
"loss": 1.6776,
"num_input_tokens_seen": 1482817536,
"step": 1257
},
{
"epoch": 0.030435295505485432,
"grad_norm": 0.98828125,
"learning_rate": 7.789701227572977e-05,
"loss": 1.8233,
"num_input_tokens_seen": 1483997184,
"step": 1258
},
{
"epoch": 0.0305532617671346,
"grad_norm": 1.0390625,
"learning_rate": 7.789201690496157e-05,
"loss": 1.563,
"num_input_tokens_seen": 1485176832,
"step": 1259
},
{
"epoch": 0.03067122802878377,
"grad_norm": 1.0859375,
"learning_rate": 7.788701576886283e-05,
"loss": 1.6999,
"num_input_tokens_seen": 1486356480,
"step": 1260
},
{
"epoch": 0.030789194290432937,
"grad_norm": 1.0,
"learning_rate": 7.788200886819447e-05,
"loss": 1.627,
"num_input_tokens_seen": 1487536128,
"step": 1261
},
{
"epoch": 0.030907160552082105,
"grad_norm": 0.9296875,
"learning_rate": 7.787699620371831e-05,
"loss": 1.7276,
"num_input_tokens_seen": 1488715776,
"step": 1262
},
{
"epoch": 0.031025126813731273,
"grad_norm": 1.0625,
"learning_rate": 7.787197777619703e-05,
"loss": 1.6639,
"num_input_tokens_seen": 1489895424,
"step": 1263
},
{
"epoch": 0.03114309307538044,
"grad_norm": 1.1015625,
"learning_rate": 7.786695358639419e-05,
"loss": 1.7681,
"num_input_tokens_seen": 1491075072,
"step": 1264
},
{
"epoch": 0.031261059337029606,
"grad_norm": 1.046875,
"learning_rate": 7.786192363507423e-05,
"loss": 1.6764,
"num_input_tokens_seen": 1492254720,
"step": 1265
},
{
"epoch": 0.03137902559867878,
"grad_norm": 1.0,
"learning_rate": 7.785688792300247e-05,
"loss": 1.626,
"num_input_tokens_seen": 1493434368,
"step": 1266
},
{
"epoch": 0.03149699186032794,
"grad_norm": 0.94140625,
"learning_rate": 7.785184645094509e-05,
"loss": 1.6718,
"num_input_tokens_seen": 1494614016,
"step": 1267
},
{
"epoch": 0.031614958121977114,
"grad_norm": 0.96875,
"learning_rate": 7.784679921966916e-05,
"loss": 1.6437,
"num_input_tokens_seen": 1495793664,
"step": 1268
},
{
"epoch": 0.031732924383626286,
"grad_norm": 0.9921875,
"learning_rate": 7.784174622994265e-05,
"loss": 1.6631,
"num_input_tokens_seen": 1496973312,
"step": 1269
},
{
"epoch": 0.03185089064527545,
"grad_norm": 0.97265625,
"learning_rate": 7.783668748253435e-05,
"loss": 1.6691,
"num_input_tokens_seen": 1498152960,
"step": 1270
},
{
"epoch": 0.03196885690692462,
"grad_norm": 0.8984375,
"learning_rate": 7.783162297821396e-05,
"loss": 1.7088,
"num_input_tokens_seen": 1499332608,
"step": 1271
},
{
"epoch": 0.03208682316857379,
"grad_norm": 1.1015625,
"learning_rate": 7.782655271775206e-05,
"loss": 1.6545,
"num_input_tokens_seen": 1500512256,
"step": 1272
},
{
"epoch": 0.03220478943022296,
"grad_norm": 0.7890625,
"learning_rate": 7.78214767019201e-05,
"loss": 1.701,
"num_input_tokens_seen": 1501691904,
"step": 1273
},
{
"epoch": 0.032322755691872124,
"grad_norm": 1.03125,
"learning_rate": 7.781639493149041e-05,
"loss": 1.5984,
"num_input_tokens_seen": 1502871552,
"step": 1274
},
{
"epoch": 0.032440721953521295,
"grad_norm": 0.90234375,
"learning_rate": 7.781130740723616e-05,
"loss": 1.7417,
"num_input_tokens_seen": 1504051200,
"step": 1275
},
{
"epoch": 0.03255868821517046,
"grad_norm": 1.125,
"learning_rate": 7.780621412993146e-05,
"loss": 1.6722,
"num_input_tokens_seen": 1505230848,
"step": 1276
},
{
"epoch": 0.03267665447681963,
"grad_norm": 0.99609375,
"learning_rate": 7.780111510035124e-05,
"loss": 1.6619,
"num_input_tokens_seen": 1506410496,
"step": 1277
},
{
"epoch": 0.0327946207384688,
"grad_norm": 0.86328125,
"learning_rate": 7.779601031927133e-05,
"loss": 1.7004,
"num_input_tokens_seen": 1507590144,
"step": 1278
},
{
"epoch": 0.03291258700011797,
"grad_norm": 1.03125,
"learning_rate": 7.779089978746844e-05,
"loss": 1.5487,
"num_input_tokens_seen": 1508769792,
"step": 1279
},
{
"epoch": 0.03303055326176713,
"grad_norm": 0.9765625,
"learning_rate": 7.778578350572012e-05,
"loss": 1.5906,
"num_input_tokens_seen": 1509949440,
"step": 1280
},
{
"epoch": 0.033148519523416305,
"grad_norm": 0.91796875,
"learning_rate": 7.778066147480484e-05,
"loss": 1.7207,
"num_input_tokens_seen": 1511129088,
"step": 1281
},
{
"epoch": 0.03326648578506547,
"grad_norm": 1.0078125,
"learning_rate": 7.777553369550193e-05,
"loss": 1.7121,
"num_input_tokens_seen": 1512308736,
"step": 1282
},
{
"epoch": 0.03338445204671464,
"grad_norm": 1.015625,
"learning_rate": 7.777040016859158e-05,
"loss": 1.6065,
"num_input_tokens_seen": 1513488384,
"step": 1283
},
{
"epoch": 0.033502418308363806,
"grad_norm": 1.0234375,
"learning_rate": 7.776526089485487e-05,
"loss": 1.5206,
"num_input_tokens_seen": 1514668032,
"step": 1284
},
{
"epoch": 0.03362038457001298,
"grad_norm": 0.9609375,
"learning_rate": 7.776011587507374e-05,
"loss": 1.6641,
"num_input_tokens_seen": 1515847680,
"step": 1285
},
{
"epoch": 0.03373835083166214,
"grad_norm": 1.03125,
"learning_rate": 7.775496511003101e-05,
"loss": 1.6269,
"num_input_tokens_seen": 1517027328,
"step": 1286
},
{
"epoch": 0.033856317093311314,
"grad_norm": 1.0546875,
"learning_rate": 7.774980860051039e-05,
"loss": 1.7265,
"num_input_tokens_seen": 1518206976,
"step": 1287
},
{
"epoch": 0.03397428335496048,
"grad_norm": 0.90234375,
"learning_rate": 7.774464634729645e-05,
"loss": 1.6951,
"num_input_tokens_seen": 1519386624,
"step": 1288
},
{
"epoch": 0.03409224961660965,
"grad_norm": 0.92578125,
"learning_rate": 7.773947835117464e-05,
"loss": 1.6725,
"num_input_tokens_seen": 1520566272,
"step": 1289
},
{
"epoch": 0.034210215878258815,
"grad_norm": 1.015625,
"learning_rate": 7.773430461293124e-05,
"loss": 1.6742,
"num_input_tokens_seen": 1521745920,
"step": 1290
},
{
"epoch": 0.03432818213990799,
"grad_norm": 1.0,
"learning_rate": 7.77291251333535e-05,
"loss": 1.7581,
"num_input_tokens_seen": 1522925568,
"step": 1291
},
{
"epoch": 0.03444614840155715,
"grad_norm": 1.0703125,
"learning_rate": 7.772393991322946e-05,
"loss": 1.668,
"num_input_tokens_seen": 1524105216,
"step": 1292
},
{
"epoch": 0.03456411466320632,
"grad_norm": 1.1875,
"learning_rate": 7.771874895334805e-05,
"loss": 1.5293,
"num_input_tokens_seen": 1525284864,
"step": 1293
},
{
"epoch": 0.03468208092485549,
"grad_norm": 0.8828125,
"learning_rate": 7.771355225449908e-05,
"loss": 1.6373,
"num_input_tokens_seen": 1526464512,
"step": 1294
},
{
"epoch": 0.03480004718650466,
"grad_norm": 0.92578125,
"learning_rate": 7.770834981747326e-05,
"loss": 1.6129,
"num_input_tokens_seen": 1527644160,
"step": 1295
},
{
"epoch": 0.03491801344815383,
"grad_norm": 1.1875,
"learning_rate": 7.770314164306213e-05,
"loss": 1.693,
"num_input_tokens_seen": 1528823808,
"step": 1296
},
{
"epoch": 0.035035979709802996,
"grad_norm": 0.859375,
"learning_rate": 7.769792773205813e-05,
"loss": 1.7876,
"num_input_tokens_seen": 1530003456,
"step": 1297
},
{
"epoch": 0.03515394597145217,
"grad_norm": 0.953125,
"learning_rate": 7.769270808525456e-05,
"loss": 1.7498,
"num_input_tokens_seen": 1531183104,
"step": 1298
},
{
"epoch": 0.03527191223310133,
"grad_norm": 0.88671875,
"learning_rate": 7.76874827034456e-05,
"loss": 1.7111,
"num_input_tokens_seen": 1532362752,
"step": 1299
},
{
"epoch": 0.035389878494750504,
"grad_norm": 0.875,
"learning_rate": 7.76822515874263e-05,
"loss": 1.8188,
"num_input_tokens_seen": 1533542400,
"step": 1300
},
{
"epoch": 0.03550784475639967,
"grad_norm": 0.86328125,
"learning_rate": 7.767701473799259e-05,
"loss": 1.6833,
"num_input_tokens_seen": 1534722048,
"step": 1301
},
{
"epoch": 0.03562581101804884,
"grad_norm": 0.95703125,
"learning_rate": 7.767177215594125e-05,
"loss": 1.6609,
"num_input_tokens_seen": 1535901696,
"step": 1302
},
{
"epoch": 0.035743777279698005,
"grad_norm": 0.8671875,
"learning_rate": 7.766652384206994e-05,
"loss": 1.7225,
"num_input_tokens_seen": 1537081344,
"step": 1303
},
{
"epoch": 0.03586174354134718,
"grad_norm": 0.9375,
"learning_rate": 7.766126979717725e-05,
"loss": 1.6236,
"num_input_tokens_seen": 1538260992,
"step": 1304
},
{
"epoch": 0.03597970980299634,
"grad_norm": 0.8828125,
"learning_rate": 7.765601002206253e-05,
"loss": 1.6974,
"num_input_tokens_seen": 1539440640,
"step": 1305
},
{
"epoch": 0.03609767606464551,
"grad_norm": 0.96484375,
"learning_rate": 7.76507445175261e-05,
"loss": 1.8008,
"num_input_tokens_seen": 1540620288,
"step": 1306
},
{
"epoch": 0.03621564232629468,
"grad_norm": 1.0390625,
"learning_rate": 7.764547328436909e-05,
"loss": 1.5457,
"num_input_tokens_seen": 1541799936,
"step": 1307
},
{
"epoch": 0.03633360858794385,
"grad_norm": 1.0078125,
"learning_rate": 7.764019632339355e-05,
"loss": 1.6641,
"num_input_tokens_seen": 1542979584,
"step": 1308
},
{
"epoch": 0.036451574849593014,
"grad_norm": 1.1796875,
"learning_rate": 7.763491363540237e-05,
"loss": 1.6156,
"num_input_tokens_seen": 1544159232,
"step": 1309
},
{
"epoch": 0.036569541111242186,
"grad_norm": 1.1015625,
"learning_rate": 7.762962522119931e-05,
"loss": 1.5664,
"num_input_tokens_seen": 1545338880,
"step": 1310
},
{
"epoch": 0.03668750737289135,
"grad_norm": 1.0,
"learning_rate": 7.762433108158903e-05,
"loss": 1.7106,
"num_input_tokens_seen": 1546518528,
"step": 1311
},
{
"epoch": 0.03680547363454052,
"grad_norm": 1.046875,
"learning_rate": 7.761903121737702e-05,
"loss": 1.7238,
"num_input_tokens_seen": 1547698176,
"step": 1312
},
{
"epoch": 0.03692343989618969,
"grad_norm": 0.94140625,
"learning_rate": 7.761372562936966e-05,
"loss": 1.6679,
"num_input_tokens_seen": 1548877824,
"step": 1313
},
{
"epoch": 0.03704140615783886,
"grad_norm": 1.234375,
"learning_rate": 7.760841431837424e-05,
"loss": 1.5861,
"num_input_tokens_seen": 1550057472,
"step": 1314
},
{
"epoch": 0.037159372419488024,
"grad_norm": 1.0546875,
"learning_rate": 7.760309728519884e-05,
"loss": 1.5827,
"num_input_tokens_seen": 1551237120,
"step": 1315
},
{
"epoch": 0.037277338681137195,
"grad_norm": 1.0,
"learning_rate": 7.75977745306525e-05,
"loss": 1.6241,
"num_input_tokens_seen": 1552416768,
"step": 1316
},
{
"epoch": 0.03739530494278636,
"grad_norm": 0.94921875,
"learning_rate": 7.759244605554504e-05,
"loss": 1.5647,
"num_input_tokens_seen": 1553596416,
"step": 1317
},
{
"epoch": 0.03751327120443553,
"grad_norm": 1.0390625,
"learning_rate": 7.758711186068723e-05,
"loss": 1.6642,
"num_input_tokens_seen": 1554776064,
"step": 1318
},
{
"epoch": 0.0376312374660847,
"grad_norm": 1.0,
"learning_rate": 7.758177194689065e-05,
"loss": 1.6156,
"num_input_tokens_seen": 1555955712,
"step": 1319
},
{
"epoch": 0.03774920372773387,
"grad_norm": 0.98046875,
"learning_rate": 7.75764263149678e-05,
"loss": 1.6103,
"num_input_tokens_seen": 1557135360,
"step": 1320
},
{
"epoch": 0.03786716998938304,
"grad_norm": 0.98828125,
"learning_rate": 7.757107496573202e-05,
"loss": 1.6254,
"num_input_tokens_seen": 1558315008,
"step": 1321
},
{
"epoch": 0.037985136251032205,
"grad_norm": 0.96484375,
"learning_rate": 7.756571789999751e-05,
"loss": 1.6553,
"num_input_tokens_seen": 1559494656,
"step": 1322
},
{
"epoch": 0.038103102512681376,
"grad_norm": 1.046875,
"learning_rate": 7.756035511857937e-05,
"loss": 1.7126,
"num_input_tokens_seen": 1560674304,
"step": 1323
},
{
"epoch": 0.03822106877433054,
"grad_norm": 0.87890625,
"learning_rate": 7.755498662229356e-05,
"loss": 1.6802,
"num_input_tokens_seen": 1561853952,
"step": 1324
},
{
"epoch": 0.03833903503597971,
"grad_norm": 0.984375,
"learning_rate": 7.754961241195689e-05,
"loss": 1.6442,
"num_input_tokens_seen": 1563033600,
"step": 1325
},
{
"epoch": 0.03845700129762888,
"grad_norm": 0.9453125,
"learning_rate": 7.754423248838708e-05,
"loss": 1.648,
"num_input_tokens_seen": 1564213248,
"step": 1326
},
{
"epoch": 0.03857496755927805,
"grad_norm": 1.046875,
"learning_rate": 7.753884685240267e-05,
"loss": 1.6544,
"num_input_tokens_seen": 1565392896,
"step": 1327
},
{
"epoch": 0.038692933820927214,
"grad_norm": 0.99609375,
"learning_rate": 7.75334555048231e-05,
"loss": 1.6477,
"num_input_tokens_seen": 1566572544,
"step": 1328
},
{
"epoch": 0.038810900082576386,
"grad_norm": 1.0625,
"learning_rate": 7.752805844646867e-05,
"loss": 1.6371,
"num_input_tokens_seen": 1567752192,
"step": 1329
},
{
"epoch": 0.03892886634422555,
"grad_norm": 1.171875,
"learning_rate": 7.752265567816055e-05,
"loss": 1.7027,
"num_input_tokens_seen": 1568931840,
"step": 1330
},
{
"epoch": 0.03904683260587472,
"grad_norm": 0.9140625,
"learning_rate": 7.75172472007208e-05,
"loss": 1.6617,
"num_input_tokens_seen": 1570111488,
"step": 1331
},
{
"epoch": 0.03916479886752389,
"grad_norm": 0.9140625,
"learning_rate": 7.75118330149723e-05,
"loss": 1.7437,
"num_input_tokens_seen": 1571291136,
"step": 1332
},
{
"epoch": 0.03928276512917306,
"grad_norm": 1.015625,
"learning_rate": 7.750641312173886e-05,
"loss": 1.6968,
"num_input_tokens_seen": 1572470784,
"step": 1333
},
{
"epoch": 0.03940073139082222,
"grad_norm": 1.0234375,
"learning_rate": 7.750098752184507e-05,
"loss": 1.5903,
"num_input_tokens_seen": 1573650432,
"step": 1334
},
{
"epoch": 0.039518697652471395,
"grad_norm": 0.85546875,
"learning_rate": 7.74955562161165e-05,
"loss": 1.6188,
"num_input_tokens_seen": 1574830080,
"step": 1335
},
{
"epoch": 0.03963666391412056,
"grad_norm": 1.09375,
"learning_rate": 7.749011920537951e-05,
"loss": 1.5991,
"num_input_tokens_seen": 1576009728,
"step": 1336
},
{
"epoch": 0.03975463017576973,
"grad_norm": 0.94140625,
"learning_rate": 7.748467649046135e-05,
"loss": 1.5751,
"num_input_tokens_seen": 1577189376,
"step": 1337
},
{
"epoch": 0.039872596437418896,
"grad_norm": 0.90625,
"learning_rate": 7.747922807219012e-05,
"loss": 1.6036,
"num_input_tokens_seen": 1578369024,
"step": 1338
},
{
"epoch": 0.03999056269906807,
"grad_norm": 0.9375,
"learning_rate": 7.747377395139484e-05,
"loss": 1.6377,
"num_input_tokens_seen": 1579548672,
"step": 1339
},
{
"epoch": 0.04010852896071723,
"grad_norm": 1.0390625,
"learning_rate": 7.746831412890534e-05,
"loss": 1.6241,
"num_input_tokens_seen": 1580728320,
"step": 1340
},
{
"epoch": 0.040226495222366404,
"grad_norm": 0.875,
"learning_rate": 7.746284860555235e-05,
"loss": 1.6902,
"num_input_tokens_seen": 1581907968,
"step": 1341
},
{
"epoch": 0.04034446148401557,
"grad_norm": 0.98828125,
"learning_rate": 7.745737738216745e-05,
"loss": 1.7062,
"num_input_tokens_seen": 1583087616,
"step": 1342
},
{
"epoch": 0.04046242774566474,
"grad_norm": 0.9375,
"learning_rate": 7.74519004595831e-05,
"loss": 1.6637,
"num_input_tokens_seen": 1584267264,
"step": 1343
},
{
"epoch": 0.040580394007313905,
"grad_norm": 0.8359375,
"learning_rate": 7.744641783863263e-05,
"loss": 1.725,
"num_input_tokens_seen": 1585446912,
"step": 1344
},
{
"epoch": 0.04069836026896308,
"grad_norm": 1.046875,
"learning_rate": 7.744092952015023e-05,
"loss": 1.6461,
"num_input_tokens_seen": 1586626560,
"step": 1345
},
{
"epoch": 0.04081632653061224,
"grad_norm": 0.97265625,
"learning_rate": 7.743543550497093e-05,
"loss": 1.7593,
"num_input_tokens_seen": 1587806208,
"step": 1346
},
{
"epoch": 0.04093429279226141,
"grad_norm": 0.9609375,
"learning_rate": 7.742993579393067e-05,
"loss": 1.5863,
"num_input_tokens_seen": 1588985856,
"step": 1347
},
{
"epoch": 0.041052259053910585,
"grad_norm": 0.9140625,
"learning_rate": 7.742443038786625e-05,
"loss": 1.6812,
"num_input_tokens_seen": 1590165504,
"step": 1348
},
{
"epoch": 0.04117022531555975,
"grad_norm": 1.0703125,
"learning_rate": 7.741891928761533e-05,
"loss": 1.5854,
"num_input_tokens_seen": 1591345152,
"step": 1349
},
{
"epoch": 0.04128819157720892,
"grad_norm": 0.90625,
"learning_rate": 7.74134024940164e-05,
"loss": 1.7451,
"num_input_tokens_seen": 1592524800,
"step": 1350
},
{
"epoch": 0.041406157838858086,
"grad_norm": 0.859375,
"learning_rate": 7.740788000790888e-05,
"loss": 1.7746,
"num_input_tokens_seen": 1593704448,
"step": 1351
},
{
"epoch": 0.04152412410050726,
"grad_norm": 0.8828125,
"learning_rate": 7.740235183013301e-05,
"loss": 1.6677,
"num_input_tokens_seen": 1594884096,
"step": 1352
},
{
"epoch": 0.04164209036215642,
"grad_norm": 1.109375,
"learning_rate": 7.739681796152992e-05,
"loss": 1.6485,
"num_input_tokens_seen": 1596063744,
"step": 1353
},
{
"epoch": 0.041760056623805594,
"grad_norm": 0.98046875,
"learning_rate": 7.739127840294159e-05,
"loss": 1.7374,
"num_input_tokens_seen": 1597243392,
"step": 1354
},
{
"epoch": 0.04187802288545476,
"grad_norm": 0.91796875,
"learning_rate": 7.738573315521088e-05,
"loss": 1.6461,
"num_input_tokens_seen": 1598423040,
"step": 1355
},
{
"epoch": 0.04199598914710393,
"grad_norm": 0.99609375,
"learning_rate": 7.738018221918148e-05,
"loss": 1.7057,
"num_input_tokens_seen": 1599602688,
"step": 1356
},
{
"epoch": 0.042113955408753095,
"grad_norm": 1.2109375,
"learning_rate": 7.737462559569803e-05,
"loss": 1.5765,
"num_input_tokens_seen": 1600782336,
"step": 1357
},
{
"epoch": 0.04223192167040227,
"grad_norm": 0.921875,
"learning_rate": 7.736906328560593e-05,
"loss": 1.5281,
"num_input_tokens_seen": 1601961984,
"step": 1358
},
{
"epoch": 0.04234988793205143,
"grad_norm": 1.1484375,
"learning_rate": 7.736349528975151e-05,
"loss": 1.7288,
"num_input_tokens_seen": 1603141632,
"step": 1359
},
{
"epoch": 0.042467854193700603,
"grad_norm": 1.078125,
"learning_rate": 7.735792160898194e-05,
"loss": 1.5428,
"num_input_tokens_seen": 1604321280,
"step": 1360
},
{
"epoch": 0.04258582045534977,
"grad_norm": 0.9140625,
"learning_rate": 7.735234224414528e-05,
"loss": 1.5481,
"num_input_tokens_seen": 1605500928,
"step": 1361
},
{
"epoch": 0.04270378671699894,
"grad_norm": 1.0078125,
"learning_rate": 7.734675719609046e-05,
"loss": 1.6189,
"num_input_tokens_seen": 1606680576,
"step": 1362
},
{
"epoch": 0.042821752978648105,
"grad_norm": 0.7890625,
"learning_rate": 7.73411664656672e-05,
"loss": 1.6773,
"num_input_tokens_seen": 1607860224,
"step": 1363
},
{
"epoch": 0.042939719240297276,
"grad_norm": 1.1640625,
"learning_rate": 7.733557005372616e-05,
"loss": 1.482,
"num_input_tokens_seen": 1609039872,
"step": 1364
},
{
"epoch": 0.04305768550194644,
"grad_norm": 0.94921875,
"learning_rate": 7.732996796111887e-05,
"loss": 1.6277,
"num_input_tokens_seen": 1610219520,
"step": 1365
},
{
"epoch": 0.04317565176359561,
"grad_norm": 1.1015625,
"learning_rate": 7.732436018869765e-05,
"loss": 1.5224,
"num_input_tokens_seen": 1611399168,
"step": 1366
},
{
"epoch": 0.04329361802524478,
"grad_norm": 1.0625,
"learning_rate": 7.731874673731579e-05,
"loss": 1.581,
"num_input_tokens_seen": 1612578816,
"step": 1367
},
{
"epoch": 0.04341158428689395,
"grad_norm": 1.1171875,
"learning_rate": 7.731312760782732e-05,
"loss": 1.5601,
"num_input_tokens_seen": 1613758464,
"step": 1368
},
{
"epoch": 0.043529550548543114,
"grad_norm": 0.87890625,
"learning_rate": 7.730750280108725e-05,
"loss": 1.5671,
"num_input_tokens_seen": 1614938112,
"step": 1369
},
{
"epoch": 0.043647516810192286,
"grad_norm": 0.9296875,
"learning_rate": 7.730187231795139e-05,
"loss": 1.5482,
"num_input_tokens_seen": 1616117760,
"step": 1370
},
{
"epoch": 0.04376548307184145,
"grad_norm": 1.046875,
"learning_rate": 7.72962361592764e-05,
"loss": 1.516,
"num_input_tokens_seen": 1617297408,
"step": 1371
},
{
"epoch": 0.04388344933349062,
"grad_norm": 1.046875,
"learning_rate": 7.729059432591989e-05,
"loss": 1.5741,
"num_input_tokens_seen": 1618477056,
"step": 1372
},
{
"epoch": 0.04400141559513979,
"grad_norm": 0.90625,
"learning_rate": 7.72849468187402e-05,
"loss": 1.5939,
"num_input_tokens_seen": 1619656704,
"step": 1373
},
{
"epoch": 0.04411938185678896,
"grad_norm": 1.21875,
"learning_rate": 7.727929363859668e-05,
"loss": 1.503,
"num_input_tokens_seen": 1620836352,
"step": 1374
},
{
"epoch": 0.04423734811843813,
"grad_norm": 0.984375,
"learning_rate": 7.727363478634941e-05,
"loss": 1.6708,
"num_input_tokens_seen": 1622016000,
"step": 1375
},
{
"epoch": 0.044355314380087295,
"grad_norm": 1.0,
"learning_rate": 7.726797026285941e-05,
"loss": 1.7104,
"num_input_tokens_seen": 1623195648,
"step": 1376
},
{
"epoch": 0.044473280641736467,
"grad_norm": 1.015625,
"learning_rate": 7.726230006898857e-05,
"loss": 1.5368,
"num_input_tokens_seen": 1624375296,
"step": 1377
},
{
"epoch": 0.04459124690338563,
"grad_norm": 1.0390625,
"learning_rate": 7.725662420559961e-05,
"loss": 1.5915,
"num_input_tokens_seen": 1625554944,
"step": 1378
},
{
"epoch": 0.0447092131650348,
"grad_norm": 0.97265625,
"learning_rate": 7.725094267355611e-05,
"loss": 1.5744,
"num_input_tokens_seen": 1626734592,
"step": 1379
},
{
"epoch": 0.04482717942668397,
"grad_norm": 0.98046875,
"learning_rate": 7.724525547372251e-05,
"loss": 1.5304,
"num_input_tokens_seen": 1627914240,
"step": 1380
},
{
"epoch": 0.04494514568833314,
"grad_norm": 1.0,
"learning_rate": 7.723956260696416e-05,
"loss": 1.6273,
"num_input_tokens_seen": 1629093888,
"step": 1381
},
{
"epoch": 0.045063111949982304,
"grad_norm": 0.875,
"learning_rate": 7.723386407414723e-05,
"loss": 1.5527,
"num_input_tokens_seen": 1630273536,
"step": 1382
},
{
"epoch": 0.045181078211631476,
"grad_norm": 0.90625,
"learning_rate": 7.722815987613875e-05,
"loss": 1.6727,
"num_input_tokens_seen": 1631453184,
"step": 1383
},
{
"epoch": 0.04529904447328064,
"grad_norm": 0.9765625,
"learning_rate": 7.722245001380663e-05,
"loss": 1.5183,
"num_input_tokens_seen": 1632632832,
"step": 1384
},
{
"epoch": 0.04541701073492981,
"grad_norm": 0.86328125,
"learning_rate": 7.721673448801963e-05,
"loss": 1.5318,
"num_input_tokens_seen": 1633812480,
"step": 1385
},
{
"epoch": 0.04553497699657898,
"grad_norm": 0.98046875,
"learning_rate": 7.721101329964737e-05,
"loss": 1.5919,
"num_input_tokens_seen": 1634992128,
"step": 1386
},
{
"epoch": 0.04565294325822815,
"grad_norm": 1.140625,
"learning_rate": 7.720528644956035e-05,
"loss": 1.566,
"num_input_tokens_seen": 1636171776,
"step": 1387
},
{
"epoch": 0.04577090951987731,
"grad_norm": 0.92578125,
"learning_rate": 7.719955393862993e-05,
"loss": 1.5904,
"num_input_tokens_seen": 1637351424,
"step": 1388
},
{
"epoch": 0.045888875781526485,
"grad_norm": 0.83203125,
"learning_rate": 7.719381576772829e-05,
"loss": 1.7284,
"num_input_tokens_seen": 1638531072,
"step": 1389
},
{
"epoch": 0.04600684204317565,
"grad_norm": 0.98828125,
"learning_rate": 7.718807193772853e-05,
"loss": 1.5317,
"num_input_tokens_seen": 1639710720,
"step": 1390
},
{
"epoch": 0.04612480830482482,
"grad_norm": 0.96875,
"learning_rate": 7.718232244950455e-05,
"loss": 1.5254,
"num_input_tokens_seen": 1640890368,
"step": 1391
},
{
"epoch": 0.046242774566473986,
"grad_norm": 0.99609375,
"learning_rate": 7.717656730393119e-05,
"loss": 1.5747,
"num_input_tokens_seen": 1642070016,
"step": 1392
},
{
"epoch": 0.04636074082812316,
"grad_norm": 1.0,
"learning_rate": 7.717080650188407e-05,
"loss": 1.606,
"num_input_tokens_seen": 1643249664,
"step": 1393
},
{
"epoch": 0.04647870708977232,
"grad_norm": 1.0234375,
"learning_rate": 7.716504004423972e-05,
"loss": 1.5735,
"num_input_tokens_seen": 1644429312,
"step": 1394
},
{
"epoch": 0.046596673351421494,
"grad_norm": 1.015625,
"learning_rate": 7.715926793187551e-05,
"loss": 1.6629,
"num_input_tokens_seen": 1645608960,
"step": 1395
},
{
"epoch": 0.04671463961307066,
"grad_norm": 1.109375,
"learning_rate": 7.715349016566968e-05,
"loss": 1.6771,
"num_input_tokens_seen": 1646788608,
"step": 1396
},
{
"epoch": 0.04683260587471983,
"grad_norm": 0.87890625,
"learning_rate": 7.71477067465013e-05,
"loss": 1.8481,
"num_input_tokens_seen": 1647968256,
"step": 1397
},
{
"epoch": 0.046950572136368995,
"grad_norm": 1.1171875,
"learning_rate": 7.714191767525036e-05,
"loss": 1.6027,
"num_input_tokens_seen": 1649147904,
"step": 1398
},
{
"epoch": 0.04706853839801817,
"grad_norm": 1.015625,
"learning_rate": 7.713612295279767e-05,
"loss": 1.6044,
"num_input_tokens_seen": 1650327552,
"step": 1399
},
{
"epoch": 0.04718650465966733,
"grad_norm": 1.0,
"learning_rate": 7.713032258002491e-05,
"loss": 1.6524,
"num_input_tokens_seen": 1651507200,
"step": 1400
},
{
"epoch": 0.04718650465966733,
"eval_wikipedia_loss": 2.2798447608947754,
"eval_wikipedia_runtime": 161.5268,
"eval_wikipedia_samples_per_second": 4.346,
"eval_wikipedia_steps_per_second": 0.186,
"num_input_tokens_seen": 1651507200,
"step": 1400
},
{
"epoch": 0.04718650465966733,
"eval_toxicity_loss": 4.040336608886719,
"eval_toxicity_runtime": 0.9662,
"eval_toxicity_samples_per_second": 2.07,
"eval_toxicity_steps_per_second": 1.035,
"num_input_tokens_seen": 1651507200,
"step": 1400
},
{
"epoch": 0.047304470921316503,
"grad_norm": 1.03125,
"learning_rate": 7.712451655781459e-05,
"loss": 1.7851,
"num_input_tokens_seen": 1652686848,
"step": 1401
},
{
"epoch": 0.047422437182965675,
"grad_norm": 1.0859375,
"learning_rate": 7.711870488705014e-05,
"loss": 1.6764,
"num_input_tokens_seen": 1653866496,
"step": 1402
},
{
"epoch": 0.04754040344461484,
"grad_norm": 1.015625,
"learning_rate": 7.711288756861579e-05,
"loss": 1.7252,
"num_input_tokens_seen": 1655046144,
"step": 1403
},
{
"epoch": 0.04765836970626401,
"grad_norm": 0.94140625,
"learning_rate": 7.710706460339665e-05,
"loss": 1.7524,
"num_input_tokens_seen": 1656225792,
"step": 1404
},
{
"epoch": 0.047776335967913176,
"grad_norm": 1.0625,
"learning_rate": 7.710123599227873e-05,
"loss": 1.7855,
"num_input_tokens_seen": 1657405440,
"step": 1405
},
{
"epoch": 0.04789430222956235,
"grad_norm": 0.81640625,
"learning_rate": 7.709540173614883e-05,
"loss": 1.77,
"num_input_tokens_seen": 1658585088,
"step": 1406
},
{
"epoch": 0.04801226849121151,
"grad_norm": 1.0390625,
"learning_rate": 7.708956183589466e-05,
"loss": 1.7499,
"num_input_tokens_seen": 1659764736,
"step": 1407
},
{
"epoch": 0.048130234752860684,
"grad_norm": 0.953125,
"learning_rate": 7.708371629240474e-05,
"loss": 1.5442,
"num_input_tokens_seen": 1660944384,
"step": 1408
},
{
"epoch": 0.04824820101450985,
"grad_norm": 0.953125,
"learning_rate": 7.707786510656849e-05,
"loss": 1.7384,
"num_input_tokens_seen": 1662124032,
"step": 1409
},
{
"epoch": 0.04836616727615902,
"grad_norm": 0.88671875,
"learning_rate": 7.70720082792762e-05,
"loss": 1.7988,
"num_input_tokens_seen": 1663303680,
"step": 1410
},
{
"epoch": 0.048484133537808186,
"grad_norm": 0.8828125,
"learning_rate": 7.706614581141898e-05,
"loss": 1.6953,
"num_input_tokens_seen": 1664483328,
"step": 1411
},
{
"epoch": 0.04860209979945736,
"grad_norm": 0.82421875,
"learning_rate": 7.706027770388881e-05,
"loss": 1.7006,
"num_input_tokens_seen": 1665662976,
"step": 1412
},
{
"epoch": 0.04872006606110652,
"grad_norm": 0.85546875,
"learning_rate": 7.705440395757855e-05,
"loss": 1.8123,
"num_input_tokens_seen": 1666842624,
"step": 1413
},
{
"epoch": 0.048838032322755694,
"grad_norm": 0.87890625,
"learning_rate": 7.704852457338188e-05,
"loss": 1.5978,
"num_input_tokens_seen": 1668022272,
"step": 1414
},
{
"epoch": 0.04895599858440486,
"grad_norm": 0.7421875,
"learning_rate": 7.704263955219336e-05,
"loss": 1.8147,
"num_input_tokens_seen": 1669201920,
"step": 1415
},
{
"epoch": 0.04907396484605403,
"grad_norm": 0.87890625,
"learning_rate": 7.703674889490842e-05,
"loss": 1.6817,
"num_input_tokens_seen": 1670381568,
"step": 1416
},
{
"epoch": 0.049191931107703195,
"grad_norm": 0.8359375,
"learning_rate": 7.703085260242331e-05,
"loss": 1.7774,
"num_input_tokens_seen": 1671561216,
"step": 1417
},
{
"epoch": 0.049309897369352367,
"grad_norm": 0.77734375,
"learning_rate": 7.702495067563519e-05,
"loss": 1.7312,
"num_input_tokens_seen": 1672740864,
"step": 1418
},
{
"epoch": 0.04942786363100153,
"grad_norm": 0.82421875,
"learning_rate": 7.701904311544202e-05,
"loss": 1.6888,
"num_input_tokens_seen": 1673920512,
"step": 1419
},
{
"epoch": 0.0495458298926507,
"grad_norm": 0.7578125,
"learning_rate": 7.701312992274265e-05,
"loss": 1.7375,
"num_input_tokens_seen": 1675100160,
"step": 1420
},
{
"epoch": 0.04966379615429987,
"grad_norm": 0.875,
"learning_rate": 7.700721109843679e-05,
"loss": 1.8633,
"num_input_tokens_seen": 1676279808,
"step": 1421
},
{
"epoch": 0.04978176241594904,
"grad_norm": 0.640625,
"learning_rate": 7.7001286643425e-05,
"loss": 1.7872,
"num_input_tokens_seen": 1677459456,
"step": 1422
},
{
"epoch": 0.049899728677598204,
"grad_norm": 0.8828125,
"learning_rate": 7.699535655860868e-05,
"loss": 1.7318,
"num_input_tokens_seen": 1678639104,
"step": 1423
},
{
"epoch": 0.050017694939247376,
"grad_norm": 0.76171875,
"learning_rate": 7.698942084489012e-05,
"loss": 1.7188,
"num_input_tokens_seen": 1679818752,
"step": 1424
},
{
"epoch": 0.05013566120089654,
"grad_norm": 0.859375,
"learning_rate": 7.698347950317244e-05,
"loss": 1.7577,
"num_input_tokens_seen": 1680998400,
"step": 1425
},
{
"epoch": 0.05025362746254571,
"grad_norm": 0.69921875,
"learning_rate": 7.697753253435962e-05,
"loss": 1.7042,
"num_input_tokens_seen": 1682178048,
"step": 1426
},
{
"epoch": 0.05037159372419488,
"grad_norm": 0.76953125,
"learning_rate": 7.697157993935651e-05,
"loss": 1.7375,
"num_input_tokens_seen": 1683357696,
"step": 1427
},
{
"epoch": 0.05048955998584405,
"grad_norm": 0.87109375,
"learning_rate": 7.696562171906881e-05,
"loss": 1.6805,
"num_input_tokens_seen": 1684537344,
"step": 1428
},
{
"epoch": 0.05060752624749322,
"grad_norm": 0.8515625,
"learning_rate": 7.695965787440305e-05,
"loss": 1.6456,
"num_input_tokens_seen": 1685716992,
"step": 1429
},
{
"epoch": 0.050725492509142385,
"grad_norm": 0.93359375,
"learning_rate": 7.695368840626666e-05,
"loss": 1.7703,
"num_input_tokens_seen": 1686896640,
"step": 1430
},
{
"epoch": 0.05084345877079156,
"grad_norm": 0.89453125,
"learning_rate": 7.694771331556792e-05,
"loss": 1.6603,
"num_input_tokens_seen": 1688076288,
"step": 1431
},
{
"epoch": 0.05096142503244072,
"grad_norm": 0.96875,
"learning_rate": 7.694173260321589e-05,
"loss": 1.6734,
"num_input_tokens_seen": 1689255936,
"step": 1432
},
{
"epoch": 0.05107939129408989,
"grad_norm": 0.82421875,
"learning_rate": 7.693574627012059e-05,
"loss": 1.705,
"num_input_tokens_seen": 1690435584,
"step": 1433
},
{
"epoch": 0.05119735755573906,
"grad_norm": 0.97265625,
"learning_rate": 7.692975431719285e-05,
"loss": 1.795,
"num_input_tokens_seen": 1691615232,
"step": 1434
},
{
"epoch": 0.05131532381738823,
"grad_norm": 0.90234375,
"learning_rate": 7.692375674534436e-05,
"loss": 1.7853,
"num_input_tokens_seen": 1692794880,
"step": 1435
},
{
"epoch": 0.051433290079037394,
"grad_norm": 0.76171875,
"learning_rate": 7.691775355548763e-05,
"loss": 1.7208,
"num_input_tokens_seen": 1693974528,
"step": 1436
},
{
"epoch": 0.051551256340686566,
"grad_norm": 0.83984375,
"learning_rate": 7.691174474853608e-05,
"loss": 1.7193,
"num_input_tokens_seen": 1695154176,
"step": 1437
},
{
"epoch": 0.05166922260233573,
"grad_norm": 0.83984375,
"learning_rate": 7.690573032540395e-05,
"loss": 1.6211,
"num_input_tokens_seen": 1696333824,
"step": 1438
},
{
"epoch": 0.0517871888639849,
"grad_norm": 0.83984375,
"learning_rate": 7.689971028700635e-05,
"loss": 1.8099,
"num_input_tokens_seen": 1697513472,
"step": 1439
},
{
"epoch": 0.05190515512563407,
"grad_norm": 0.8671875,
"learning_rate": 7.689368463425922e-05,
"loss": 1.6112,
"num_input_tokens_seen": 1698693120,
"step": 1440
},
{
"epoch": 0.05202312138728324,
"grad_norm": 0.79296875,
"learning_rate": 7.68876533680794e-05,
"loss": 1.7277,
"num_input_tokens_seen": 1699872768,
"step": 1441
},
{
"epoch": 0.052141087648932403,
"grad_norm": 0.87109375,
"learning_rate": 7.688161648938454e-05,
"loss": 1.6243,
"num_input_tokens_seen": 1701052416,
"step": 1442
},
{
"epoch": 0.052259053910581575,
"grad_norm": 0.80078125,
"learning_rate": 7.687557399909317e-05,
"loss": 1.6371,
"num_input_tokens_seen": 1702232064,
"step": 1443
},
{
"epoch": 0.05237702017223074,
"grad_norm": 0.94140625,
"learning_rate": 7.686952589812465e-05,
"loss": 1.605,
"num_input_tokens_seen": 1703411712,
"step": 1444
},
{
"epoch": 0.05249498643387991,
"grad_norm": 0.765625,
"learning_rate": 7.686347218739922e-05,
"loss": 1.731,
"num_input_tokens_seen": 1704591360,
"step": 1445
},
{
"epoch": 0.052612952695529076,
"grad_norm": 0.83984375,
"learning_rate": 7.685741286783795e-05,
"loss": 1.838,
"num_input_tokens_seen": 1705771008,
"step": 1446
},
{
"epoch": 0.05273091895717825,
"grad_norm": 0.81640625,
"learning_rate": 7.68513479403628e-05,
"loss": 1.6418,
"num_input_tokens_seen": 1706950656,
"step": 1447
},
{
"epoch": 0.05284888521882741,
"grad_norm": 0.78125,
"learning_rate": 7.684527740589654e-05,
"loss": 1.7607,
"num_input_tokens_seen": 1708130304,
"step": 1448
},
{
"epoch": 0.052966851480476584,
"grad_norm": 0.859375,
"learning_rate": 7.68392012653628e-05,
"loss": 1.7545,
"num_input_tokens_seen": 1709309952,
"step": 1449
},
{
"epoch": 0.05308481774212575,
"grad_norm": 0.78125,
"learning_rate": 7.68331195196861e-05,
"loss": 1.712,
"num_input_tokens_seen": 1710489600,
"step": 1450
},
{
"epoch": 0.05320278400377492,
"grad_norm": 0.83984375,
"learning_rate": 7.682703216979178e-05,
"loss": 1.7965,
"num_input_tokens_seen": 1711669248,
"step": 1451
},
{
"epoch": 0.053320750265424086,
"grad_norm": 0.80078125,
"learning_rate": 7.682093921660603e-05,
"loss": 1.6449,
"num_input_tokens_seen": 1712848896,
"step": 1452
},
{
"epoch": 0.05343871652707326,
"grad_norm": 0.8046875,
"learning_rate": 7.68148406610559e-05,
"loss": 1.6439,
"num_input_tokens_seen": 1714028544,
"step": 1453
},
{
"epoch": 0.05355668278872242,
"grad_norm": 1.0078125,
"learning_rate": 7.680873650406931e-05,
"loss": 1.5848,
"num_input_tokens_seen": 1715208192,
"step": 1454
},
{
"epoch": 0.053674649050371594,
"grad_norm": 0.828125,
"learning_rate": 7.6802626746575e-05,
"loss": 1.6039,
"num_input_tokens_seen": 1716387840,
"step": 1455
},
{
"epoch": 0.053792615312020765,
"grad_norm": 0.890625,
"learning_rate": 7.679651138950259e-05,
"loss": 1.6822,
"num_input_tokens_seen": 1717567488,
"step": 1456
},
{
"epoch": 0.05391058157366993,
"grad_norm": 0.92578125,
"learning_rate": 7.679039043378256e-05,
"loss": 1.8276,
"num_input_tokens_seen": 1718747136,
"step": 1457
},
{
"epoch": 0.0540285478353191,
"grad_norm": 0.95703125,
"learning_rate": 7.678426388034618e-05,
"loss": 1.7017,
"num_input_tokens_seen": 1719926784,
"step": 1458
},
{
"epoch": 0.054146514096968267,
"grad_norm": 0.96484375,
"learning_rate": 7.677813173012566e-05,
"loss": 1.6949,
"num_input_tokens_seen": 1721106432,
"step": 1459
},
{
"epoch": 0.05426448035861744,
"grad_norm": 1.0546875,
"learning_rate": 7.677199398405397e-05,
"loss": 1.7601,
"num_input_tokens_seen": 1722286080,
"step": 1460
},
{
"epoch": 0.0543824466202666,
"grad_norm": 1.0078125,
"learning_rate": 7.676585064306503e-05,
"loss": 1.6369,
"num_input_tokens_seen": 1723465728,
"step": 1461
},
{
"epoch": 0.054500412881915775,
"grad_norm": 0.78125,
"learning_rate": 7.67597017080935e-05,
"loss": 1.7194,
"num_input_tokens_seen": 1724645376,
"step": 1462
},
{
"epoch": 0.05461837914356494,
"grad_norm": 1.109375,
"learning_rate": 7.675354718007501e-05,
"loss": 1.6558,
"num_input_tokens_seen": 1725825024,
"step": 1463
},
{
"epoch": 0.05473634540521411,
"grad_norm": 0.92578125,
"learning_rate": 7.674738705994595e-05,
"loss": 1.6609,
"num_input_tokens_seen": 1727004672,
"step": 1464
},
{
"epoch": 0.054854311666863276,
"grad_norm": 0.8359375,
"learning_rate": 7.674122134864359e-05,
"loss": 1.7041,
"num_input_tokens_seen": 1728184320,
"step": 1465
},
{
"epoch": 0.05497227792851245,
"grad_norm": 0.86328125,
"learning_rate": 7.673505004710604e-05,
"loss": 1.563,
"num_input_tokens_seen": 1729363968,
"step": 1466
},
{
"epoch": 0.05509024419016161,
"grad_norm": 0.7109375,
"learning_rate": 7.672887315627232e-05,
"loss": 1.6657,
"num_input_tokens_seen": 1730543616,
"step": 1467
},
{
"epoch": 0.055208210451810784,
"grad_norm": 0.82421875,
"learning_rate": 7.67226906770822e-05,
"loss": 1.7601,
"num_input_tokens_seen": 1731723264,
"step": 1468
},
{
"epoch": 0.05532617671345995,
"grad_norm": 0.75390625,
"learning_rate": 7.67165026104764e-05,
"loss": 1.6698,
"num_input_tokens_seen": 1732902912,
"step": 1469
},
{
"epoch": 0.05544414297510912,
"grad_norm": 0.84765625,
"learning_rate": 7.671030895739641e-05,
"loss": 1.6914,
"num_input_tokens_seen": 1734082560,
"step": 1470
},
{
"epoch": 0.055562109236758285,
"grad_norm": 0.765625,
"learning_rate": 7.670410971878464e-05,
"loss": 2.0048,
"num_input_tokens_seen": 1735262208,
"step": 1471
},
{
"epoch": 0.05568007549840746,
"grad_norm": 0.93359375,
"learning_rate": 7.669790489558426e-05,
"loss": 1.6403,
"num_input_tokens_seen": 1736441856,
"step": 1472
},
{
"epoch": 0.05579804176005662,
"grad_norm": 0.92578125,
"learning_rate": 7.66916944887394e-05,
"loss": 1.78,
"num_input_tokens_seen": 1737621504,
"step": 1473
},
{
"epoch": 0.05591600802170579,
"grad_norm": 0.81640625,
"learning_rate": 7.668547849919494e-05,
"loss": 1.667,
"num_input_tokens_seen": 1738801152,
"step": 1474
},
{
"epoch": 0.05603397428335496,
"grad_norm": 0.87109375,
"learning_rate": 7.667925692789668e-05,
"loss": 1.683,
"num_input_tokens_seen": 1739980800,
"step": 1475
},
{
"epoch": 0.05615194054500413,
"grad_norm": 0.921875,
"learning_rate": 7.667302977579124e-05,
"loss": 1.6781,
"num_input_tokens_seen": 1741160448,
"step": 1476
},
{
"epoch": 0.056269906806653294,
"grad_norm": 0.81640625,
"learning_rate": 7.666679704382607e-05,
"loss": 1.6615,
"num_input_tokens_seen": 1742340096,
"step": 1477
},
{
"epoch": 0.056387873068302466,
"grad_norm": 0.83203125,
"learning_rate": 7.66605587329495e-05,
"loss": 1.6439,
"num_input_tokens_seen": 1743519744,
"step": 1478
},
{
"epoch": 0.05650583932995163,
"grad_norm": 0.80859375,
"learning_rate": 7.665431484411072e-05,
"loss": 1.6852,
"num_input_tokens_seen": 1744699392,
"step": 1479
},
{
"epoch": 0.0566238055916008,
"grad_norm": 0.97265625,
"learning_rate": 7.664806537825972e-05,
"loss": 1.7587,
"num_input_tokens_seen": 1745879040,
"step": 1480
},
{
"epoch": 0.05674177185324997,
"grad_norm": 0.82421875,
"learning_rate": 7.664181033634738e-05,
"loss": 1.6441,
"num_input_tokens_seen": 1747058688,
"step": 1481
},
{
"epoch": 0.05685973811489914,
"grad_norm": 0.69921875,
"learning_rate": 7.66355497193254e-05,
"loss": 1.7134,
"num_input_tokens_seen": 1748238336,
"step": 1482
},
{
"epoch": 0.05697770437654831,
"grad_norm": 0.8828125,
"learning_rate": 7.662928352814638e-05,
"loss": 1.6675,
"num_input_tokens_seen": 1749417984,
"step": 1483
},
{
"epoch": 0.057095670638197475,
"grad_norm": 0.8515625,
"learning_rate": 7.662301176376367e-05,
"loss": 1.657,
"num_input_tokens_seen": 1750597632,
"step": 1484
},
{
"epoch": 0.05721363689984665,
"grad_norm": 0.875,
"learning_rate": 7.661673442713159e-05,
"loss": 1.6178,
"num_input_tokens_seen": 1751777280,
"step": 1485
},
{
"epoch": 0.05733160316149581,
"grad_norm": 0.93359375,
"learning_rate": 7.661045151920522e-05,
"loss": 1.5022,
"num_input_tokens_seen": 1752956928,
"step": 1486
},
{
"epoch": 0.05744956942314498,
"grad_norm": 1.0,
"learning_rate": 7.66041630409405e-05,
"loss": 1.5419,
"num_input_tokens_seen": 1754136576,
"step": 1487
},
{
"epoch": 0.05756753568479415,
"grad_norm": 0.9609375,
"learning_rate": 7.659786899329426e-05,
"loss": 1.6542,
"num_input_tokens_seen": 1755316224,
"step": 1488
},
{
"epoch": 0.05768550194644332,
"grad_norm": 0.83984375,
"learning_rate": 7.659156937722413e-05,
"loss": 1.6363,
"num_input_tokens_seen": 1756495872,
"step": 1489
},
{
"epoch": 0.057803468208092484,
"grad_norm": 0.953125,
"learning_rate": 7.658526419368863e-05,
"loss": 1.7117,
"num_input_tokens_seen": 1757675520,
"step": 1490
},
{
"epoch": 0.057921434469741656,
"grad_norm": 0.94140625,
"learning_rate": 7.657895344364707e-05,
"loss": 1.6056,
"num_input_tokens_seen": 1758855168,
"step": 1491
},
{
"epoch": 0.05803940073139082,
"grad_norm": 0.7890625,
"learning_rate": 7.657263712805966e-05,
"loss": 1.7623,
"num_input_tokens_seen": 1760034816,
"step": 1492
},
{
"epoch": 0.05815736699303999,
"grad_norm": 0.93359375,
"learning_rate": 7.656631524788743e-05,
"loss": 1.5595,
"num_input_tokens_seen": 1761214464,
"step": 1493
},
{
"epoch": 0.05827533325468916,
"grad_norm": 0.70703125,
"learning_rate": 7.655998780409228e-05,
"loss": 1.7031,
"num_input_tokens_seen": 1762394112,
"step": 1494
},
{
"epoch": 0.05839329951633833,
"grad_norm": 0.98828125,
"learning_rate": 7.655365479763693e-05,
"loss": 1.5904,
"num_input_tokens_seen": 1763573760,
"step": 1495
},
{
"epoch": 0.058511265777987494,
"grad_norm": 0.83203125,
"learning_rate": 7.654731622948494e-05,
"loss": 1.7596,
"num_input_tokens_seen": 1764753408,
"step": 1496
},
{
"epoch": 0.058629232039636665,
"grad_norm": 0.87109375,
"learning_rate": 7.654097210060076e-05,
"loss": 1.7201,
"num_input_tokens_seen": 1765933056,
"step": 1497
},
{
"epoch": 0.05874719830128583,
"grad_norm": 0.7578125,
"learning_rate": 7.653462241194965e-05,
"loss": 1.5919,
"num_input_tokens_seen": 1767112704,
"step": 1498
},
{
"epoch": 0.058865164562935,
"grad_norm": 0.77734375,
"learning_rate": 7.652826716449773e-05,
"loss": 1.718,
"num_input_tokens_seen": 1768292352,
"step": 1499
},
{
"epoch": 0.058983130824584167,
"grad_norm": 0.9609375,
"learning_rate": 7.652190635921195e-05,
"loss": 1.6418,
"num_input_tokens_seen": 1769472000,
"step": 1500
},
{
"epoch": 0.05910109708623334,
"grad_norm": 0.78125,
"learning_rate": 7.651553999706011e-05,
"loss": 1.601,
"num_input_tokens_seen": 1770651648,
"step": 1501
},
{
"epoch": 0.0592190633478825,
"grad_norm": 0.94921875,
"learning_rate": 7.65091680790109e-05,
"loss": 1.7251,
"num_input_tokens_seen": 1771831296,
"step": 1502
},
{
"epoch": 0.059337029609531675,
"grad_norm": 0.8828125,
"learning_rate": 7.650279060603376e-05,
"loss": 1.611,
"num_input_tokens_seen": 1773010944,
"step": 1503
},
{
"epoch": 0.05945499587118084,
"grad_norm": 0.8671875,
"learning_rate": 7.649640757909907e-05,
"loss": 1.6156,
"num_input_tokens_seen": 1774190592,
"step": 1504
},
{
"epoch": 0.05957296213283001,
"grad_norm": 0.796875,
"learning_rate": 7.649001899917804e-05,
"loss": 1.6175,
"num_input_tokens_seen": 1775370240,
"step": 1505
},
{
"epoch": 0.059690928394479176,
"grad_norm": 0.9140625,
"learning_rate": 7.648362486724266e-05,
"loss": 1.6087,
"num_input_tokens_seen": 1776549888,
"step": 1506
},
{
"epoch": 0.05980889465612835,
"grad_norm": 0.82421875,
"learning_rate": 7.64772251842658e-05,
"loss": 1.6076,
"num_input_tokens_seen": 1777729536,
"step": 1507
},
{
"epoch": 0.05992686091777751,
"grad_norm": 0.953125,
"learning_rate": 7.647081995122122e-05,
"loss": 1.5446,
"num_input_tokens_seen": 1778909184,
"step": 1508
},
{
"epoch": 0.060044827179426684,
"grad_norm": 0.76953125,
"learning_rate": 7.646440916908347e-05,
"loss": 1.8014,
"num_input_tokens_seen": 1780088832,
"step": 1509
},
{
"epoch": 0.060162793441075856,
"grad_norm": 0.88671875,
"learning_rate": 7.645799283882797e-05,
"loss": 1.736,
"num_input_tokens_seen": 1781268480,
"step": 1510
},
{
"epoch": 0.06028075970272502,
"grad_norm": 0.97265625,
"learning_rate": 7.645157096143096e-05,
"loss": 1.6802,
"num_input_tokens_seen": 1782448128,
"step": 1511
},
{
"epoch": 0.06039872596437419,
"grad_norm": 0.92578125,
"learning_rate": 7.644514353786954e-05,
"loss": 1.7258,
"num_input_tokens_seen": 1783627776,
"step": 1512
},
{
"epoch": 0.06051669222602336,
"grad_norm": 0.9140625,
"learning_rate": 7.643871056912165e-05,
"loss": 1.5829,
"num_input_tokens_seen": 1784807424,
"step": 1513
},
{
"epoch": 0.06063465848767253,
"grad_norm": 0.96484375,
"learning_rate": 7.64322720561661e-05,
"loss": 1.6509,
"num_input_tokens_seen": 1785987072,
"step": 1514
},
{
"epoch": 0.06075262474932169,
"grad_norm": 0.81640625,
"learning_rate": 7.642582799998249e-05,
"loss": 1.6295,
"num_input_tokens_seen": 1787166720,
"step": 1515
},
{
"epoch": 0.060870591010970865,
"grad_norm": 0.8671875,
"learning_rate": 7.641937840155132e-05,
"loss": 1.6291,
"num_input_tokens_seen": 1788346368,
"step": 1516
},
{
"epoch": 0.06098855727262003,
"grad_norm": 0.90234375,
"learning_rate": 7.641292326185389e-05,
"loss": 1.6605,
"num_input_tokens_seen": 1789526016,
"step": 1517
},
{
"epoch": 0.0611065235342692,
"grad_norm": 0.80078125,
"learning_rate": 7.640646258187236e-05,
"loss": 1.6616,
"num_input_tokens_seen": 1790705664,
"step": 1518
},
{
"epoch": 0.061224489795918366,
"grad_norm": 0.75,
"learning_rate": 7.639999636258972e-05,
"loss": 1.6898,
"num_input_tokens_seen": 1791885312,
"step": 1519
},
{
"epoch": 0.06134245605756754,
"grad_norm": 0.875,
"learning_rate": 7.639352460498985e-05,
"loss": 1.7334,
"num_input_tokens_seen": 1793064960,
"step": 1520
},
{
"epoch": 0.0614604223192167,
"grad_norm": 0.83203125,
"learning_rate": 7.638704731005741e-05,
"loss": 1.5819,
"num_input_tokens_seen": 1794244608,
"step": 1521
},
{
"epoch": 0.061578388580865874,
"grad_norm": 0.7890625,
"learning_rate": 7.638056447877795e-05,
"loss": 1.7682,
"num_input_tokens_seen": 1795424256,
"step": 1522
},
{
"epoch": 0.06169635484251504,
"grad_norm": 0.88671875,
"learning_rate": 7.637407611213783e-05,
"loss": 1.6932,
"num_input_tokens_seen": 1796603904,
"step": 1523
},
{
"epoch": 0.06181432110416421,
"grad_norm": 0.734375,
"learning_rate": 7.636758221112428e-05,
"loss": 1.6488,
"num_input_tokens_seen": 1797783552,
"step": 1524
},
{
"epoch": 0.061932287365813375,
"grad_norm": 0.84765625,
"learning_rate": 7.636108277672533e-05,
"loss": 1.7371,
"num_input_tokens_seen": 1798963200,
"step": 1525
},
{
"epoch": 0.06205025362746255,
"grad_norm": 0.80078125,
"learning_rate": 7.635457780992989e-05,
"loss": 1.7218,
"num_input_tokens_seen": 1800142848,
"step": 1526
},
{
"epoch": 0.06216821988911171,
"grad_norm": 0.81640625,
"learning_rate": 7.634806731172772e-05,
"loss": 1.7231,
"num_input_tokens_seen": 1801322496,
"step": 1527
},
{
"epoch": 0.06228618615076088,
"grad_norm": 0.81640625,
"learning_rate": 7.634155128310938e-05,
"loss": 1.6153,
"num_input_tokens_seen": 1802502144,
"step": 1528
},
{
"epoch": 0.06240415241241005,
"grad_norm": 0.92578125,
"learning_rate": 7.63350297250663e-05,
"loss": 1.6892,
"num_input_tokens_seen": 1803681792,
"step": 1529
},
{
"epoch": 0.06252211867405921,
"grad_norm": 0.890625,
"learning_rate": 7.632850263859077e-05,
"loss": 1.5461,
"num_input_tokens_seen": 1804861440,
"step": 1530
},
{
"epoch": 0.06264008493570838,
"grad_norm": 0.87890625,
"learning_rate": 7.632197002467586e-05,
"loss": 1.7642,
"num_input_tokens_seen": 1806041088,
"step": 1531
},
{
"epoch": 0.06275805119735756,
"grad_norm": 0.8671875,
"learning_rate": 7.631543188431552e-05,
"loss": 1.6223,
"num_input_tokens_seen": 1807220736,
"step": 1532
},
{
"epoch": 0.06287601745900673,
"grad_norm": 0.86328125,
"learning_rate": 7.630888821850457e-05,
"loss": 1.5815,
"num_input_tokens_seen": 1808400384,
"step": 1533
},
{
"epoch": 0.06299398372065589,
"grad_norm": 0.8828125,
"learning_rate": 7.630233902823862e-05,
"loss": 1.5879,
"num_input_tokens_seen": 1809580032,
"step": 1534
},
{
"epoch": 0.06311194998230506,
"grad_norm": 0.7578125,
"learning_rate": 7.629578431451415e-05,
"loss": 1.6571,
"num_input_tokens_seen": 1810759680,
"step": 1535
},
{
"epoch": 0.06322991624395423,
"grad_norm": 0.78125,
"learning_rate": 7.628922407832843e-05,
"loss": 1.6632,
"num_input_tokens_seen": 1811939328,
"step": 1536
},
{
"epoch": 0.0633478825056034,
"grad_norm": 0.82421875,
"learning_rate": 7.628265832067967e-05,
"loss": 1.8045,
"num_input_tokens_seen": 1813118976,
"step": 1537
},
{
"epoch": 0.06346584876725257,
"grad_norm": 0.828125,
"learning_rate": 7.627608704256681e-05,
"loss": 1.6111,
"num_input_tokens_seen": 1814298624,
"step": 1538
},
{
"epoch": 0.06358381502890173,
"grad_norm": 1.0234375,
"learning_rate": 7.62695102449897e-05,
"loss": 1.7442,
"num_input_tokens_seen": 1815478272,
"step": 1539
},
{
"epoch": 0.0637017812905509,
"grad_norm": 0.8203125,
"learning_rate": 7.626292792894902e-05,
"loss": 1.7246,
"num_input_tokens_seen": 1816657920,
"step": 1540
},
{
"epoch": 0.06381974755220007,
"grad_norm": 0.7890625,
"learning_rate": 7.625634009544627e-05,
"loss": 1.7061,
"num_input_tokens_seen": 1817837568,
"step": 1541
},
{
"epoch": 0.06393771381384925,
"grad_norm": 1.03125,
"learning_rate": 7.62497467454838e-05,
"loss": 1.5667,
"num_input_tokens_seen": 1819017216,
"step": 1542
},
{
"epoch": 0.0640556800754984,
"grad_norm": 0.82421875,
"learning_rate": 7.62431478800648e-05,
"loss": 1.6419,
"num_input_tokens_seen": 1820196864,
"step": 1543
},
{
"epoch": 0.06417364633714757,
"grad_norm": 0.8984375,
"learning_rate": 7.62365435001933e-05,
"loss": 1.6755,
"num_input_tokens_seen": 1821376512,
"step": 1544
},
{
"epoch": 0.06429161259879675,
"grad_norm": 0.84765625,
"learning_rate": 7.622993360687416e-05,
"loss": 1.6076,
"num_input_tokens_seen": 1822556160,
"step": 1545
},
{
"epoch": 0.06440957886044592,
"grad_norm": 0.84765625,
"learning_rate": 7.622331820111309e-05,
"loss": 1.7897,
"num_input_tokens_seen": 1823735808,
"step": 1546
},
{
"epoch": 0.06452754512209508,
"grad_norm": 0.78125,
"learning_rate": 7.621669728391663e-05,
"loss": 1.7062,
"num_input_tokens_seen": 1824915456,
"step": 1547
},
{
"epoch": 0.06464551138374425,
"grad_norm": 0.95703125,
"learning_rate": 7.621007085629218e-05,
"loss": 1.8787,
"num_input_tokens_seen": 1826095104,
"step": 1548
},
{
"epoch": 0.06476347764539342,
"grad_norm": 0.90625,
"learning_rate": 7.620343891924794e-05,
"loss": 1.6609,
"num_input_tokens_seen": 1827274752,
"step": 1549
},
{
"epoch": 0.06488144390704259,
"grad_norm": 0.8671875,
"learning_rate": 7.619680147379297e-05,
"loss": 1.6619,
"num_input_tokens_seen": 1828454400,
"step": 1550
},
{
"epoch": 0.06499941016869175,
"grad_norm": 0.94140625,
"learning_rate": 7.61901585209372e-05,
"loss": 1.5511,
"num_input_tokens_seen": 1829634048,
"step": 1551
},
{
"epoch": 0.06511737643034092,
"grad_norm": 0.82421875,
"learning_rate": 7.618351006169133e-05,
"loss": 1.6867,
"num_input_tokens_seen": 1830813696,
"step": 1552
},
{
"epoch": 0.06523534269199009,
"grad_norm": 0.8125,
"learning_rate": 7.617685609706693e-05,
"loss": 1.6889,
"num_input_tokens_seen": 1831993344,
"step": 1553
},
{
"epoch": 0.06535330895363926,
"grad_norm": 0.84765625,
"learning_rate": 7.617019662807645e-05,
"loss": 1.6725,
"num_input_tokens_seen": 1833172992,
"step": 1554
},
{
"epoch": 0.06547127521528842,
"grad_norm": 0.921875,
"learning_rate": 7.61635316557331e-05,
"loss": 1.6894,
"num_input_tokens_seen": 1834352640,
"step": 1555
},
{
"epoch": 0.0655892414769376,
"grad_norm": 0.734375,
"learning_rate": 7.6156861181051e-05,
"loss": 1.596,
"num_input_tokens_seen": 1835532288,
"step": 1556
},
{
"epoch": 0.06570720773858676,
"grad_norm": 0.9140625,
"learning_rate": 7.615018520504503e-05,
"loss": 1.6515,
"num_input_tokens_seen": 1836711936,
"step": 1557
},
{
"epoch": 0.06582517400023594,
"grad_norm": 0.98046875,
"learning_rate": 7.6143503728731e-05,
"loss": 1.642,
"num_input_tokens_seen": 1837891584,
"step": 1558
},
{
"epoch": 0.0659431402618851,
"grad_norm": 0.88671875,
"learning_rate": 7.613681675312548e-05,
"loss": 1.7575,
"num_input_tokens_seen": 1839071232,
"step": 1559
},
{
"epoch": 0.06606110652353427,
"grad_norm": 1.0,
"learning_rate": 7.61301242792459e-05,
"loss": 1.571,
"num_input_tokens_seen": 1840250880,
"step": 1560
},
{
"epoch": 0.06617907278518344,
"grad_norm": 0.875,
"learning_rate": 7.612342630811053e-05,
"loss": 1.495,
"num_input_tokens_seen": 1841430528,
"step": 1561
},
{
"epoch": 0.06629703904683261,
"grad_norm": 1.0546875,
"learning_rate": 7.61167228407385e-05,
"loss": 1.5052,
"num_input_tokens_seen": 1842610176,
"step": 1562
},
{
"epoch": 0.06641500530848178,
"grad_norm": 0.7734375,
"learning_rate": 7.611001387814973e-05,
"loss": 1.6763,
"num_input_tokens_seen": 1843789824,
"step": 1563
},
{
"epoch": 0.06653297157013094,
"grad_norm": 0.921875,
"learning_rate": 7.610329942136501e-05,
"loss": 1.683,
"num_input_tokens_seen": 1844969472,
"step": 1564
},
{
"epoch": 0.06665093783178011,
"grad_norm": 0.8359375,
"learning_rate": 7.609657947140594e-05,
"loss": 1.5549,
"num_input_tokens_seen": 1846149120,
"step": 1565
},
{
"epoch": 0.06676890409342928,
"grad_norm": 0.890625,
"learning_rate": 7.608985402929499e-05,
"loss": 1.556,
"num_input_tokens_seen": 1847328768,
"step": 1566
},
{
"epoch": 0.06688687035507845,
"grad_norm": 0.82421875,
"learning_rate": 7.608312309605545e-05,
"loss": 1.6592,
"num_input_tokens_seen": 1848508416,
"step": 1567
},
{
"epoch": 0.06700483661672761,
"grad_norm": 0.75390625,
"learning_rate": 7.607638667271142e-05,
"loss": 1.7267,
"num_input_tokens_seen": 1849688064,
"step": 1568
},
{
"epoch": 0.06712280287837678,
"grad_norm": 0.77734375,
"learning_rate": 7.606964476028788e-05,
"loss": 1.709,
"num_input_tokens_seen": 1850867712,
"step": 1569
},
{
"epoch": 0.06724076914002595,
"grad_norm": 0.80859375,
"learning_rate": 7.60628973598106e-05,
"loss": 1.5965,
"num_input_tokens_seen": 1852047360,
"step": 1570
},
{
"epoch": 0.06735873540167513,
"grad_norm": 0.71875,
"learning_rate": 7.605614447230622e-05,
"loss": 1.6316,
"num_input_tokens_seen": 1853227008,
"step": 1571
},
{
"epoch": 0.06747670166332428,
"grad_norm": 0.828125,
"learning_rate": 7.604938609880221e-05,
"loss": 1.7218,
"num_input_tokens_seen": 1854406656,
"step": 1572
},
{
"epoch": 0.06759466792497346,
"grad_norm": 0.7890625,
"learning_rate": 7.604262224032687e-05,
"loss": 1.6131,
"num_input_tokens_seen": 1855586304,
"step": 1573
},
{
"epoch": 0.06771263418662263,
"grad_norm": 0.90625,
"learning_rate": 7.603585289790931e-05,
"loss": 1.5626,
"num_input_tokens_seen": 1856765952,
"step": 1574
},
{
"epoch": 0.0678306004482718,
"grad_norm": 0.79296875,
"learning_rate": 7.60290780725795e-05,
"loss": 1.6016,
"num_input_tokens_seen": 1857945600,
"step": 1575
},
{
"epoch": 0.06794856670992096,
"grad_norm": 0.72265625,
"learning_rate": 7.602229776536826e-05,
"loss": 1.6785,
"num_input_tokens_seen": 1859125248,
"step": 1576
},
{
"epoch": 0.06806653297157013,
"grad_norm": 0.765625,
"learning_rate": 7.60155119773072e-05,
"loss": 1.824,
"num_input_tokens_seen": 1860304896,
"step": 1577
},
{
"epoch": 0.0681844992332193,
"grad_norm": 0.796875,
"learning_rate": 7.600872070942882e-05,
"loss": 1.6268,
"num_input_tokens_seen": 1861484544,
"step": 1578
},
{
"epoch": 0.06830246549486847,
"grad_norm": 0.78515625,
"learning_rate": 7.600192396276638e-05,
"loss": 1.6131,
"num_input_tokens_seen": 1862664192,
"step": 1579
},
{
"epoch": 0.06842043175651763,
"grad_norm": 0.7734375,
"learning_rate": 7.599512173835406e-05,
"loss": 1.6428,
"num_input_tokens_seen": 1863843840,
"step": 1580
},
{
"epoch": 0.0685383980181668,
"grad_norm": 0.86328125,
"learning_rate": 7.598831403722681e-05,
"loss": 1.6124,
"num_input_tokens_seen": 1865023488,
"step": 1581
},
{
"epoch": 0.06865636427981597,
"grad_norm": 0.70703125,
"learning_rate": 7.598150086042042e-05,
"loss": 1.6437,
"num_input_tokens_seen": 1866203136,
"step": 1582
},
{
"epoch": 0.06877433054146515,
"grad_norm": 0.8671875,
"learning_rate": 7.597468220897154e-05,
"loss": 1.4808,
"num_input_tokens_seen": 1867382784,
"step": 1583
},
{
"epoch": 0.0688922968031143,
"grad_norm": 0.83984375,
"learning_rate": 7.596785808391764e-05,
"loss": 1.6543,
"num_input_tokens_seen": 1868562432,
"step": 1584
},
{
"epoch": 0.06901026306476347,
"grad_norm": 0.8515625,
"learning_rate": 7.596102848629701e-05,
"loss": 1.5045,
"num_input_tokens_seen": 1869742080,
"step": 1585
},
{
"epoch": 0.06912822932641265,
"grad_norm": 0.78125,
"learning_rate": 7.595419341714879e-05,
"loss": 1.6585,
"num_input_tokens_seen": 1870921728,
"step": 1586
},
{
"epoch": 0.06924619558806182,
"grad_norm": 0.80078125,
"learning_rate": 7.594735287751293e-05,
"loss": 1.5746,
"num_input_tokens_seen": 1872101376,
"step": 1587
},
{
"epoch": 0.06936416184971098,
"grad_norm": 0.78125,
"learning_rate": 7.594050686843028e-05,
"loss": 1.5959,
"num_input_tokens_seen": 1873281024,
"step": 1588
},
{
"epoch": 0.06948212811136015,
"grad_norm": 0.8671875,
"learning_rate": 7.593365539094242e-05,
"loss": 1.4867,
"num_input_tokens_seen": 1874460672,
"step": 1589
},
{
"epoch": 0.06960009437300932,
"grad_norm": 0.875,
"learning_rate": 7.592679844609184e-05,
"loss": 1.5991,
"num_input_tokens_seen": 1875640320,
"step": 1590
},
{
"epoch": 0.06971806063465849,
"grad_norm": 0.7109375,
"learning_rate": 7.591993603492182e-05,
"loss": 1.7018,
"num_input_tokens_seen": 1876819968,
"step": 1591
},
{
"epoch": 0.06983602689630766,
"grad_norm": 0.77734375,
"learning_rate": 7.591306815847649e-05,
"loss": 1.5462,
"num_input_tokens_seen": 1877999616,
"step": 1592
},
{
"epoch": 0.06995399315795682,
"grad_norm": 0.84765625,
"learning_rate": 7.590619481780081e-05,
"loss": 1.4772,
"num_input_tokens_seen": 1879179264,
"step": 1593
},
{
"epoch": 0.07007195941960599,
"grad_norm": 0.765625,
"learning_rate": 7.589931601394057e-05,
"loss": 1.622,
"num_input_tokens_seen": 1880358912,
"step": 1594
},
{
"epoch": 0.07018992568125516,
"grad_norm": 0.8984375,
"learning_rate": 7.58924317479424e-05,
"loss": 1.7494,
"num_input_tokens_seen": 1881538560,
"step": 1595
},
{
"epoch": 0.07030789194290434,
"grad_norm": 0.828125,
"learning_rate": 7.588554202085375e-05,
"loss": 1.6065,
"num_input_tokens_seen": 1882718208,
"step": 1596
},
{
"epoch": 0.0704258582045535,
"grad_norm": 0.828125,
"learning_rate": 7.587864683372288e-05,
"loss": 1.6051,
"num_input_tokens_seen": 1883897856,
"step": 1597
},
{
"epoch": 0.07054382446620266,
"grad_norm": 0.828125,
"learning_rate": 7.587174618759893e-05,
"loss": 1.4877,
"num_input_tokens_seen": 1885077504,
"step": 1598
},
{
"epoch": 0.07066179072785184,
"grad_norm": 0.8125,
"learning_rate": 7.586484008353182e-05,
"loss": 1.519,
"num_input_tokens_seen": 1886257152,
"step": 1599
},
{
"epoch": 0.07077975698950101,
"grad_norm": 0.79296875,
"learning_rate": 7.585792852257236e-05,
"loss": 1.5906,
"num_input_tokens_seen": 1887436800,
"step": 1600
},
{
"epoch": 0.07077975698950101,
"eval_wikipedia_loss": 2.2380566596984863,
"eval_wikipedia_runtime": 161.8513,
"eval_wikipedia_samples_per_second": 4.337,
"eval_wikipedia_steps_per_second": 0.185,
"num_input_tokens_seen": 1887436800,
"step": 1600
},
{
"epoch": 0.07077975698950101,
"eval_toxicity_loss": 3.9978690147399902,
"eval_toxicity_runtime": 0.9108,
"eval_toxicity_samples_per_second": 2.196,
"eval_toxicity_steps_per_second": 1.098,
"num_input_tokens_seen": 1887436800,
"step": 1600
},
{
"epoch": 0.00011796626164916834,
"grad_norm": 0.69921875,
"learning_rate": 7.585101150577211e-05,
"loss": 1.6086,
"num_input_tokens_seen": 1888616448,
"step": 1601
},
{
"epoch": 0.0002359325232983367,
"grad_norm": 0.640625,
"learning_rate": 7.584408903418355e-05,
"loss": 1.4421,
"num_input_tokens_seen": 1889796096,
"step": 1602
},
{
"epoch": 0.000353898784947505,
"grad_norm": 0.69140625,
"learning_rate": 7.583716110885992e-05,
"loss": 1.3924,
"num_input_tokens_seen": 1890975744,
"step": 1603
},
{
"epoch": 0.0004718650465966734,
"grad_norm": 0.83203125,
"learning_rate": 7.583022773085532e-05,
"loss": 1.4947,
"num_input_tokens_seen": 1892155392,
"step": 1604
},
{
"epoch": 0.0005898313082458417,
"grad_norm": 0.79296875,
"learning_rate": 7.582328890122466e-05,
"loss": 1.5896,
"num_input_tokens_seen": 1893335040,
"step": 1605
},
{
"epoch": 0.00070779756989501,
"grad_norm": 0.73828125,
"learning_rate": 7.581634462102373e-05,
"loss": 1.4465,
"num_input_tokens_seen": 1894514688,
"step": 1606
},
{
"epoch": 0.0008257638315441783,
"grad_norm": 0.78515625,
"learning_rate": 7.580939489130906e-05,
"loss": 1.48,
"num_input_tokens_seen": 1895694336,
"step": 1607
},
{
"epoch": 0.0009437300931933467,
"grad_norm": 0.7109375,
"learning_rate": 7.580243971313811e-05,
"loss": 1.5052,
"num_input_tokens_seen": 1896873984,
"step": 1608
},
{
"epoch": 0.001061696354842515,
"grad_norm": 0.8046875,
"learning_rate": 7.579547908756911e-05,
"loss": 1.5373,
"num_input_tokens_seen": 1898053632,
"step": 1609
},
{
"epoch": 0.0011796626164916834,
"grad_norm": 0.828125,
"learning_rate": 7.578851301566112e-05,
"loss": 1.4559,
"num_input_tokens_seen": 1899233280,
"step": 1610
},
{
"epoch": 0.0012976288781408518,
"grad_norm": 1.015625,
"learning_rate": 7.578154149847404e-05,
"loss": 1.5937,
"num_input_tokens_seen": 1900412928,
"step": 1611
},
{
"epoch": 0.00141559513979002,
"grad_norm": 0.85546875,
"learning_rate": 7.57745645370686e-05,
"loss": 1.6112,
"num_input_tokens_seen": 1901592576,
"step": 1612
},
{
"epoch": 0.0015335614014391884,
"grad_norm": 0.83203125,
"learning_rate": 7.576758213250638e-05,
"loss": 1.4209,
"num_input_tokens_seen": 1902772224,
"step": 1613
},
{
"epoch": 0.0016515276630883566,
"grad_norm": 0.94921875,
"learning_rate": 7.576059428584972e-05,
"loss": 1.6171,
"num_input_tokens_seen": 1903951872,
"step": 1614
},
{
"epoch": 0.001769493924737525,
"grad_norm": 0.9453125,
"learning_rate": 7.575360099816185e-05,
"loss": 1.4181,
"num_input_tokens_seen": 1905131520,
"step": 1615
},
{
"epoch": 0.0018874601863866935,
"grad_norm": 1.0078125,
"learning_rate": 7.574660227050681e-05,
"loss": 1.529,
"num_input_tokens_seen": 1906311168,
"step": 1616
},
{
"epoch": 0.0020054264480358617,
"grad_norm": 0.91015625,
"learning_rate": 7.573959810394948e-05,
"loss": 1.6515,
"num_input_tokens_seen": 1907490816,
"step": 1617
},
{
"epoch": 0.00212339270968503,
"grad_norm": 1.015625,
"learning_rate": 7.573258849955555e-05,
"loss": 1.6023,
"num_input_tokens_seen": 1908670464,
"step": 1618
},
{
"epoch": 0.0022413589713341986,
"grad_norm": 0.890625,
"learning_rate": 7.572557345839153e-05,
"loss": 1.4418,
"num_input_tokens_seen": 1909850112,
"step": 1619
},
{
"epoch": 0.0023593252329833668,
"grad_norm": 1.0703125,
"learning_rate": 7.571855298152477e-05,
"loss": 1.5494,
"num_input_tokens_seen": 1911029760,
"step": 1620
},
{
"epoch": 0.002477291494632535,
"grad_norm": 0.84375,
"learning_rate": 7.571152707002347e-05,
"loss": 1.5324,
"num_input_tokens_seen": 1912209408,
"step": 1621
},
{
"epoch": 0.0025952577562817036,
"grad_norm": 1.0078125,
"learning_rate": 7.57044957249566e-05,
"loss": 1.5491,
"num_input_tokens_seen": 1913389056,
"step": 1622
},
{
"epoch": 0.002713224017930872,
"grad_norm": 0.81640625,
"learning_rate": 7.569745894739402e-05,
"loss": 1.5509,
"num_input_tokens_seen": 1914568704,
"step": 1623
},
{
"epoch": 0.00283119027958004,
"grad_norm": 0.97265625,
"learning_rate": 7.569041673840637e-05,
"loss": 1.5173,
"num_input_tokens_seen": 1915748352,
"step": 1624
},
{
"epoch": 0.0029491565412292082,
"grad_norm": 0.69921875,
"learning_rate": 7.568336909906514e-05,
"loss": 1.6629,
"num_input_tokens_seen": 1916928000,
"step": 1625
},
{
"epoch": 0.003067122802878377,
"grad_norm": 0.91015625,
"learning_rate": 7.567631603044264e-05,
"loss": 1.5456,
"num_input_tokens_seen": 1918107648,
"step": 1626
},
{
"epoch": 0.003185089064527545,
"grad_norm": 0.87890625,
"learning_rate": 7.5669257533612e-05,
"loss": 1.4868,
"num_input_tokens_seen": 1919287296,
"step": 1627
},
{
"epoch": 0.0033030553261767133,
"grad_norm": 0.8671875,
"learning_rate": 7.566219360964719e-05,
"loss": 1.475,
"num_input_tokens_seen": 1920466944,
"step": 1628
},
{
"epoch": 0.003421021587825882,
"grad_norm": 0.84375,
"learning_rate": 7.565512425962298e-05,
"loss": 1.4149,
"num_input_tokens_seen": 1921646592,
"step": 1629
},
{
"epoch": 0.00353898784947505,
"grad_norm": 0.84375,
"learning_rate": 7.5648049484615e-05,
"loss": 1.4973,
"num_input_tokens_seen": 1922826240,
"step": 1630
},
{
"epoch": 0.0036569541111242184,
"grad_norm": 0.859375,
"learning_rate": 7.564096928569969e-05,
"loss": 1.5576,
"num_input_tokens_seen": 1924005888,
"step": 1631
},
{
"epoch": 0.003774920372773387,
"grad_norm": 0.83984375,
"learning_rate": 7.56338836639543e-05,
"loss": 1.5025,
"num_input_tokens_seen": 1925185536,
"step": 1632
},
{
"epoch": 0.003892886634422555,
"grad_norm": 0.87109375,
"learning_rate": 7.562679262045692e-05,
"loss": 1.5112,
"num_input_tokens_seen": 1926365184,
"step": 1633
},
{
"epoch": 0.004010852896071723,
"grad_norm": 0.6875,
"learning_rate": 7.561969615628649e-05,
"loss": 1.6745,
"num_input_tokens_seen": 1927544832,
"step": 1634
},
{
"epoch": 0.004128819157720892,
"grad_norm": 0.85546875,
"learning_rate": 7.56125942725227e-05,
"loss": 1.5378,
"num_input_tokens_seen": 1928724480,
"step": 1635
},
{
"epoch": 0.00424678541937006,
"grad_norm": 0.7890625,
"learning_rate": 7.560548697024616e-05,
"loss": 1.4558,
"num_input_tokens_seen": 1929904128,
"step": 1636
},
{
"epoch": 0.004364751681019229,
"grad_norm": 0.8984375,
"learning_rate": 7.559837425053822e-05,
"loss": 1.5246,
"num_input_tokens_seen": 1931083776,
"step": 1637
},
{
"epoch": 0.004482717942668397,
"grad_norm": 0.7890625,
"learning_rate": 7.559125611448112e-05,
"loss": 1.6329,
"num_input_tokens_seen": 1932263424,
"step": 1638
},
{
"epoch": 0.004600684204317565,
"grad_norm": 0.8515625,
"learning_rate": 7.558413256315788e-05,
"loss": 1.6571,
"num_input_tokens_seen": 1933443072,
"step": 1639
},
{
"epoch": 0.0047186504659667335,
"grad_norm": 0.7109375,
"learning_rate": 7.557700359765238e-05,
"loss": 1.75,
"num_input_tokens_seen": 1934622720,
"step": 1640
},
{
"epoch": 0.004836616727615902,
"grad_norm": 0.78515625,
"learning_rate": 7.556986921904927e-05,
"loss": 1.7082,
"num_input_tokens_seen": 1935802368,
"step": 1641
},
{
"epoch": 0.00495458298926507,
"grad_norm": 0.75390625,
"learning_rate": 7.556272942843407e-05,
"loss": 1.6003,
"num_input_tokens_seen": 1936982016,
"step": 1642
},
{
"epoch": 0.005072549250914238,
"grad_norm": 0.80859375,
"learning_rate": 7.555558422689312e-05,
"loss": 1.6011,
"num_input_tokens_seen": 1938161664,
"step": 1643
},
{
"epoch": 0.005190515512563407,
"grad_norm": 0.7109375,
"learning_rate": 7.554843361551357e-05,
"loss": 1.5739,
"num_input_tokens_seen": 1939341312,
"step": 1644
},
{
"epoch": 0.0053084817742125754,
"grad_norm": 0.7890625,
"learning_rate": 7.55412775953834e-05,
"loss": 1.5394,
"num_input_tokens_seen": 1940520960,
"step": 1645
},
{
"epoch": 0.005426448035861744,
"grad_norm": 0.9609375,
"learning_rate": 7.553411616759141e-05,
"loss": 1.3772,
"num_input_tokens_seen": 1941700608,
"step": 1646
},
{
"epoch": 0.005544414297510912,
"grad_norm": 0.81640625,
"learning_rate": 7.55269493332272e-05,
"loss": 1.4369,
"num_input_tokens_seen": 1942880256,
"step": 1647
},
{
"epoch": 0.00566238055916008,
"grad_norm": 0.73828125,
"learning_rate": 7.551977709338125e-05,
"loss": 1.7093,
"num_input_tokens_seen": 1944059904,
"step": 1648
},
{
"epoch": 0.005780346820809248,
"grad_norm": 0.7734375,
"learning_rate": 7.55125994491448e-05,
"loss": 1.5003,
"num_input_tokens_seen": 1945239552,
"step": 1649
},
{
"epoch": 0.0058983130824584165,
"grad_norm": 0.81640625,
"learning_rate": 7.550541640160996e-05,
"loss": 1.6225,
"num_input_tokens_seen": 1946419200,
"step": 1650
},
{
"epoch": 0.0060162793441075856,
"grad_norm": 0.86328125,
"learning_rate": 7.549822795186963e-05,
"loss": 1.4609,
"num_input_tokens_seen": 1947598848,
"step": 1651
},
{
"epoch": 0.006134245605756754,
"grad_norm": 0.80859375,
"learning_rate": 7.549103410101754e-05,
"loss": 1.5634,
"num_input_tokens_seen": 1948778496,
"step": 1652
},
{
"epoch": 0.006252211867405922,
"grad_norm": 0.77734375,
"learning_rate": 7.548383485014826e-05,
"loss": 1.5238,
"num_input_tokens_seen": 1949958144,
"step": 1653
},
{
"epoch": 0.00637017812905509,
"grad_norm": 0.78125,
"learning_rate": 7.547663020035717e-05,
"loss": 1.6528,
"num_input_tokens_seen": 1951137792,
"step": 1654
},
{
"epoch": 0.006488144390704258,
"grad_norm": 0.765625,
"learning_rate": 7.546942015274046e-05,
"loss": 1.6072,
"num_input_tokens_seen": 1952317440,
"step": 1655
},
{
"epoch": 0.006606110652353427,
"grad_norm": 0.7890625,
"learning_rate": 7.546220470839512e-05,
"loss": 1.4707,
"num_input_tokens_seen": 1953497088,
"step": 1656
},
{
"epoch": 0.006724076914002596,
"grad_norm": 0.7109375,
"learning_rate": 7.545498386841904e-05,
"loss": 1.678,
"num_input_tokens_seen": 1954676736,
"step": 1657
},
{
"epoch": 0.006842043175651764,
"grad_norm": 0.92578125,
"learning_rate": 7.544775763391086e-05,
"loss": 1.5031,
"num_input_tokens_seen": 1955856384,
"step": 1658
},
{
"epoch": 0.006960009437300932,
"grad_norm": 0.75390625,
"learning_rate": 7.544052600597009e-05,
"loss": 1.4664,
"num_input_tokens_seen": 1957036032,
"step": 1659
},
{
"epoch": 0.0070779756989501,
"grad_norm": 1.0078125,
"learning_rate": 7.543328898569698e-05,
"loss": 1.5593,
"num_input_tokens_seen": 1958215680,
"step": 1660
},
{
"epoch": 0.0071959419605992685,
"grad_norm": 0.87109375,
"learning_rate": 7.542604657419268e-05,
"loss": 1.5318,
"num_input_tokens_seen": 1959395328,
"step": 1661
},
{
"epoch": 0.007313908222248437,
"grad_norm": 0.80859375,
"learning_rate": 7.541879877255915e-05,
"loss": 1.6062,
"num_input_tokens_seen": 1960574976,
"step": 1662
},
{
"epoch": 0.007431874483897605,
"grad_norm": 0.8125,
"learning_rate": 7.541154558189914e-05,
"loss": 1.5415,
"num_input_tokens_seen": 1961754624,
"step": 1663
},
{
"epoch": 0.007549840745546774,
"grad_norm": 0.80078125,
"learning_rate": 7.540428700331625e-05,
"loss": 1.4204,
"num_input_tokens_seen": 1962934272,
"step": 1664
},
{
"epoch": 0.007667807007195942,
"grad_norm": 0.76171875,
"learning_rate": 7.539702303791486e-05,
"loss": 1.5433,
"num_input_tokens_seen": 1964113920,
"step": 1665
},
{
"epoch": 0.00778577326884511,
"grad_norm": 0.80859375,
"learning_rate": 7.53897536868002e-05,
"loss": 1.6512,
"num_input_tokens_seen": 1965293568,
"step": 1666
},
{
"epoch": 0.007903739530494279,
"grad_norm": 0.81640625,
"learning_rate": 7.538247895107835e-05,
"loss": 1.5139,
"num_input_tokens_seen": 1966473216,
"step": 1667
},
{
"epoch": 0.008021705792143447,
"grad_norm": 0.74609375,
"learning_rate": 7.53751988318561e-05,
"loss": 1.7058,
"num_input_tokens_seen": 1967652864,
"step": 1668
},
{
"epoch": 0.008139672053792615,
"grad_norm": 0.828125,
"learning_rate": 7.53679133302412e-05,
"loss": 1.5792,
"num_input_tokens_seen": 1968832512,
"step": 1669
},
{
"epoch": 0.008257638315441783,
"grad_norm": 0.8046875,
"learning_rate": 7.536062244734212e-05,
"loss": 1.4524,
"num_input_tokens_seen": 1970012160,
"step": 1670
},
{
"epoch": 0.008375604577090951,
"grad_norm": 0.75390625,
"learning_rate": 7.535332618426816e-05,
"loss": 1.6209,
"num_input_tokens_seen": 1971191808,
"step": 1671
},
{
"epoch": 0.00849357083874012,
"grad_norm": 0.71484375,
"learning_rate": 7.53460245421295e-05,
"loss": 1.5364,
"num_input_tokens_seen": 1972371456,
"step": 1672
},
{
"epoch": 0.008611537100389288,
"grad_norm": 0.734375,
"learning_rate": 7.533871752203708e-05,
"loss": 1.6611,
"num_input_tokens_seen": 1973551104,
"step": 1673
},
{
"epoch": 0.008729503362038458,
"grad_norm": 0.82421875,
"learning_rate": 7.533140512510267e-05,
"loss": 1.5247,
"num_input_tokens_seen": 1974730752,
"step": 1674
},
{
"epoch": 0.008847469623687626,
"grad_norm": 0.703125,
"learning_rate": 7.532408735243887e-05,
"loss": 1.6105,
"num_input_tokens_seen": 1975910400,
"step": 1675
},
{
"epoch": 0.008965435885336794,
"grad_norm": 0.71484375,
"learning_rate": 7.531676420515908e-05,
"loss": 1.621,
"num_input_tokens_seen": 1977090048,
"step": 1676
},
{
"epoch": 0.009083402146985962,
"grad_norm": 0.71484375,
"learning_rate": 7.530943568437753e-05,
"loss": 1.5137,
"num_input_tokens_seen": 1978269696,
"step": 1677
},
{
"epoch": 0.00920136840863513,
"grad_norm": 0.7265625,
"learning_rate": 7.530210179120927e-05,
"loss": 1.4861,
"num_input_tokens_seen": 1979449344,
"step": 1678
},
{
"epoch": 0.009319334670284299,
"grad_norm": 0.74609375,
"learning_rate": 7.529476252677016e-05,
"loss": 1.6147,
"num_input_tokens_seen": 1980628992,
"step": 1679
},
{
"epoch": 0.009437300931933467,
"grad_norm": 0.8046875,
"learning_rate": 7.528741789217692e-05,
"loss": 1.506,
"num_input_tokens_seen": 1981808640,
"step": 1680
},
{
"epoch": 0.009555267193582635,
"grad_norm": 0.73046875,
"learning_rate": 7.528006788854697e-05,
"loss": 1.5949,
"num_input_tokens_seen": 1982988288,
"step": 1681
},
{
"epoch": 0.009673233455231803,
"grad_norm": 0.79296875,
"learning_rate": 7.527271251699867e-05,
"loss": 1.5086,
"num_input_tokens_seen": 1984167936,
"step": 1682
},
{
"epoch": 0.009791199716880972,
"grad_norm": 0.79296875,
"learning_rate": 7.526535177865118e-05,
"loss": 1.5301,
"num_input_tokens_seen": 1985347584,
"step": 1683
},
{
"epoch": 0.00990916597853014,
"grad_norm": 0.76953125,
"learning_rate": 7.525798567462439e-05,
"loss": 1.6647,
"num_input_tokens_seen": 1986527232,
"step": 1684
},
{
"epoch": 0.010027132240179308,
"grad_norm": 0.77734375,
"learning_rate": 7.525061420603911e-05,
"loss": 1.6048,
"num_input_tokens_seen": 1987706880,
"step": 1685
},
{
"epoch": 0.010145098501828476,
"grad_norm": 0.87109375,
"learning_rate": 7.524323737401688e-05,
"loss": 1.7124,
"num_input_tokens_seen": 1988886528,
"step": 1686
},
{
"epoch": 0.010263064763477646,
"grad_norm": 0.765625,
"learning_rate": 7.523585517968013e-05,
"loss": 1.6371,
"num_input_tokens_seen": 1990066176,
"step": 1687
},
{
"epoch": 0.010381031025126814,
"grad_norm": 0.79296875,
"learning_rate": 7.522846762415207e-05,
"loss": 1.5074,
"num_input_tokens_seen": 1991245824,
"step": 1688
},
{
"epoch": 0.010498997286775983,
"grad_norm": 0.734375,
"learning_rate": 7.52210747085567e-05,
"loss": 1.5505,
"num_input_tokens_seen": 1992425472,
"step": 1689
},
{
"epoch": 0.010616963548425151,
"grad_norm": 0.81640625,
"learning_rate": 7.521367643401889e-05,
"loss": 1.5206,
"num_input_tokens_seen": 1993605120,
"step": 1690
},
{
"epoch": 0.010734929810074319,
"grad_norm": 0.66796875,
"learning_rate": 7.52062728016643e-05,
"loss": 1.6137,
"num_input_tokens_seen": 1994784768,
"step": 1691
},
{
"epoch": 0.010852896071723487,
"grad_norm": 0.765625,
"learning_rate": 7.519886381261938e-05,
"loss": 1.5537,
"num_input_tokens_seen": 1995964416,
"step": 1692
},
{
"epoch": 0.010970862333372655,
"grad_norm": 0.74609375,
"learning_rate": 7.519144946801145e-05,
"loss": 1.5487,
"num_input_tokens_seen": 1997144064,
"step": 1693
},
{
"epoch": 0.011088828595021824,
"grad_norm": 0.80078125,
"learning_rate": 7.518402976896861e-05,
"loss": 1.5808,
"num_input_tokens_seen": 1998323712,
"step": 1694
},
{
"epoch": 0.011206794856670992,
"grad_norm": 0.7890625,
"learning_rate": 7.517660471661976e-05,
"loss": 1.5318,
"num_input_tokens_seen": 1999503360,
"step": 1695
},
{
"epoch": 0.01132476111832016,
"grad_norm": 0.703125,
"learning_rate": 7.516917431209462e-05,
"loss": 1.6044,
"num_input_tokens_seen": 2000683008,
"step": 1696
},
{
"epoch": 0.011442727379969328,
"grad_norm": 0.72265625,
"learning_rate": 7.51617385565238e-05,
"loss": 1.6691,
"num_input_tokens_seen": 2001862656,
"step": 1697
},
{
"epoch": 0.011560693641618497,
"grad_norm": 0.8359375,
"learning_rate": 7.51542974510386e-05,
"loss": 1.4109,
"num_input_tokens_seen": 2003042304,
"step": 1698
},
{
"epoch": 0.011678659903267665,
"grad_norm": 0.8359375,
"learning_rate": 7.514685099677122e-05,
"loss": 1.5029,
"num_input_tokens_seen": 2004221952,
"step": 1699
},
{
"epoch": 0.011796626164916833,
"grad_norm": 0.85546875,
"learning_rate": 7.513939919485466e-05,
"loss": 1.5524,
"num_input_tokens_seen": 2005401600,
"step": 1700
},
{
"epoch": 0.011914592426566003,
"grad_norm": 0.82421875,
"learning_rate": 7.51319420464227e-05,
"loss": 1.4823,
"num_input_tokens_seen": 2006581248,
"step": 1701
},
{
"epoch": 0.012032558688215171,
"grad_norm": 0.75390625,
"learning_rate": 7.512447955260998e-05,
"loss": 1.66,
"num_input_tokens_seen": 2007760896,
"step": 1702
},
{
"epoch": 0.01215052494986434,
"grad_norm": 0.80078125,
"learning_rate": 7.51170117145519e-05,
"loss": 1.6474,
"num_input_tokens_seen": 2008940544,
"step": 1703
},
{
"epoch": 0.012268491211513508,
"grad_norm": 0.6875,
"learning_rate": 7.510953853338474e-05,
"loss": 1.6331,
"num_input_tokens_seen": 2010120192,
"step": 1704
},
{
"epoch": 0.012386457473162676,
"grad_norm": 0.79296875,
"learning_rate": 7.510206001024554e-05,
"loss": 1.7039,
"num_input_tokens_seen": 2011299840,
"step": 1705
},
{
"epoch": 0.012504423734811844,
"grad_norm": 0.765625,
"learning_rate": 7.509457614627217e-05,
"loss": 1.5544,
"num_input_tokens_seen": 2012479488,
"step": 1706
},
{
"epoch": 0.012622389996461012,
"grad_norm": 0.765625,
"learning_rate": 7.50870869426033e-05,
"loss": 1.5711,
"num_input_tokens_seen": 2013659136,
"step": 1707
},
{
"epoch": 0.01274035625811018,
"grad_norm": 0.77734375,
"learning_rate": 7.507959240037844e-05,
"loss": 1.5601,
"num_input_tokens_seen": 2014838784,
"step": 1708
},
{
"epoch": 0.012858322519759349,
"grad_norm": 0.70703125,
"learning_rate": 7.50720925207379e-05,
"loss": 1.7537,
"num_input_tokens_seen": 2016018432,
"step": 1709
},
{
"epoch": 0.012976288781408517,
"grad_norm": 0.7421875,
"learning_rate": 7.506458730482277e-05,
"loss": 1.5504,
"num_input_tokens_seen": 2017198080,
"step": 1710
},
{
"epoch": 0.013094255043057685,
"grad_norm": 0.71484375,
"learning_rate": 7.505707675377502e-05,
"loss": 1.6148,
"num_input_tokens_seen": 2018377728,
"step": 1711
},
{
"epoch": 0.013212221304706853,
"grad_norm": 0.71484375,
"learning_rate": 7.504956086873735e-05,
"loss": 1.541,
"num_input_tokens_seen": 2019557376,
"step": 1712
},
{
"epoch": 0.013330187566356021,
"grad_norm": 0.765625,
"learning_rate": 7.504203965085335e-05,
"loss": 1.6243,
"num_input_tokens_seen": 2020737024,
"step": 1713
},
{
"epoch": 0.013448153828005191,
"grad_norm": 0.796875,
"learning_rate": 7.503451310126738e-05,
"loss": 1.4932,
"num_input_tokens_seen": 2021916672,
"step": 1714
},
{
"epoch": 0.01356612008965436,
"grad_norm": 0.765625,
"learning_rate": 7.50269812211246e-05,
"loss": 1.7139,
"num_input_tokens_seen": 2023096320,
"step": 1715
},
{
"epoch": 0.013684086351303528,
"grad_norm": 0.8515625,
"learning_rate": 7.5019444011571e-05,
"loss": 1.5194,
"num_input_tokens_seen": 2024275968,
"step": 1716
},
{
"epoch": 0.013802052612952696,
"grad_norm": 0.76171875,
"learning_rate": 7.501190147375338e-05,
"loss": 1.6824,
"num_input_tokens_seen": 2025455616,
"step": 1717
},
{
"epoch": 0.013920018874601864,
"grad_norm": 0.73046875,
"learning_rate": 7.500435360881937e-05,
"loss": 1.6044,
"num_input_tokens_seen": 2026635264,
"step": 1718
},
{
"epoch": 0.014037985136251032,
"grad_norm": 0.66796875,
"learning_rate": 7.499680041791737e-05,
"loss": 1.5265,
"num_input_tokens_seen": 2027814912,
"step": 1719
},
{
"epoch": 0.0141559513979002,
"grad_norm": 0.82421875,
"learning_rate": 7.49892419021966e-05,
"loss": 1.5371,
"num_input_tokens_seen": 2028994560,
"step": 1720
},
{
"epoch": 0.014273917659549369,
"grad_norm": 0.7109375,
"learning_rate": 7.498167806280712e-05,
"loss": 1.5835,
"num_input_tokens_seen": 2030174208,
"step": 1721
},
{
"epoch": 0.014391883921198537,
"grad_norm": 0.7109375,
"learning_rate": 7.497410890089976e-05,
"loss": 1.5378,
"num_input_tokens_seen": 2031353856,
"step": 1722
},
{
"epoch": 0.014509850182847705,
"grad_norm": 0.69140625,
"learning_rate": 7.496653441762621e-05,
"loss": 1.6016,
"num_input_tokens_seen": 2032533504,
"step": 1723
},
{
"epoch": 0.014627816444496873,
"grad_norm": 0.7421875,
"learning_rate": 7.495895461413891e-05,
"loss": 1.5343,
"num_input_tokens_seen": 2033713152,
"step": 1724
},
{
"epoch": 0.014745782706146042,
"grad_norm": 0.71484375,
"learning_rate": 7.495136949159117e-05,
"loss": 1.6498,
"num_input_tokens_seen": 2034892800,
"step": 1725
},
{
"epoch": 0.01486374896779521,
"grad_norm": 0.80859375,
"learning_rate": 7.494377905113704e-05,
"loss": 1.4332,
"num_input_tokens_seen": 2036072448,
"step": 1726
},
{
"epoch": 0.014981715229444378,
"grad_norm": 0.66796875,
"learning_rate": 7.493618329393145e-05,
"loss": 1.6878,
"num_input_tokens_seen": 2037252096,
"step": 1727
},
{
"epoch": 0.015099681491093548,
"grad_norm": 0.69921875,
"learning_rate": 7.49285822211301e-05,
"loss": 1.557,
"num_input_tokens_seen": 2038431744,
"step": 1728
},
{
"epoch": 0.015217647752742716,
"grad_norm": 0.7109375,
"learning_rate": 7.492097583388949e-05,
"loss": 1.6027,
"num_input_tokens_seen": 2039611392,
"step": 1729
},
{
"epoch": 0.015335614014391884,
"grad_norm": 0.73828125,
"learning_rate": 7.491336413336695e-05,
"loss": 1.5635,
"num_input_tokens_seen": 2040791040,
"step": 1730
},
{
"epoch": 0.015453580276041053,
"grad_norm": 0.7890625,
"learning_rate": 7.490574712072063e-05,
"loss": 1.5244,
"num_input_tokens_seen": 2041970688,
"step": 1731
},
{
"epoch": 0.01557154653769022,
"grad_norm": 0.75,
"learning_rate": 7.489812479710944e-05,
"loss": 1.4318,
"num_input_tokens_seen": 2043150336,
"step": 1732
},
{
"epoch": 0.01568951279933939,
"grad_norm": 0.7265625,
"learning_rate": 7.489049716369316e-05,
"loss": 1.5501,
"num_input_tokens_seen": 2044329984,
"step": 1733
},
{
"epoch": 0.015807479060988557,
"grad_norm": 0.62890625,
"learning_rate": 7.488286422163232e-05,
"loss": 1.6179,
"num_input_tokens_seen": 2045509632,
"step": 1734
},
{
"epoch": 0.015925445322637725,
"grad_norm": 0.6796875,
"learning_rate": 7.48752259720883e-05,
"loss": 1.5711,
"num_input_tokens_seen": 2046689280,
"step": 1735
},
{
"epoch": 0.016043411584286894,
"grad_norm": 0.71875,
"learning_rate": 7.486758241622327e-05,
"loss": 1.5267,
"num_input_tokens_seen": 2047868928,
"step": 1736
},
{
"epoch": 0.016161377845936062,
"grad_norm": 0.72265625,
"learning_rate": 7.48599335552002e-05,
"loss": 1.622,
"num_input_tokens_seen": 2049048576,
"step": 1737
},
{
"epoch": 0.01627934410758523,
"grad_norm": 0.98828125,
"learning_rate": 7.485227939018287e-05,
"loss": 1.5862,
"num_input_tokens_seen": 2050228224,
"step": 1738
},
{
"epoch": 0.0163973103692344,
"grad_norm": 0.78125,
"learning_rate": 7.484461992233592e-05,
"loss": 1.5703,
"num_input_tokens_seen": 2051407872,
"step": 1739
},
{
"epoch": 0.016515276630883566,
"grad_norm": 0.80078125,
"learning_rate": 7.48369551528247e-05,
"loss": 1.6746,
"num_input_tokens_seen": 2052587520,
"step": 1740
},
{
"epoch": 0.016633242892532735,
"grad_norm": 0.85546875,
"learning_rate": 7.482928508281543e-05,
"loss": 1.4853,
"num_input_tokens_seen": 2053767168,
"step": 1741
},
{
"epoch": 0.016751209154181903,
"grad_norm": 0.7265625,
"learning_rate": 7.482160971347514e-05,
"loss": 1.5609,
"num_input_tokens_seen": 2054946816,
"step": 1742
},
{
"epoch": 0.01686917541583107,
"grad_norm": 0.796875,
"learning_rate": 7.481392904597161e-05,
"loss": 1.4062,
"num_input_tokens_seen": 2056126464,
"step": 1743
},
{
"epoch": 0.01698714167748024,
"grad_norm": 0.6796875,
"learning_rate": 7.480624308147352e-05,
"loss": 1.6638,
"num_input_tokens_seen": 2057306112,
"step": 1744
},
{
"epoch": 0.017105107939129408,
"grad_norm": 0.84375,
"learning_rate": 7.479855182115027e-05,
"loss": 1.4458,
"num_input_tokens_seen": 2058485760,
"step": 1745
},
{
"epoch": 0.017223074200778576,
"grad_norm": 0.65625,
"learning_rate": 7.479085526617209e-05,
"loss": 1.7725,
"num_input_tokens_seen": 2059665408,
"step": 1746
},
{
"epoch": 0.017341040462427744,
"grad_norm": 0.80078125,
"learning_rate": 7.478315341771003e-05,
"loss": 1.5117,
"num_input_tokens_seen": 2060845056,
"step": 1747
},
{
"epoch": 0.017459006724076916,
"grad_norm": 0.68359375,
"learning_rate": 7.477544627693595e-05,
"loss": 1.5183,
"num_input_tokens_seen": 2062024704,
"step": 1748
},
{
"epoch": 0.017576972985726084,
"grad_norm": 0.828125,
"learning_rate": 7.476773384502251e-05,
"loss": 1.5633,
"num_input_tokens_seen": 2063204352,
"step": 1749
},
{
"epoch": 0.017694939247375252,
"grad_norm": 0.66796875,
"learning_rate": 7.476001612314314e-05,
"loss": 1.5814,
"num_input_tokens_seen": 2064384000,
"step": 1750
},
{
"epoch": 0.01781290550902442,
"grad_norm": 0.80859375,
"learning_rate": 7.475229311247212e-05,
"loss": 1.5078,
"num_input_tokens_seen": 2065563648,
"step": 1751
},
{
"epoch": 0.01793087177067359,
"grad_norm": 0.7421875,
"learning_rate": 7.474456481418452e-05,
"loss": 1.5594,
"num_input_tokens_seen": 2066743296,
"step": 1752
},
{
"epoch": 0.018048838032322757,
"grad_norm": 0.83203125,
"learning_rate": 7.473683122945622e-05,
"loss": 1.5369,
"num_input_tokens_seen": 2067922944,
"step": 1753
},
{
"epoch": 0.018166804293971925,
"grad_norm": 0.69921875,
"learning_rate": 7.472909235946388e-05,
"loss": 1.5476,
"num_input_tokens_seen": 2069102592,
"step": 1754
},
{
"epoch": 0.018284770555621093,
"grad_norm": 0.71875,
"learning_rate": 7.472134820538498e-05,
"loss": 1.525,
"num_input_tokens_seen": 2070282240,
"step": 1755
},
{
"epoch": 0.01840273681727026,
"grad_norm": 0.75,
"learning_rate": 7.471359876839782e-05,
"loss": 1.6491,
"num_input_tokens_seen": 2071461888,
"step": 1756
},
{
"epoch": 0.01852070307891943,
"grad_norm": 0.7109375,
"learning_rate": 7.470584404968149e-05,
"loss": 1.5732,
"num_input_tokens_seen": 2072641536,
"step": 1757
},
{
"epoch": 0.018638669340568598,
"grad_norm": 0.71875,
"learning_rate": 7.469808405041587e-05,
"loss": 1.5811,
"num_input_tokens_seen": 2073821184,
"step": 1758
},
{
"epoch": 0.018756635602217766,
"grad_norm": 0.65625,
"learning_rate": 7.469031877178167e-05,
"loss": 1.6869,
"num_input_tokens_seen": 2075000832,
"step": 1759
},
{
"epoch": 0.018874601863866934,
"grad_norm": 0.64453125,
"learning_rate": 7.468254821496037e-05,
"loss": 1.7133,
"num_input_tokens_seen": 2076180480,
"step": 1760
},
{
"epoch": 0.018992568125516102,
"grad_norm": 0.6953125,
"learning_rate": 7.46747723811343e-05,
"loss": 1.5979,
"num_input_tokens_seen": 2077360128,
"step": 1761
},
{
"epoch": 0.01911053438716527,
"grad_norm": 0.64453125,
"learning_rate": 7.466699127148653e-05,
"loss": 1.6148,
"num_input_tokens_seen": 2078539776,
"step": 1762
},
{
"epoch": 0.01922850064881444,
"grad_norm": 0.67578125,
"learning_rate": 7.465920488720099e-05,
"loss": 1.5162,
"num_input_tokens_seen": 2079719424,
"step": 1763
},
{
"epoch": 0.019346466910463607,
"grad_norm": 0.7265625,
"learning_rate": 7.465141322946241e-05,
"loss": 1.4266,
"num_input_tokens_seen": 2080899072,
"step": 1764
},
{
"epoch": 0.019464433172112775,
"grad_norm": 0.703125,
"learning_rate": 7.464361629945625e-05,
"loss": 1.6248,
"num_input_tokens_seen": 2082078720,
"step": 1765
},
{
"epoch": 0.019582399433761943,
"grad_norm": 0.70703125,
"learning_rate": 7.463581409836888e-05,
"loss": 1.4722,
"num_input_tokens_seen": 2083258368,
"step": 1766
},
{
"epoch": 0.01970036569541111,
"grad_norm": 0.73046875,
"learning_rate": 7.462800662738739e-05,
"loss": 1.5744,
"num_input_tokens_seen": 2084438016,
"step": 1767
},
{
"epoch": 0.01981833195706028,
"grad_norm": 0.703125,
"learning_rate": 7.46201938876997e-05,
"loss": 1.5756,
"num_input_tokens_seen": 2085617664,
"step": 1768
},
{
"epoch": 0.019936298218709448,
"grad_norm": 0.68359375,
"learning_rate": 7.461237588049453e-05,
"loss": 1.4826,
"num_input_tokens_seen": 2086797312,
"step": 1769
},
{
"epoch": 0.020054264480358616,
"grad_norm": 0.74609375,
"learning_rate": 7.460455260696142e-05,
"loss": 1.7039,
"num_input_tokens_seen": 2087976960,
"step": 1770
},
{
"epoch": 0.020172230742007784,
"grad_norm": 0.75390625,
"learning_rate": 7.459672406829068e-05,
"loss": 1.4206,
"num_input_tokens_seen": 2089156608,
"step": 1771
},
{
"epoch": 0.020290197003656953,
"grad_norm": 0.71875,
"learning_rate": 7.458889026567342e-05,
"loss": 1.6331,
"num_input_tokens_seen": 2090336256,
"step": 1772
},
{
"epoch": 0.02040816326530612,
"grad_norm": 0.7421875,
"learning_rate": 7.458105120030159e-05,
"loss": 1.5279,
"num_input_tokens_seen": 2091515904,
"step": 1773
},
{
"epoch": 0.020526129526955292,
"grad_norm": 0.6640625,
"learning_rate": 7.457320687336792e-05,
"loss": 1.6853,
"num_input_tokens_seen": 2092695552,
"step": 1774
},
{
"epoch": 0.02064409578860446,
"grad_norm": 0.80859375,
"learning_rate": 7.45653572860659e-05,
"loss": 1.5579,
"num_input_tokens_seen": 2093875200,
"step": 1775
},
{
"epoch": 0.02076206205025363,
"grad_norm": 0.6953125,
"learning_rate": 7.455750243958989e-05,
"loss": 1.6844,
"num_input_tokens_seen": 2095054848,
"step": 1776
},
{
"epoch": 0.020880028311902797,
"grad_norm": 0.76953125,
"learning_rate": 7.4549642335135e-05,
"loss": 1.6191,
"num_input_tokens_seen": 2096234496,
"step": 1777
},
{
"epoch": 0.020997994573551965,
"grad_norm": 0.83984375,
"learning_rate": 7.454177697389717e-05,
"loss": 1.5507,
"num_input_tokens_seen": 2097414144,
"step": 1778
},
{
"epoch": 0.021115960835201134,
"grad_norm": 0.82421875,
"learning_rate": 7.453390635707312e-05,
"loss": 1.4706,
"num_input_tokens_seen": 2098593792,
"step": 1779
},
{
"epoch": 0.021233927096850302,
"grad_norm": 0.6484375,
"learning_rate": 7.452603048586039e-05,
"loss": 1.5721,
"num_input_tokens_seen": 2099773440,
"step": 1780
},
{
"epoch": 0.02135189335849947,
"grad_norm": 0.8671875,
"learning_rate": 7.451814936145728e-05,
"loss": 1.344,
"num_input_tokens_seen": 2100953088,
"step": 1781
},
{
"epoch": 0.021469859620148638,
"grad_norm": 0.6953125,
"learning_rate": 7.451026298506294e-05,
"loss": 1.5718,
"num_input_tokens_seen": 2102132736,
"step": 1782
},
{
"epoch": 0.021587825881797806,
"grad_norm": 0.79296875,
"learning_rate": 7.450237135787728e-05,
"loss": 1.5887,
"num_input_tokens_seen": 2103312384,
"step": 1783
},
{
"epoch": 0.021705792143446975,
"grad_norm": 0.77734375,
"learning_rate": 7.449447448110104e-05,
"loss": 1.5317,
"num_input_tokens_seen": 2104492032,
"step": 1784
},
{
"epoch": 0.021823758405096143,
"grad_norm": 0.78515625,
"learning_rate": 7.448657235593572e-05,
"loss": 1.645,
"num_input_tokens_seen": 2105671680,
"step": 1785
},
{
"epoch": 0.02194172466674531,
"grad_norm": 0.76953125,
"learning_rate": 7.447866498358366e-05,
"loss": 1.533,
"num_input_tokens_seen": 2106851328,
"step": 1786
},
{
"epoch": 0.02205969092839448,
"grad_norm": 0.71875,
"learning_rate": 7.447075236524797e-05,
"loss": 1.6696,
"num_input_tokens_seen": 2108030976,
"step": 1787
},
{
"epoch": 0.022177657190043647,
"grad_norm": 0.87890625,
"learning_rate": 7.446283450213256e-05,
"loss": 1.4246,
"num_input_tokens_seen": 2109210624,
"step": 1788
},
{
"epoch": 0.022295623451692816,
"grad_norm": 0.703125,
"learning_rate": 7.445491139544218e-05,
"loss": 1.6228,
"num_input_tokens_seen": 2110390272,
"step": 1789
},
{
"epoch": 0.022413589713341984,
"grad_norm": 0.828125,
"learning_rate": 7.444698304638229e-05,
"loss": 1.6991,
"num_input_tokens_seen": 2111569920,
"step": 1790
},
{
"epoch": 0.022531555974991152,
"grad_norm": 0.8359375,
"learning_rate": 7.443904945615924e-05,
"loss": 1.5197,
"num_input_tokens_seen": 2112749568,
"step": 1791
},
{
"epoch": 0.02264952223664032,
"grad_norm": 0.7890625,
"learning_rate": 7.443111062598013e-05,
"loss": 1.5836,
"num_input_tokens_seen": 2113929216,
"step": 1792
},
{
"epoch": 0.02276748849828949,
"grad_norm": 0.8203125,
"learning_rate": 7.442316655705285e-05,
"loss": 1.6029,
"num_input_tokens_seen": 2115108864,
"step": 1793
},
{
"epoch": 0.022885454759938657,
"grad_norm": 0.8125,
"learning_rate": 7.441521725058613e-05,
"loss": 1.6011,
"num_input_tokens_seen": 2116288512,
"step": 1794
},
{
"epoch": 0.023003421021587825,
"grad_norm": 0.75390625,
"learning_rate": 7.440726270778945e-05,
"loss": 1.6227,
"num_input_tokens_seen": 2117468160,
"step": 1795
},
{
"epoch": 0.023121387283236993,
"grad_norm": 0.8046875,
"learning_rate": 7.43993029298731e-05,
"loss": 1.4413,
"num_input_tokens_seen": 2118647808,
"step": 1796
},
{
"epoch": 0.02323935354488616,
"grad_norm": 0.69140625,
"learning_rate": 7.43913379180482e-05,
"loss": 1.5052,
"num_input_tokens_seen": 2119827456,
"step": 1797
},
{
"epoch": 0.02335731980653533,
"grad_norm": 0.71484375,
"learning_rate": 7.438336767352662e-05,
"loss": 1.5939,
"num_input_tokens_seen": 2121007104,
"step": 1798
},
{
"epoch": 0.023475286068184498,
"grad_norm": 0.68359375,
"learning_rate": 7.437539219752105e-05,
"loss": 1.649,
"num_input_tokens_seen": 2122186752,
"step": 1799
},
{
"epoch": 0.023593252329833666,
"grad_norm": 0.76953125,
"learning_rate": 7.436741149124496e-05,
"loss": 1.5859,
"num_input_tokens_seen": 2123366400,
"step": 1800
},
{
"epoch": 0.023593252329833666,
"eval_wikipedia_loss": 2.2597548961639404,
"eval_wikipedia_runtime": 162.0396,
"eval_wikipedia_samples_per_second": 4.332,
"eval_wikipedia_steps_per_second": 0.185,
"num_input_tokens_seen": 2123366400,
"step": 1800
},
{
"epoch": 0.023593252329833666,
"eval_toxicity_loss": 4.024182319641113,
"eval_toxicity_runtime": 0.9591,
"eval_toxicity_samples_per_second": 2.085,
"eval_toxicity_steps_per_second": 1.043,
"num_input_tokens_seen": 2123366400,
"step": 1800
},
{
"epoch": 0.023711218591482838,
"grad_norm": 0.73046875,
"learning_rate": 7.435942555591265e-05,
"loss": 1.5798,
"num_input_tokens_seen": 2124546048,
"step": 1801
},
{
"epoch": 0.023829184853132006,
"grad_norm": 0.8203125,
"learning_rate": 7.435143439273918e-05,
"loss": 1.6395,
"num_input_tokens_seen": 2125725696,
"step": 1802
},
{
"epoch": 0.023947151114781174,
"grad_norm": 0.7421875,
"learning_rate": 7.434343800294041e-05,
"loss": 1.6122,
"num_input_tokens_seen": 2126905344,
"step": 1803
},
{
"epoch": 0.024065117376430342,
"grad_norm": 0.76171875,
"learning_rate": 7.433543638773302e-05,
"loss": 1.5967,
"num_input_tokens_seen": 2128084992,
"step": 1804
},
{
"epoch": 0.02418308363807951,
"grad_norm": 0.7734375,
"learning_rate": 7.432742954833445e-05,
"loss": 1.5463,
"num_input_tokens_seen": 2129264640,
"step": 1805
},
{
"epoch": 0.02430104989972868,
"grad_norm": 0.76171875,
"learning_rate": 7.431941748596297e-05,
"loss": 1.5567,
"num_input_tokens_seen": 2130444288,
"step": 1806
},
{
"epoch": 0.024419016161377847,
"grad_norm": 0.70703125,
"learning_rate": 7.431140020183763e-05,
"loss": 1.5494,
"num_input_tokens_seen": 2131623936,
"step": 1807
},
{
"epoch": 0.024536982423027015,
"grad_norm": 0.76171875,
"learning_rate": 7.430337769717824e-05,
"loss": 1.6508,
"num_input_tokens_seen": 2132803584,
"step": 1808
},
{
"epoch": 0.024654948684676183,
"grad_norm": 0.65234375,
"learning_rate": 7.429534997320546e-05,
"loss": 1.7077,
"num_input_tokens_seen": 2133983232,
"step": 1809
},
{
"epoch": 0.02477291494632535,
"grad_norm": 0.8046875,
"learning_rate": 7.428731703114075e-05,
"loss": 1.4564,
"num_input_tokens_seen": 2135162880,
"step": 1810
},
{
"epoch": 0.02489088120797452,
"grad_norm": 0.68359375,
"learning_rate": 7.427927887220629e-05,
"loss": 1.6478,
"num_input_tokens_seen": 2136342528,
"step": 1811
},
{
"epoch": 0.025008847469623688,
"grad_norm": 0.6953125,
"learning_rate": 7.427123549762511e-05,
"loss": 1.6442,
"num_input_tokens_seen": 2137522176,
"step": 1812
},
{
"epoch": 0.025126813731272856,
"grad_norm": 0.77734375,
"learning_rate": 7.426318690862104e-05,
"loss": 1.4693,
"num_input_tokens_seen": 2138701824,
"step": 1813
},
{
"epoch": 0.025244779992922024,
"grad_norm": 0.69140625,
"learning_rate": 7.425513310641867e-05,
"loss": 1.5982,
"num_input_tokens_seen": 2139881472,
"step": 1814
},
{
"epoch": 0.025362746254571193,
"grad_norm": 0.7421875,
"learning_rate": 7.42470740922434e-05,
"loss": 1.4796,
"num_input_tokens_seen": 2141061120,
"step": 1815
},
{
"epoch": 0.02548071251622036,
"grad_norm": 0.8203125,
"learning_rate": 7.423900986732143e-05,
"loss": 1.4444,
"num_input_tokens_seen": 2142240768,
"step": 1816
},
{
"epoch": 0.02559867877786953,
"grad_norm": 0.6640625,
"learning_rate": 7.423094043287974e-05,
"loss": 1.5656,
"num_input_tokens_seen": 2143420416,
"step": 1817
},
{
"epoch": 0.025716645039518697,
"grad_norm": 0.78515625,
"learning_rate": 7.422286579014609e-05,
"loss": 1.4905,
"num_input_tokens_seen": 2144600064,
"step": 1818
},
{
"epoch": 0.025834611301167865,
"grad_norm": 0.68359375,
"learning_rate": 7.421478594034907e-05,
"loss": 1.688,
"num_input_tokens_seen": 2145779712,
"step": 1819
},
{
"epoch": 0.025952577562817034,
"grad_norm": 0.82421875,
"learning_rate": 7.420670088471803e-05,
"loss": 1.5098,
"num_input_tokens_seen": 2146959360,
"step": 1820
},
{
"epoch": 0.026070543824466202,
"grad_norm": 0.7109375,
"learning_rate": 7.419861062448314e-05,
"loss": 1.5118,
"num_input_tokens_seen": 2148139008,
"step": 1821
},
{
"epoch": 0.02618851008611537,
"grad_norm": 0.7734375,
"learning_rate": 7.419051516087535e-05,
"loss": 1.5599,
"num_input_tokens_seen": 2149318656,
"step": 1822
},
{
"epoch": 0.026306476347764538,
"grad_norm": 0.72265625,
"learning_rate": 7.418241449512638e-05,
"loss": 1.5937,
"num_input_tokens_seen": 2150498304,
"step": 1823
},
{
"epoch": 0.026424442609413706,
"grad_norm": 0.72265625,
"learning_rate": 7.417430862846875e-05,
"loss": 1.5222,
"num_input_tokens_seen": 2151677952,
"step": 1824
},
{
"epoch": 0.026542408871062875,
"grad_norm": 0.76171875,
"learning_rate": 7.416619756213581e-05,
"loss": 1.5716,
"num_input_tokens_seen": 2152857600,
"step": 1825
},
{
"epoch": 0.026660375132712043,
"grad_norm": 0.74609375,
"learning_rate": 7.415808129736164e-05,
"loss": 1.5728,
"num_input_tokens_seen": 2154037248,
"step": 1826
},
{
"epoch": 0.02677834139436121,
"grad_norm": 0.6484375,
"learning_rate": 7.414995983538116e-05,
"loss": 1.7184,
"num_input_tokens_seen": 2155216896,
"step": 1827
},
{
"epoch": 0.026896307656010383,
"grad_norm": 0.72265625,
"learning_rate": 7.414183317743008e-05,
"loss": 1.5229,
"num_input_tokens_seen": 2156396544,
"step": 1828
},
{
"epoch": 0.02701427391765955,
"grad_norm": 0.6875,
"learning_rate": 7.413370132474485e-05,
"loss": 1.5489,
"num_input_tokens_seen": 2157576192,
"step": 1829
},
{
"epoch": 0.02713224017930872,
"grad_norm": 0.6484375,
"learning_rate": 7.412556427856276e-05,
"loss": 1.642,
"num_input_tokens_seen": 2158755840,
"step": 1830
},
{
"epoch": 0.027250206440957887,
"grad_norm": 0.6640625,
"learning_rate": 7.411742204012188e-05,
"loss": 1.6251,
"num_input_tokens_seen": 2159935488,
"step": 1831
},
{
"epoch": 0.027368172702607056,
"grad_norm": 0.73046875,
"learning_rate": 7.410927461066107e-05,
"loss": 1.5202,
"num_input_tokens_seen": 2161115136,
"step": 1832
},
{
"epoch": 0.027486138964256224,
"grad_norm": 0.82421875,
"learning_rate": 7.410112199141994e-05,
"loss": 1.5337,
"num_input_tokens_seen": 2162294784,
"step": 1833
},
{
"epoch": 0.027604105225905392,
"grad_norm": 0.80859375,
"learning_rate": 7.409296418363897e-05,
"loss": 1.4675,
"num_input_tokens_seen": 2163474432,
"step": 1834
},
{
"epoch": 0.02772207148755456,
"grad_norm": 0.7109375,
"learning_rate": 7.408480118855935e-05,
"loss": 1.6147,
"num_input_tokens_seen": 2164654080,
"step": 1835
},
{
"epoch": 0.02784003774920373,
"grad_norm": 0.95703125,
"learning_rate": 7.407663300742309e-05,
"loss": 1.4314,
"num_input_tokens_seen": 2165833728,
"step": 1836
},
{
"epoch": 0.027958004010852897,
"grad_norm": 0.80078125,
"learning_rate": 7.406845964147303e-05,
"loss": 1.5776,
"num_input_tokens_seen": 2167013376,
"step": 1837
},
{
"epoch": 0.028075970272502065,
"grad_norm": 0.78125,
"learning_rate": 7.406028109195273e-05,
"loss": 1.6119,
"num_input_tokens_seen": 2168193024,
"step": 1838
},
{
"epoch": 0.028193936534151233,
"grad_norm": 0.828125,
"learning_rate": 7.405209736010656e-05,
"loss": 1.5874,
"num_input_tokens_seen": 2169372672,
"step": 1839
},
{
"epoch": 0.0283119027958004,
"grad_norm": 0.75,
"learning_rate": 7.404390844717971e-05,
"loss": 1.583,
"num_input_tokens_seen": 2170552320,
"step": 1840
},
{
"epoch": 0.02842986905744957,
"grad_norm": 0.6328125,
"learning_rate": 7.403571435441814e-05,
"loss": 1.6481,
"num_input_tokens_seen": 2171731968,
"step": 1841
},
{
"epoch": 0.028547835319098738,
"grad_norm": 0.71875,
"learning_rate": 7.402751508306858e-05,
"loss": 1.5372,
"num_input_tokens_seen": 2172911616,
"step": 1842
},
{
"epoch": 0.028665801580747906,
"grad_norm": 0.6484375,
"learning_rate": 7.401931063437855e-05,
"loss": 1.5178,
"num_input_tokens_seen": 2174091264,
"step": 1843
},
{
"epoch": 0.028783767842397074,
"grad_norm": 0.68359375,
"learning_rate": 7.40111010095964e-05,
"loss": 1.5221,
"num_input_tokens_seen": 2175270912,
"step": 1844
},
{
"epoch": 0.028901734104046242,
"grad_norm": 0.65625,
"learning_rate": 7.400288620997123e-05,
"loss": 1.7355,
"num_input_tokens_seen": 2176450560,
"step": 1845
},
{
"epoch": 0.02901970036569541,
"grad_norm": 0.68359375,
"learning_rate": 7.399466623675292e-05,
"loss": 1.5432,
"num_input_tokens_seen": 2177630208,
"step": 1846
},
{
"epoch": 0.02913766662734458,
"grad_norm": 0.6796875,
"learning_rate": 7.398644109119214e-05,
"loss": 1.4819,
"num_input_tokens_seen": 2178809856,
"step": 1847
},
{
"epoch": 0.029255632888993747,
"grad_norm": 0.703125,
"learning_rate": 7.39782107745404e-05,
"loss": 1.5471,
"num_input_tokens_seen": 2179989504,
"step": 1848
},
{
"epoch": 0.029373599150642915,
"grad_norm": 0.7265625,
"learning_rate": 7.396997528804994e-05,
"loss": 1.6242,
"num_input_tokens_seen": 2181169152,
"step": 1849
},
{
"epoch": 0.029491565412292083,
"grad_norm": 0.6328125,
"learning_rate": 7.396173463297379e-05,
"loss": 1.5332,
"num_input_tokens_seen": 2182348800,
"step": 1850
},
{
"epoch": 0.02960953167394125,
"grad_norm": 0.66796875,
"learning_rate": 7.395348881056578e-05,
"loss": 1.4848,
"num_input_tokens_seen": 2183528448,
"step": 1851
},
{
"epoch": 0.02972749793559042,
"grad_norm": 0.71484375,
"learning_rate": 7.394523782208053e-05,
"loss": 1.5674,
"num_input_tokens_seen": 2184708096,
"step": 1852
},
{
"epoch": 0.029845464197239588,
"grad_norm": 0.703125,
"learning_rate": 7.393698166877346e-05,
"loss": 1.7141,
"num_input_tokens_seen": 2185887744,
"step": 1853
},
{
"epoch": 0.029963430458888756,
"grad_norm": 0.68359375,
"learning_rate": 7.39287203519007e-05,
"loss": 1.5141,
"num_input_tokens_seen": 2187067392,
"step": 1854
},
{
"epoch": 0.030081396720537928,
"grad_norm": 0.6640625,
"learning_rate": 7.39204538727193e-05,
"loss": 1.6283,
"num_input_tokens_seen": 2188247040,
"step": 1855
},
{
"epoch": 0.030199362982187096,
"grad_norm": 0.66015625,
"learning_rate": 7.391218223248695e-05,
"loss": 1.7063,
"num_input_tokens_seen": 2189426688,
"step": 1856
},
{
"epoch": 0.030317329243836264,
"grad_norm": 0.765625,
"learning_rate": 7.390390543246224e-05,
"loss": 1.5577,
"num_input_tokens_seen": 2190606336,
"step": 1857
},
{
"epoch": 0.030435295505485432,
"grad_norm": 0.7578125,
"learning_rate": 7.389562347390447e-05,
"loss": 1.5518,
"num_input_tokens_seen": 2191785984,
"step": 1858
},
{
"epoch": 0.0305532617671346,
"grad_norm": 0.7890625,
"learning_rate": 7.388733635807378e-05,
"loss": 1.5015,
"num_input_tokens_seen": 2192965632,
"step": 1859
},
{
"epoch": 0.03067122802878377,
"grad_norm": 0.6640625,
"learning_rate": 7.387904408623103e-05,
"loss": 1.4798,
"num_input_tokens_seen": 2194145280,
"step": 1860
},
{
"epoch": 0.030789194290432937,
"grad_norm": 0.7578125,
"learning_rate": 7.387074665963794e-05,
"loss": 1.5943,
"num_input_tokens_seen": 2195324928,
"step": 1861
},
{
"epoch": 0.030907160552082105,
"grad_norm": 0.6796875,
"learning_rate": 7.386244407955695e-05,
"loss": 1.6022,
"num_input_tokens_seen": 2196504576,
"step": 1862
},
{
"epoch": 0.031025126813731273,
"grad_norm": 0.7421875,
"learning_rate": 7.385413634725132e-05,
"loss": 1.5654,
"num_input_tokens_seen": 2197684224,
"step": 1863
},
{
"epoch": 0.03114309307538044,
"grad_norm": 0.77734375,
"learning_rate": 7.384582346398509e-05,
"loss": 1.597,
"num_input_tokens_seen": 2198863872,
"step": 1864
},
{
"epoch": 0.031261059337029606,
"grad_norm": 0.765625,
"learning_rate": 7.383750543102308e-05,
"loss": 1.4255,
"num_input_tokens_seen": 2200043520,
"step": 1865
},
{
"epoch": 0.03137902559867878,
"grad_norm": 0.72265625,
"learning_rate": 7.382918224963087e-05,
"loss": 1.6604,
"num_input_tokens_seen": 2201223168,
"step": 1866
},
{
"epoch": 0.03149699186032794,
"grad_norm": 0.765625,
"learning_rate": 7.382085392107486e-05,
"loss": 1.6227,
"num_input_tokens_seen": 2202402816,
"step": 1867
},
{
"epoch": 0.031614958121977114,
"grad_norm": 0.94140625,
"learning_rate": 7.381252044662223e-05,
"loss": 1.5137,
"num_input_tokens_seen": 2203582464,
"step": 1868
},
{
"epoch": 0.031732924383626286,
"grad_norm": 0.71875,
"learning_rate": 7.380418182754093e-05,
"loss": 1.6039,
"num_input_tokens_seen": 2204762112,
"step": 1869
},
{
"epoch": 0.03185089064527545,
"grad_norm": 0.76953125,
"learning_rate": 7.379583806509967e-05,
"loss": 1.4421,
"num_input_tokens_seen": 2205941760,
"step": 1870
},
{
"epoch": 0.03196885690692462,
"grad_norm": 0.66796875,
"learning_rate": 7.378748916056798e-05,
"loss": 1.6332,
"num_input_tokens_seen": 2207121408,
"step": 1871
},
{
"epoch": 0.03208682316857379,
"grad_norm": 0.71484375,
"learning_rate": 7.377913511521617e-05,
"loss": 1.5797,
"num_input_tokens_seen": 2208301056,
"step": 1872
},
{
"epoch": 0.03220478943022296,
"grad_norm": 0.66796875,
"learning_rate": 7.377077593031531e-05,
"loss": 1.598,
"num_input_tokens_seen": 2209480704,
"step": 1873
},
{
"epoch": 0.032322755691872124,
"grad_norm": 0.73046875,
"learning_rate": 7.376241160713727e-05,
"loss": 1.5616,
"num_input_tokens_seen": 2210660352,
"step": 1874
},
{
"epoch": 0.032440721953521295,
"grad_norm": 0.765625,
"learning_rate": 7.37540421469547e-05,
"loss": 1.4614,
"num_input_tokens_seen": 2211840000,
"step": 1875
},
{
"epoch": 0.03255868821517046,
"grad_norm": 0.6796875,
"learning_rate": 7.374566755104098e-05,
"loss": 1.5495,
"num_input_tokens_seen": 2213019648,
"step": 1876
},
{
"epoch": 0.03267665447681963,
"grad_norm": 0.81640625,
"learning_rate": 7.373728782067038e-05,
"loss": 1.358,
"num_input_tokens_seen": 2214199296,
"step": 1877
},
{
"epoch": 0.0327946207384688,
"grad_norm": 0.79296875,
"learning_rate": 7.372890295711786e-05,
"loss": 1.4047,
"num_input_tokens_seen": 2215378944,
"step": 1878
},
{
"epoch": 0.03291258700011797,
"grad_norm": 0.7578125,
"learning_rate": 7.37205129616592e-05,
"loss": 1.4799,
"num_input_tokens_seen": 2216558592,
"step": 1879
},
{
"epoch": 0.03303055326176713,
"grad_norm": 0.71875,
"learning_rate": 7.371211783557095e-05,
"loss": 1.3531,
"num_input_tokens_seen": 2217738240,
"step": 1880
},
{
"epoch": 0.033148519523416305,
"grad_norm": 0.70703125,
"learning_rate": 7.370371758013042e-05,
"loss": 1.5988,
"num_input_tokens_seen": 2218917888,
"step": 1881
},
{
"epoch": 0.03326648578506547,
"grad_norm": 0.69140625,
"learning_rate": 7.369531219661575e-05,
"loss": 1.5523,
"num_input_tokens_seen": 2220097536,
"step": 1882
},
{
"epoch": 0.03338445204671464,
"grad_norm": 0.6484375,
"learning_rate": 7.368690168630582e-05,
"loss": 1.5946,
"num_input_tokens_seen": 2221277184,
"step": 1883
},
{
"epoch": 0.033502418308363806,
"grad_norm": 0.7109375,
"learning_rate": 7.367848605048031e-05,
"loss": 1.4925,
"num_input_tokens_seen": 2222456832,
"step": 1884
},
{
"epoch": 0.03362038457001298,
"grad_norm": 0.76171875,
"learning_rate": 7.367006529041967e-05,
"loss": 1.5686,
"num_input_tokens_seen": 2223636480,
"step": 1885
},
{
"epoch": 0.03373835083166214,
"grad_norm": 0.8359375,
"learning_rate": 7.366163940740511e-05,
"loss": 1.5127,
"num_input_tokens_seen": 2224816128,
"step": 1886
},
{
"epoch": 0.033856317093311314,
"grad_norm": 0.83203125,
"learning_rate": 7.365320840271867e-05,
"loss": 1.5116,
"num_input_tokens_seen": 2225995776,
"step": 1887
},
{
"epoch": 0.03397428335496048,
"grad_norm": 0.796875,
"learning_rate": 7.364477227764314e-05,
"loss": 1.486,
"num_input_tokens_seen": 2227175424,
"step": 1888
},
{
"epoch": 0.03409224961660965,
"grad_norm": 1.046875,
"learning_rate": 7.363633103346207e-05,
"loss": 1.6249,
"num_input_tokens_seen": 2228355072,
"step": 1889
},
{
"epoch": 0.034210215878258815,
"grad_norm": 0.8984375,
"learning_rate": 7.362788467145983e-05,
"loss": 1.6136,
"num_input_tokens_seen": 2229534720,
"step": 1890
},
{
"epoch": 0.03432818213990799,
"grad_norm": 0.84765625,
"learning_rate": 7.361943319292153e-05,
"loss": 1.5335,
"num_input_tokens_seen": 2230714368,
"step": 1891
},
{
"epoch": 0.03444614840155715,
"grad_norm": 0.7734375,
"learning_rate": 7.36109765991331e-05,
"loss": 1.4603,
"num_input_tokens_seen": 2231894016,
"step": 1892
},
{
"epoch": 0.03456411466320632,
"grad_norm": 0.83984375,
"learning_rate": 7.360251489138119e-05,
"loss": 1.4003,
"num_input_tokens_seen": 2233073664,
"step": 1893
},
{
"epoch": 0.03468208092485549,
"grad_norm": 0.73828125,
"learning_rate": 7.359404807095328e-05,
"loss": 1.5943,
"num_input_tokens_seen": 2234253312,
"step": 1894
},
{
"epoch": 0.03480004718650466,
"grad_norm": 0.8203125,
"learning_rate": 7.358557613913761e-05,
"loss": 1.5785,
"num_input_tokens_seen": 2235432960,
"step": 1895
},
{
"epoch": 0.03491801344815383,
"grad_norm": 0.8359375,
"learning_rate": 7.357709909722319e-05,
"loss": 1.3838,
"num_input_tokens_seen": 2236612608,
"step": 1896
},
{
"epoch": 0.035035979709802996,
"grad_norm": 0.69140625,
"learning_rate": 7.356861694649985e-05,
"loss": 1.5458,
"num_input_tokens_seen": 2237792256,
"step": 1897
},
{
"epoch": 0.03515394597145217,
"grad_norm": 0.73828125,
"learning_rate": 7.35601296882581e-05,
"loss": 1.5471,
"num_input_tokens_seen": 2238971904,
"step": 1898
},
{
"epoch": 0.03527191223310133,
"grad_norm": 0.8046875,
"learning_rate": 7.355163732378937e-05,
"loss": 1.4163,
"num_input_tokens_seen": 2240151552,
"step": 1899
},
{
"epoch": 0.035389878494750504,
"grad_norm": 0.703125,
"learning_rate": 7.35431398543857e-05,
"loss": 1.5865,
"num_input_tokens_seen": 2241331200,
"step": 1900
},
{
"epoch": 0.03550784475639967,
"grad_norm": 0.8203125,
"learning_rate": 7.353463728134005e-05,
"loss": 1.3711,
"num_input_tokens_seen": 2242510848,
"step": 1901
},
{
"epoch": 0.03562581101804884,
"grad_norm": 0.68359375,
"learning_rate": 7.352612960594609e-05,
"loss": 1.5282,
"num_input_tokens_seen": 2243690496,
"step": 1902
},
{
"epoch": 0.035743777279698005,
"grad_norm": 0.765625,
"learning_rate": 7.351761682949827e-05,
"loss": 1.6066,
"num_input_tokens_seen": 2244870144,
"step": 1903
},
{
"epoch": 0.03586174354134718,
"grad_norm": 0.76953125,
"learning_rate": 7.350909895329183e-05,
"loss": 1.5216,
"num_input_tokens_seen": 2246049792,
"step": 1904
},
{
"epoch": 0.03597970980299634,
"grad_norm": 0.7421875,
"learning_rate": 7.350057597862277e-05,
"loss": 1.5401,
"num_input_tokens_seen": 2247229440,
"step": 1905
},
{
"epoch": 0.03609767606464551,
"grad_norm": 0.77734375,
"learning_rate": 7.349204790678788e-05,
"loss": 1.4897,
"num_input_tokens_seen": 2248409088,
"step": 1906
},
{
"epoch": 0.03621564232629468,
"grad_norm": 0.8359375,
"learning_rate": 7.348351473908469e-05,
"loss": 1.5963,
"num_input_tokens_seen": 2249588736,
"step": 1907
},
{
"epoch": 0.03633360858794385,
"grad_norm": 0.765625,
"learning_rate": 7.34749764768116e-05,
"loss": 1.6018,
"num_input_tokens_seen": 2250768384,
"step": 1908
},
{
"epoch": 0.036451574849593014,
"grad_norm": 0.84765625,
"learning_rate": 7.346643312126766e-05,
"loss": 1.5299,
"num_input_tokens_seen": 2251948032,
"step": 1909
},
{
"epoch": 0.036569541111242186,
"grad_norm": 0.68359375,
"learning_rate": 7.345788467375278e-05,
"loss": 1.4308,
"num_input_tokens_seen": 2253127680,
"step": 1910
},
{
"epoch": 0.03668750737289135,
"grad_norm": 0.984375,
"learning_rate": 7.344933113556762e-05,
"loss": 1.6228,
"num_input_tokens_seen": 2254307328,
"step": 1911
},
{
"epoch": 0.03680547363454052,
"grad_norm": 0.78515625,
"learning_rate": 7.344077250801361e-05,
"loss": 1.5428,
"num_input_tokens_seen": 2255486976,
"step": 1912
},
{
"epoch": 0.03692343989618969,
"grad_norm": 0.8828125,
"learning_rate": 7.343220879239296e-05,
"loss": 1.4764,
"num_input_tokens_seen": 2256666624,
"step": 1913
},
{
"epoch": 0.03704140615783886,
"grad_norm": 0.9140625,
"learning_rate": 7.342363999000865e-05,
"loss": 1.4045,
"num_input_tokens_seen": 2257846272,
"step": 1914
},
{
"epoch": 0.037159372419488024,
"grad_norm": 0.87109375,
"learning_rate": 7.341506610216445e-05,
"loss": 1.4639,
"num_input_tokens_seen": 2259025920,
"step": 1915
},
{
"epoch": 0.037277338681137195,
"grad_norm": 0.8125,
"learning_rate": 7.340648713016487e-05,
"loss": 1.4876,
"num_input_tokens_seen": 2260205568,
"step": 1916
},
{
"epoch": 0.03739530494278636,
"grad_norm": 0.7890625,
"learning_rate": 7.339790307531523e-05,
"loss": 1.4313,
"num_input_tokens_seen": 2261385216,
"step": 1917
},
{
"epoch": 0.03751327120443553,
"grad_norm": 0.77734375,
"learning_rate": 7.33893139389216e-05,
"loss": 1.6842,
"num_input_tokens_seen": 2262564864,
"step": 1918
},
{
"epoch": 0.0376312374660847,
"grad_norm": 0.765625,
"learning_rate": 7.338071972229083e-05,
"loss": 1.3915,
"num_input_tokens_seen": 2263744512,
"step": 1919
},
{
"epoch": 0.03774920372773387,
"grad_norm": 0.765625,
"learning_rate": 7.337212042673055e-05,
"loss": 1.4723,
"num_input_tokens_seen": 2264924160,
"step": 1920
},
{
"epoch": 0.03786716998938304,
"grad_norm": 0.828125,
"learning_rate": 7.336351605354916e-05,
"loss": 1.3676,
"num_input_tokens_seen": 2266103808,
"step": 1921
},
{
"epoch": 0.037985136251032205,
"grad_norm": 0.66796875,
"learning_rate": 7.335490660405581e-05,
"loss": 1.512,
"num_input_tokens_seen": 2267283456,
"step": 1922
},
{
"epoch": 0.038103102512681376,
"grad_norm": 0.703125,
"learning_rate": 7.334629207956046e-05,
"loss": 1.4637,
"num_input_tokens_seen": 2268463104,
"step": 1923
},
{
"epoch": 0.03822106877433054,
"grad_norm": 0.66796875,
"learning_rate": 7.333767248137382e-05,
"loss": 1.5311,
"num_input_tokens_seen": 2269642752,
"step": 1924
},
{
"epoch": 0.03833903503597971,
"grad_norm": 0.6875,
"learning_rate": 7.332904781080736e-05,
"loss": 1.4918,
"num_input_tokens_seen": 2270822400,
"step": 1925
},
{
"epoch": 0.03845700129762888,
"grad_norm": 0.67578125,
"learning_rate": 7.332041806917337e-05,
"loss": 1.5913,
"num_input_tokens_seen": 2272002048,
"step": 1926
},
{
"epoch": 0.03857496755927805,
"grad_norm": 0.6953125,
"learning_rate": 7.331178325778485e-05,
"loss": 1.3829,
"num_input_tokens_seen": 2273181696,
"step": 1927
},
{
"epoch": 0.038692933820927214,
"grad_norm": 0.6875,
"learning_rate": 7.330314337795561e-05,
"loss": 1.4821,
"num_input_tokens_seen": 2274361344,
"step": 1928
},
{
"epoch": 0.038810900082576386,
"grad_norm": 0.7421875,
"learning_rate": 7.329449843100022e-05,
"loss": 1.5275,
"num_input_tokens_seen": 2275540992,
"step": 1929
},
{
"epoch": 0.03892886634422555,
"grad_norm": 0.63671875,
"learning_rate": 7.328584841823401e-05,
"loss": 1.634,
"num_input_tokens_seen": 2276720640,
"step": 1930
},
{
"epoch": 0.03904683260587472,
"grad_norm": 0.7109375,
"learning_rate": 7.327719334097312e-05,
"loss": 1.6776,
"num_input_tokens_seen": 2277900288,
"step": 1931
},
{
"epoch": 0.03916479886752389,
"grad_norm": 0.70703125,
"learning_rate": 7.326853320053442e-05,
"loss": 1.4269,
"num_input_tokens_seen": 2279079936,
"step": 1932
},
{
"epoch": 0.03928276512917306,
"grad_norm": 0.66015625,
"learning_rate": 7.325986799823555e-05,
"loss": 1.4935,
"num_input_tokens_seen": 2280259584,
"step": 1933
},
{
"epoch": 0.03940073139082222,
"grad_norm": 0.66015625,
"learning_rate": 7.325119773539497e-05,
"loss": 1.5445,
"num_input_tokens_seen": 2281439232,
"step": 1934
},
{
"epoch": 0.039518697652471395,
"grad_norm": 0.71484375,
"learning_rate": 7.324252241333185e-05,
"loss": 1.4365,
"num_input_tokens_seen": 2282618880,
"step": 1935
},
{
"epoch": 0.03963666391412056,
"grad_norm": 0.6484375,
"learning_rate": 7.323384203336615e-05,
"loss": 1.5626,
"num_input_tokens_seen": 2283798528,
"step": 1936
},
{
"epoch": 0.03975463017576973,
"grad_norm": 0.75390625,
"learning_rate": 7.32251565968186e-05,
"loss": 1.4364,
"num_input_tokens_seen": 2284978176,
"step": 1937
},
{
"epoch": 0.039872596437418896,
"grad_norm": 0.65234375,
"learning_rate": 7.321646610501073e-05,
"loss": 1.5292,
"num_input_tokens_seen": 2286157824,
"step": 1938
},
{
"epoch": 0.03999056269906807,
"grad_norm": 0.71484375,
"learning_rate": 7.320777055926478e-05,
"loss": 1.3565,
"num_input_tokens_seen": 2287337472,
"step": 1939
},
{
"epoch": 0.04010852896071723,
"grad_norm": 0.69921875,
"learning_rate": 7.319906996090383e-05,
"loss": 1.5397,
"num_input_tokens_seen": 2288517120,
"step": 1940
},
{
"epoch": 0.040226495222366404,
"grad_norm": 0.6640625,
"learning_rate": 7.319036431125165e-05,
"loss": 1.4882,
"num_input_tokens_seen": 2289696768,
"step": 1941
},
{
"epoch": 0.04034446148401557,
"grad_norm": 0.67578125,
"learning_rate": 7.318165361163284e-05,
"loss": 1.6145,
"num_input_tokens_seen": 2290876416,
"step": 1942
},
{
"epoch": 0.04046242774566474,
"grad_norm": 0.6484375,
"learning_rate": 7.317293786337273e-05,
"loss": 1.5312,
"num_input_tokens_seen": 2292056064,
"step": 1943
},
{
"epoch": 0.040580394007313905,
"grad_norm": 0.76171875,
"learning_rate": 7.316421706779746e-05,
"loss": 1.5138,
"num_input_tokens_seen": 2293235712,
"step": 1944
},
{
"epoch": 0.04069836026896308,
"grad_norm": 0.703125,
"learning_rate": 7.31554912262339e-05,
"loss": 1.4698,
"num_input_tokens_seen": 2294415360,
"step": 1945
},
{
"epoch": 0.04081632653061224,
"grad_norm": 0.65234375,
"learning_rate": 7.314676034000968e-05,
"loss": 1.4988,
"num_input_tokens_seen": 2295595008,
"step": 1946
},
{
"epoch": 0.04093429279226141,
"grad_norm": 0.65234375,
"learning_rate": 7.313802441045326e-05,
"loss": 1.5906,
"num_input_tokens_seen": 2296774656,
"step": 1947
},
{
"epoch": 0.041052259053910585,
"grad_norm": 0.67578125,
"learning_rate": 7.31292834388938e-05,
"loss": 1.7172,
"num_input_tokens_seen": 2297954304,
"step": 1948
},
{
"epoch": 0.04117022531555975,
"grad_norm": 0.66796875,
"learning_rate": 7.312053742666126e-05,
"loss": 1.5577,
"num_input_tokens_seen": 2299133952,
"step": 1949
},
{
"epoch": 0.04128819157720892,
"grad_norm": 0.6953125,
"learning_rate": 7.311178637508634e-05,
"loss": 1.5534,
"num_input_tokens_seen": 2300313600,
"step": 1950
},
{
"epoch": 0.041406157838858086,
"grad_norm": 0.62109375,
"learning_rate": 7.310303028550056e-05,
"loss": 1.4872,
"num_input_tokens_seen": 2301493248,
"step": 1951
},
{
"epoch": 0.04152412410050726,
"grad_norm": 0.73046875,
"learning_rate": 7.309426915923615e-05,
"loss": 1.4101,
"num_input_tokens_seen": 2302672896,
"step": 1952
},
{
"epoch": 0.04164209036215642,
"grad_norm": 0.69140625,
"learning_rate": 7.308550299762613e-05,
"loss": 1.609,
"num_input_tokens_seen": 2303852544,
"step": 1953
},
{
"epoch": 0.041760056623805594,
"grad_norm": 0.63671875,
"learning_rate": 7.307673180200429e-05,
"loss": 1.6246,
"num_input_tokens_seen": 2305032192,
"step": 1954
},
{
"epoch": 0.04187802288545476,
"grad_norm": 0.74609375,
"learning_rate": 7.306795557370519e-05,
"loss": 1.5308,
"num_input_tokens_seen": 2306211840,
"step": 1955
},
{
"epoch": 0.04199598914710393,
"grad_norm": 0.68359375,
"learning_rate": 7.305917431406413e-05,
"loss": 1.5942,
"num_input_tokens_seen": 2307391488,
"step": 1956
},
{
"epoch": 0.042113955408753095,
"grad_norm": 0.7734375,
"learning_rate": 7.30503880244172e-05,
"loss": 1.6072,
"num_input_tokens_seen": 2308571136,
"step": 1957
},
{
"epoch": 0.04223192167040227,
"grad_norm": 0.734375,
"learning_rate": 7.304159670610126e-05,
"loss": 1.3998,
"num_input_tokens_seen": 2309750784,
"step": 1958
},
{
"epoch": 0.04234988793205143,
"grad_norm": 0.71875,
"learning_rate": 7.30328003604539e-05,
"loss": 1.5292,
"num_input_tokens_seen": 2310930432,
"step": 1959
},
{
"epoch": 0.042467854193700603,
"grad_norm": 0.6875,
"learning_rate": 7.302399898881352e-05,
"loss": 1.4782,
"num_input_tokens_seen": 2312110080,
"step": 1960
},
{
"epoch": 0.04258582045534977,
"grad_norm": 0.703125,
"learning_rate": 7.301519259251925e-05,
"loss": 1.3287,
"num_input_tokens_seen": 2313289728,
"step": 1961
},
{
"epoch": 0.04270378671699894,
"grad_norm": 0.734375,
"learning_rate": 7.3006381172911e-05,
"loss": 1.5527,
"num_input_tokens_seen": 2314469376,
"step": 1962
},
{
"epoch": 0.042821752978648105,
"grad_norm": 0.734375,
"learning_rate": 7.299756473132944e-05,
"loss": 1.5148,
"num_input_tokens_seen": 2315649024,
"step": 1963
},
{
"epoch": 0.042939719240297276,
"grad_norm": 0.6640625,
"learning_rate": 7.2988743269116e-05,
"loss": 1.541,
"num_input_tokens_seen": 2316828672,
"step": 1964
},
{
"epoch": 0.04305768550194644,
"grad_norm": 0.6875,
"learning_rate": 7.297991678761289e-05,
"loss": 1.4709,
"num_input_tokens_seen": 2318008320,
"step": 1965
},
{
"epoch": 0.04317565176359561,
"grad_norm": 0.7421875,
"learning_rate": 7.297108528816308e-05,
"loss": 1.4995,
"num_input_tokens_seen": 2319187968,
"step": 1966
},
{
"epoch": 0.04329361802524478,
"grad_norm": 0.67578125,
"learning_rate": 7.296224877211029e-05,
"loss": 1.4244,
"num_input_tokens_seen": 2320367616,
"step": 1967
},
{
"epoch": 0.04341158428689395,
"grad_norm": 0.9296875,
"learning_rate": 7.295340724079899e-05,
"loss": 1.5493,
"num_input_tokens_seen": 2321547264,
"step": 1968
},
{
"epoch": 0.043529550548543114,
"grad_norm": 0.74609375,
"learning_rate": 7.294456069557445e-05,
"loss": 1.5873,
"num_input_tokens_seen": 2322726912,
"step": 1969
},
{
"epoch": 0.043647516810192286,
"grad_norm": 0.75,
"learning_rate": 7.293570913778268e-05,
"loss": 1.3775,
"num_input_tokens_seen": 2323906560,
"step": 1970
},
{
"epoch": 0.04376548307184145,
"grad_norm": 0.75,
"learning_rate": 7.292685256877049e-05,
"loss": 1.443,
"num_input_tokens_seen": 2325086208,
"step": 1971
},
{
"epoch": 0.04388344933349062,
"grad_norm": 0.703125,
"learning_rate": 7.291799098988539e-05,
"loss": 1.495,
"num_input_tokens_seen": 2326265856,
"step": 1972
},
{
"epoch": 0.04400141559513979,
"grad_norm": 0.65625,
"learning_rate": 7.290912440247567e-05,
"loss": 1.5695,
"num_input_tokens_seen": 2327445504,
"step": 1973
},
{
"epoch": 0.04411938185678896,
"grad_norm": 0.70703125,
"learning_rate": 7.290025280789042e-05,
"loss": 1.5184,
"num_input_tokens_seen": 2328625152,
"step": 1974
},
{
"epoch": 0.04423734811843813,
"grad_norm": 0.6953125,
"learning_rate": 7.289137620747947e-05,
"loss": 1.4567,
"num_input_tokens_seen": 2329804800,
"step": 1975
},
{
"epoch": 0.044355314380087295,
"grad_norm": 0.625,
"learning_rate": 7.288249460259338e-05,
"loss": 1.5241,
"num_input_tokens_seen": 2330984448,
"step": 1976
},
{
"epoch": 0.044473280641736467,
"grad_norm": 0.73828125,
"learning_rate": 7.287360799458354e-05,
"loss": 1.4828,
"num_input_tokens_seen": 2332164096,
"step": 1977
},
{
"epoch": 0.04459124690338563,
"grad_norm": 0.6640625,
"learning_rate": 7.286471638480204e-05,
"loss": 1.6184,
"num_input_tokens_seen": 2333343744,
"step": 1978
},
{
"epoch": 0.0447092131650348,
"grad_norm": 0.73828125,
"learning_rate": 7.285581977460174e-05,
"loss": 1.3618,
"num_input_tokens_seen": 2334523392,
"step": 1979
},
{
"epoch": 0.04482717942668397,
"grad_norm": 0.73046875,
"learning_rate": 7.28469181653363e-05,
"loss": 1.3807,
"num_input_tokens_seen": 2335703040,
"step": 1980
},
{
"epoch": 0.04494514568833314,
"grad_norm": 0.72265625,
"learning_rate": 7.283801155836009e-05,
"loss": 1.5257,
"num_input_tokens_seen": 2336882688,
"step": 1981
},
{
"epoch": 0.045063111949982304,
"grad_norm": 0.69140625,
"learning_rate": 7.282909995502828e-05,
"loss": 1.5478,
"num_input_tokens_seen": 2338062336,
"step": 1982
},
{
"epoch": 0.045181078211631476,
"grad_norm": 0.703125,
"learning_rate": 7.282018335669678e-05,
"loss": 1.4212,
"num_input_tokens_seen": 2339241984,
"step": 1983
},
{
"epoch": 0.04529904447328064,
"grad_norm": 0.82421875,
"learning_rate": 7.281126176472226e-05,
"loss": 1.4443,
"num_input_tokens_seen": 2340421632,
"step": 1984
},
{
"epoch": 0.04541701073492981,
"grad_norm": 0.640625,
"learning_rate": 7.280233518046217e-05,
"loss": 1.544,
"num_input_tokens_seen": 2341601280,
"step": 1985
},
{
"epoch": 0.04553497699657898,
"grad_norm": 0.796875,
"learning_rate": 7.27934036052747e-05,
"loss": 1.498,
"num_input_tokens_seen": 2342780928,
"step": 1986
},
{
"epoch": 0.04565294325822815,
"grad_norm": 0.69140625,
"learning_rate": 7.278446704051878e-05,
"loss": 1.6416,
"num_input_tokens_seen": 2343960576,
"step": 1987
},
{
"epoch": 0.04577090951987731,
"grad_norm": 0.76171875,
"learning_rate": 7.277552548755414e-05,
"loss": 1.4801,
"num_input_tokens_seen": 2345140224,
"step": 1988
},
{
"epoch": 0.045888875781526485,
"grad_norm": 0.703125,
"learning_rate": 7.276657894774126e-05,
"loss": 1.4833,
"num_input_tokens_seen": 2346319872,
"step": 1989
},
{
"epoch": 0.04600684204317565,
"grad_norm": 0.73046875,
"learning_rate": 7.275762742244135e-05,
"loss": 1.5809,
"num_input_tokens_seen": 2347499520,
"step": 1990
},
{
"epoch": 0.04612480830482482,
"grad_norm": 0.73828125,
"learning_rate": 7.274867091301642e-05,
"loss": 1.553,
"num_input_tokens_seen": 2348679168,
"step": 1991
},
{
"epoch": 0.046242774566473986,
"grad_norm": 0.890625,
"learning_rate": 7.27397094208292e-05,
"loss": 1.3766,
"num_input_tokens_seen": 2349858816,
"step": 1992
},
{
"epoch": 0.04636074082812316,
"grad_norm": 0.75390625,
"learning_rate": 7.27307429472432e-05,
"loss": 1.5416,
"num_input_tokens_seen": 2351038464,
"step": 1993
},
{
"epoch": 0.04647870708977232,
"grad_norm": 0.8828125,
"learning_rate": 7.27217714936227e-05,
"loss": 1.4611,
"num_input_tokens_seen": 2352218112,
"step": 1994
},
{
"epoch": 0.046596673351421494,
"grad_norm": 0.71875,
"learning_rate": 7.271279506133269e-05,
"loss": 1.4342,
"num_input_tokens_seen": 2353397760,
"step": 1995
},
{
"epoch": 0.04671463961307066,
"grad_norm": 0.84765625,
"learning_rate": 7.270381365173897e-05,
"loss": 1.5395,
"num_input_tokens_seen": 2354577408,
"step": 1996
},
{
"epoch": 0.04683260587471983,
"grad_norm": 0.7265625,
"learning_rate": 7.269482726620807e-05,
"loss": 1.5751,
"num_input_tokens_seen": 2355757056,
"step": 1997
},
{
"epoch": 0.046950572136368995,
"grad_norm": 0.75,
"learning_rate": 7.268583590610729e-05,
"loss": 1.5388,
"num_input_tokens_seen": 2356936704,
"step": 1998
},
{
"epoch": 0.04706853839801817,
"grad_norm": 0.7421875,
"learning_rate": 7.267683957280466e-05,
"loss": 1.4628,
"num_input_tokens_seen": 2358116352,
"step": 1999
},
{
"epoch": 0.04718650465966733,
"grad_norm": 0.6875,
"learning_rate": 7.266783826766901e-05,
"loss": 1.6311,
"num_input_tokens_seen": 2359296000,
"step": 2000
},
{
"epoch": 0.04718650465966733,
"eval_wikipedia_loss": 2.2608883380889893,
"eval_wikipedia_runtime": 163.5557,
"eval_wikipedia_samples_per_second": 4.292,
"eval_wikipedia_steps_per_second": 0.183,
"num_input_tokens_seen": 2359296000,
"step": 2000
},
{
"epoch": 0.04718650465966733,
"eval_toxicity_loss": 4.0160722732543945,
"eval_toxicity_runtime": 1.1558,
"eval_toxicity_samples_per_second": 1.73,
"eval_toxicity_steps_per_second": 0.865,
"num_input_tokens_seen": 2359296000,
"step": 2000
},
{
"epoch": 0.047304470921316503,
"grad_norm": 0.74609375,
"learning_rate": 7.265883199206989e-05,
"loss": 1.4713,
"num_input_tokens_seen": 2360475648,
"step": 2001
},
{
"epoch": 0.047422437182965675,
"grad_norm": 0.6953125,
"learning_rate": 7.264982074737762e-05,
"loss": 1.4955,
"num_input_tokens_seen": 2361655296,
"step": 2002
},
{
"epoch": 0.04754040344461484,
"grad_norm": 0.67578125,
"learning_rate": 7.264080453496328e-05,
"loss": 1.5753,
"num_input_tokens_seen": 2362834944,
"step": 2003
},
{
"epoch": 0.04765836970626401,
"grad_norm": 0.6328125,
"learning_rate": 7.263178335619868e-05,
"loss": 1.6932,
"num_input_tokens_seen": 2364014592,
"step": 2004
},
{
"epoch": 0.047776335967913176,
"grad_norm": 0.7109375,
"learning_rate": 7.262275721245644e-05,
"loss": 1.6407,
"num_input_tokens_seen": 2365194240,
"step": 2005
},
{
"epoch": 0.04789430222956235,
"grad_norm": 0.59765625,
"learning_rate": 7.26137261051099e-05,
"loss": 1.698,
"num_input_tokens_seen": 2366373888,
"step": 2006
},
{
"epoch": 0.04801226849121151,
"grad_norm": 0.76953125,
"learning_rate": 7.260469003553312e-05,
"loss": 1.4249,
"num_input_tokens_seen": 2367553536,
"step": 2007
},
{
"epoch": 0.048130234752860684,
"grad_norm": 0.61328125,
"learning_rate": 7.259564900510098e-05,
"loss": 1.7002,
"num_input_tokens_seen": 2368733184,
"step": 2008
},
{
"epoch": 0.04824820101450985,
"grad_norm": 0.6796875,
"learning_rate": 7.258660301518909e-05,
"loss": 1.7053,
"num_input_tokens_seen": 2369912832,
"step": 2009
},
{
"epoch": 0.04836616727615902,
"grad_norm": 0.6875,
"learning_rate": 7.257755206717379e-05,
"loss": 1.4856,
"num_input_tokens_seen": 2371092480,
"step": 2010
},
{
"epoch": 0.048484133537808186,
"grad_norm": 0.72265625,
"learning_rate": 7.256849616243223e-05,
"loss": 1.4462,
"num_input_tokens_seen": 2372272128,
"step": 2011
},
{
"epoch": 0.04860209979945736,
"grad_norm": 0.70703125,
"learning_rate": 7.255943530234225e-05,
"loss": 1.4224,
"num_input_tokens_seen": 2373451776,
"step": 2012
},
{
"epoch": 0.04872006606110652,
"grad_norm": 0.63671875,
"learning_rate": 7.255036948828249e-05,
"loss": 1.6058,
"num_input_tokens_seen": 2374631424,
"step": 2013
},
{
"epoch": 0.048838032322755694,
"grad_norm": 0.69921875,
"learning_rate": 7.254129872163232e-05,
"loss": 1.4915,
"num_input_tokens_seen": 2375811072,
"step": 2014
},
{
"epoch": 0.04895599858440486,
"grad_norm": 0.73828125,
"learning_rate": 7.253222300377188e-05,
"loss": 1.5527,
"num_input_tokens_seen": 2376990720,
"step": 2015
},
{
"epoch": 0.04907396484605403,
"grad_norm": 0.640625,
"learning_rate": 7.252314233608204e-05,
"loss": 1.5796,
"num_input_tokens_seen": 2378170368,
"step": 2016
},
{
"epoch": 0.049191931107703195,
"grad_norm": 0.6875,
"learning_rate": 7.251405671994446e-05,
"loss": 1.6129,
"num_input_tokens_seen": 2379350016,
"step": 2017
},
{
"epoch": 0.049309897369352367,
"grad_norm": 0.6796875,
"learning_rate": 7.250496615674152e-05,
"loss": 1.5305,
"num_input_tokens_seen": 2380529664,
"step": 2018
},
{
"epoch": 0.04942786363100153,
"grad_norm": 0.6328125,
"learning_rate": 7.249587064785636e-05,
"loss": 1.5511,
"num_input_tokens_seen": 2381709312,
"step": 2019
},
{
"epoch": 0.0495458298926507,
"grad_norm": 0.69140625,
"learning_rate": 7.248677019467286e-05,
"loss": 1.5779,
"num_input_tokens_seen": 2382888960,
"step": 2020
},
{
"epoch": 0.04966379615429987,
"grad_norm": 0.65625,
"learning_rate": 7.24776647985757e-05,
"loss": 1.5857,
"num_input_tokens_seen": 2384068608,
"step": 2021
},
{
"epoch": 0.04978176241594904,
"grad_norm": 0.66796875,
"learning_rate": 7.246855446095028e-05,
"loss": 1.496,
"num_input_tokens_seen": 2385248256,
"step": 2022
},
{
"epoch": 0.049899728677598204,
"grad_norm": 0.6171875,
"learning_rate": 7.245943918318272e-05,
"loss": 1.49,
"num_input_tokens_seen": 2386427904,
"step": 2023
},
{
"epoch": 0.050017694939247376,
"grad_norm": 0.65625,
"learning_rate": 7.245031896665995e-05,
"loss": 1.6214,
"num_input_tokens_seen": 2387607552,
"step": 2024
},
{
"epoch": 0.05013566120089654,
"grad_norm": 0.6015625,
"learning_rate": 7.244119381276963e-05,
"loss": 1.6914,
"num_input_tokens_seen": 2388787200,
"step": 2025
},
{
"epoch": 0.05025362746254571,
"grad_norm": 0.66796875,
"learning_rate": 7.243206372290014e-05,
"loss": 1.5241,
"num_input_tokens_seen": 2389966848,
"step": 2026
},
{
"epoch": 0.05037159372419488,
"grad_norm": 0.6796875,
"learning_rate": 7.242292869844067e-05,
"loss": 1.6441,
"num_input_tokens_seen": 2391146496,
"step": 2027
},
{
"epoch": 0.05048955998584405,
"grad_norm": 0.6875,
"learning_rate": 7.24137887407811e-05,
"loss": 1.6763,
"num_input_tokens_seen": 2392326144,
"step": 2028
},
{
"epoch": 0.05060752624749322,
"grad_norm": 0.66796875,
"learning_rate": 7.24046438513121e-05,
"loss": 1.4772,
"num_input_tokens_seen": 2393505792,
"step": 2029
},
{
"epoch": 0.050725492509142385,
"grad_norm": 0.66796875,
"learning_rate": 7.239549403142509e-05,
"loss": 1.5272,
"num_input_tokens_seen": 2394685440,
"step": 2030
},
{
"epoch": 0.05084345877079156,
"grad_norm": 0.7265625,
"learning_rate": 7.238633928251221e-05,
"loss": 1.3924,
"num_input_tokens_seen": 2395865088,
"step": 2031
},
{
"epoch": 0.05096142503244072,
"grad_norm": 0.6015625,
"learning_rate": 7.237717960596639e-05,
"loss": 1.6087,
"num_input_tokens_seen": 2397044736,
"step": 2032
},
{
"epoch": 0.05107939129408989,
"grad_norm": 0.71875,
"learning_rate": 7.236801500318127e-05,
"loss": 1.5425,
"num_input_tokens_seen": 2398224384,
"step": 2033
},
{
"epoch": 0.05119735755573906,
"grad_norm": 0.68359375,
"learning_rate": 7.235884547555127e-05,
"loss": 1.5686,
"num_input_tokens_seen": 2399404032,
"step": 2034
},
{
"epoch": 0.05131532381738823,
"grad_norm": 0.6328125,
"learning_rate": 7.234967102447155e-05,
"loss": 1.6643,
"num_input_tokens_seen": 2400583680,
"step": 2035
},
{
"epoch": 0.051433290079037394,
"grad_norm": 0.6875,
"learning_rate": 7.234049165133801e-05,
"loss": 1.5434,
"num_input_tokens_seen": 2401763328,
"step": 2036
},
{
"epoch": 0.051551256340686566,
"grad_norm": 0.70703125,
"learning_rate": 7.23313073575473e-05,
"loss": 1.5236,
"num_input_tokens_seen": 2402942976,
"step": 2037
},
{
"epoch": 0.05166922260233573,
"grad_norm": 0.625,
"learning_rate": 7.232211814449686e-05,
"loss": 1.59,
"num_input_tokens_seen": 2404122624,
"step": 2038
},
{
"epoch": 0.0517871888639849,
"grad_norm": 0.625,
"learning_rate": 7.23129240135848e-05,
"loss": 1.5156,
"num_input_tokens_seen": 2405302272,
"step": 2039
},
{
"epoch": 0.05190515512563407,
"grad_norm": 0.62109375,
"learning_rate": 7.230372496621003e-05,
"loss": 1.6894,
"num_input_tokens_seen": 2406481920,
"step": 2040
},
{
"epoch": 0.05202312138728324,
"grad_norm": 0.66015625,
"learning_rate": 7.229452100377223e-05,
"loss": 1.5584,
"num_input_tokens_seen": 2407661568,
"step": 2041
},
{
"epoch": 0.052141087648932403,
"grad_norm": 0.70703125,
"learning_rate": 7.228531212767178e-05,
"loss": 1.5068,
"num_input_tokens_seen": 2408841216,
"step": 2042
},
{
"epoch": 0.052259053910581575,
"grad_norm": 0.67578125,
"learning_rate": 7.22760983393098e-05,
"loss": 1.595,
"num_input_tokens_seen": 2410020864,
"step": 2043
},
{
"epoch": 0.05237702017223074,
"grad_norm": 0.64453125,
"learning_rate": 7.226687964008822e-05,
"loss": 1.4293,
"num_input_tokens_seen": 2411200512,
"step": 2044
},
{
"epoch": 0.05249498643387991,
"grad_norm": 0.65625,
"learning_rate": 7.225765603140964e-05,
"loss": 1.4935,
"num_input_tokens_seen": 2412380160,
"step": 2045
},
{
"epoch": 0.052612952695529076,
"grad_norm": 0.640625,
"learning_rate": 7.22484275146775e-05,
"loss": 1.453,
"num_input_tokens_seen": 2413559808,
"step": 2046
},
{
"epoch": 0.05273091895717825,
"grad_norm": 0.625,
"learning_rate": 7.223919409129589e-05,
"loss": 1.5559,
"num_input_tokens_seen": 2414739456,
"step": 2047
},
{
"epoch": 0.05284888521882741,
"grad_norm": 0.640625,
"learning_rate": 7.222995576266971e-05,
"loss": 1.4826,
"num_input_tokens_seen": 2415919104,
"step": 2048
},
{
"epoch": 0.052966851480476584,
"grad_norm": 0.70703125,
"learning_rate": 7.222071253020457e-05,
"loss": 1.5566,
"num_input_tokens_seen": 2417098752,
"step": 2049
},
{
"epoch": 0.05308481774212575,
"grad_norm": 0.609375,
"learning_rate": 7.221146439530687e-05,
"loss": 1.6199,
"num_input_tokens_seen": 2418278400,
"step": 2050
},
{
"epoch": 0.05320278400377492,
"grad_norm": 0.625,
"learning_rate": 7.220221135938369e-05,
"loss": 1.5922,
"num_input_tokens_seen": 2419458048,
"step": 2051
},
{
"epoch": 0.053320750265424086,
"grad_norm": 0.59375,
"learning_rate": 7.219295342384293e-05,
"loss": 1.5259,
"num_input_tokens_seen": 2420637696,
"step": 2052
},
{
"epoch": 0.05343871652707326,
"grad_norm": 0.58984375,
"learning_rate": 7.218369059009319e-05,
"loss": 1.6072,
"num_input_tokens_seen": 2421817344,
"step": 2053
},
{
"epoch": 0.05355668278872242,
"grad_norm": 0.671875,
"learning_rate": 7.217442285954381e-05,
"loss": 1.6169,
"num_input_tokens_seen": 2422996992,
"step": 2054
},
{
"epoch": 0.053674649050371594,
"grad_norm": 0.62109375,
"learning_rate": 7.21651502336049e-05,
"loss": 1.5768,
"num_input_tokens_seen": 2424176640,
"step": 2055
},
{
"epoch": 0.053792615312020765,
"grad_norm": 0.609375,
"learning_rate": 7.21558727136873e-05,
"loss": 1.5223,
"num_input_tokens_seen": 2425356288,
"step": 2056
},
{
"epoch": 0.05391058157366993,
"grad_norm": 0.59765625,
"learning_rate": 7.214659030120262e-05,
"loss": 1.5783,
"num_input_tokens_seen": 2426535936,
"step": 2057
},
{
"epoch": 0.0540285478353191,
"grad_norm": 0.71875,
"learning_rate": 7.213730299756317e-05,
"loss": 1.4265,
"num_input_tokens_seen": 2427715584,
"step": 2058
},
{
"epoch": 0.054146514096968267,
"grad_norm": 0.6484375,
"learning_rate": 7.212801080418204e-05,
"loss": 1.5677,
"num_input_tokens_seen": 2428895232,
"step": 2059
},
{
"epoch": 0.05426448035861744,
"grad_norm": 0.6171875,
"learning_rate": 7.211871372247304e-05,
"loss": 1.5219,
"num_input_tokens_seen": 2430074880,
"step": 2060
},
{
"epoch": 0.0543824466202666,
"grad_norm": 0.66796875,
"learning_rate": 7.210941175385075e-05,
"loss": 1.5097,
"num_input_tokens_seen": 2431254528,
"step": 2061
},
{
"epoch": 0.054500412881915775,
"grad_norm": 0.66015625,
"learning_rate": 7.210010489973047e-05,
"loss": 1.7001,
"num_input_tokens_seen": 2432434176,
"step": 2062
},
{
"epoch": 0.05461837914356494,
"grad_norm": 0.640625,
"learning_rate": 7.209079316152826e-05,
"loss": 1.5308,
"num_input_tokens_seen": 2433613824,
"step": 2063
},
{
"epoch": 0.05473634540521411,
"grad_norm": 0.63671875,
"learning_rate": 7.208147654066091e-05,
"loss": 1.5989,
"num_input_tokens_seen": 2434793472,
"step": 2064
},
{
"epoch": 0.054854311666863276,
"grad_norm": 0.6328125,
"learning_rate": 7.207215503854596e-05,
"loss": 1.7482,
"num_input_tokens_seen": 2435973120,
"step": 2065
},
{
"epoch": 0.05497227792851245,
"grad_norm": 0.6875,
"learning_rate": 7.206282865660169e-05,
"loss": 1.6143,
"num_input_tokens_seen": 2437152768,
"step": 2066
},
{
"epoch": 0.05509024419016161,
"grad_norm": 0.6484375,
"learning_rate": 7.205349739624711e-05,
"loss": 1.6127,
"num_input_tokens_seen": 2438332416,
"step": 2067
},
{
"epoch": 0.055208210451810784,
"grad_norm": 0.64453125,
"learning_rate": 7.204416125890203e-05,
"loss": 1.4988,
"num_input_tokens_seen": 2439512064,
"step": 2068
},
{
"epoch": 0.05532617671345995,
"grad_norm": 0.61328125,
"learning_rate": 7.20348202459869e-05,
"loss": 1.5739,
"num_input_tokens_seen": 2440691712,
"step": 2069
},
{
"epoch": 0.05544414297510912,
"grad_norm": 0.65234375,
"learning_rate": 7.202547435892302e-05,
"loss": 1.5308,
"num_input_tokens_seen": 2441871360,
"step": 2070
},
{
"epoch": 0.055562109236758285,
"grad_norm": 0.625,
"learning_rate": 7.201612359913235e-05,
"loss": 1.5948,
"num_input_tokens_seen": 2443051008,
"step": 2071
},
{
"epoch": 0.05568007549840746,
"grad_norm": 0.67578125,
"learning_rate": 7.200676796803765e-05,
"loss": 1.5753,
"num_input_tokens_seen": 2444230656,
"step": 2072
},
{
"epoch": 0.05579804176005662,
"grad_norm": 0.6484375,
"learning_rate": 7.199740746706235e-05,
"loss": 1.5194,
"num_input_tokens_seen": 2445410304,
"step": 2073
},
{
"epoch": 0.05591600802170579,
"grad_norm": 0.68359375,
"learning_rate": 7.198804209763071e-05,
"loss": 1.4933,
"num_input_tokens_seen": 2446589952,
"step": 2074
},
{
"epoch": 0.05603397428335496,
"grad_norm": 0.65234375,
"learning_rate": 7.197867186116767e-05,
"loss": 1.5015,
"num_input_tokens_seen": 2447769600,
"step": 2075
},
{
"epoch": 0.05615194054500413,
"grad_norm": 0.63671875,
"learning_rate": 7.196929675909893e-05,
"loss": 1.6335,
"num_input_tokens_seen": 2448949248,
"step": 2076
},
{
"epoch": 0.056269906806653294,
"grad_norm": 0.6171875,
"learning_rate": 7.195991679285091e-05,
"loss": 1.5782,
"num_input_tokens_seen": 2450128896,
"step": 2077
},
{
"epoch": 0.056387873068302466,
"grad_norm": 0.703125,
"learning_rate": 7.19505319638508e-05,
"loss": 1.4924,
"num_input_tokens_seen": 2451308544,
"step": 2078
},
{
"epoch": 0.05650583932995163,
"grad_norm": 0.69921875,
"learning_rate": 7.194114227352653e-05,
"loss": 1.5547,
"num_input_tokens_seen": 2452488192,
"step": 2079
},
{
"epoch": 0.0566238055916008,
"grad_norm": 0.61328125,
"learning_rate": 7.193174772330673e-05,
"loss": 1.4579,
"num_input_tokens_seen": 2453667840,
"step": 2080
},
{
"epoch": 0.05674177185324997,
"grad_norm": 0.7421875,
"learning_rate": 7.19223483146208e-05,
"loss": 1.4006,
"num_input_tokens_seen": 2454847488,
"step": 2081
},
{
"epoch": 0.05685973811489914,
"grad_norm": 0.6875,
"learning_rate": 7.191294404889891e-05,
"loss": 1.4052,
"num_input_tokens_seen": 2456027136,
"step": 2082
},
{
"epoch": 0.05697770437654831,
"grad_norm": 0.6640625,
"learning_rate": 7.190353492757186e-05,
"loss": 1.5845,
"num_input_tokens_seen": 2457206784,
"step": 2083
},
{
"epoch": 0.057095670638197475,
"grad_norm": 0.66796875,
"learning_rate": 7.189412095207136e-05,
"loss": 1.5037,
"num_input_tokens_seen": 2458386432,
"step": 2084
},
{
"epoch": 0.05721363689984665,
"grad_norm": 0.76953125,
"learning_rate": 7.188470212382968e-05,
"loss": 1.5359,
"num_input_tokens_seen": 2459566080,
"step": 2085
},
{
"epoch": 0.05733160316149581,
"grad_norm": 0.84375,
"learning_rate": 7.187527844427995e-05,
"loss": 1.665,
"num_input_tokens_seen": 2460745728,
"step": 2086
},
{
"epoch": 0.05744956942314498,
"grad_norm": 0.6875,
"learning_rate": 7.186584991485599e-05,
"loss": 1.6403,
"num_input_tokens_seen": 2461925376,
"step": 2087
},
{
"epoch": 0.05756753568479415,
"grad_norm": 0.7265625,
"learning_rate": 7.185641653699234e-05,
"loss": 1.3332,
"num_input_tokens_seen": 2463105024,
"step": 2088
},
{
"epoch": 0.05768550194644332,
"grad_norm": 0.609375,
"learning_rate": 7.184697831212435e-05,
"loss": 1.6377,
"num_input_tokens_seen": 2464284672,
"step": 2089
},
{
"epoch": 0.057803468208092484,
"grad_norm": 0.73828125,
"learning_rate": 7.183753524168803e-05,
"loss": 1.4486,
"num_input_tokens_seen": 2465464320,
"step": 2090
},
{
"epoch": 0.057921434469741656,
"grad_norm": 0.66015625,
"learning_rate": 7.182808732712018e-05,
"loss": 1.4627,
"num_input_tokens_seen": 2466643968,
"step": 2091
},
{
"epoch": 0.05803940073139082,
"grad_norm": 0.65234375,
"learning_rate": 7.181863456985827e-05,
"loss": 1.6655,
"num_input_tokens_seen": 2467823616,
"step": 2092
},
{
"epoch": 0.05815736699303999,
"grad_norm": 0.63671875,
"learning_rate": 7.180917697134061e-05,
"loss": 1.499,
"num_input_tokens_seen": 2469003264,
"step": 2093
},
{
"epoch": 0.05827533325468916,
"grad_norm": 0.79296875,
"learning_rate": 7.179971453300615e-05,
"loss": 1.5666,
"num_input_tokens_seen": 2470182912,
"step": 2094
},
{
"epoch": 0.05839329951633833,
"grad_norm": 0.62109375,
"learning_rate": 7.179024725629464e-05,
"loss": 1.754,
"num_input_tokens_seen": 2471362560,
"step": 2095
},
{
"epoch": 0.058511265777987494,
"grad_norm": 0.6484375,
"learning_rate": 7.178077514264652e-05,
"loss": 1.6832,
"num_input_tokens_seen": 2472542208,
"step": 2096
},
{
"epoch": 0.058629232039636665,
"grad_norm": 0.7109375,
"learning_rate": 7.177129819350299e-05,
"loss": 1.4757,
"num_input_tokens_seen": 2473721856,
"step": 2097
},
{
"epoch": 0.05874719830128583,
"grad_norm": 0.59765625,
"learning_rate": 7.1761816410306e-05,
"loss": 1.4754,
"num_input_tokens_seen": 2474901504,
"step": 2098
},
{
"epoch": 0.058865164562935,
"grad_norm": 0.59375,
"learning_rate": 7.175232979449819e-05,
"loss": 1.677,
"num_input_tokens_seen": 2476081152,
"step": 2099
},
{
"epoch": 0.058983130824584167,
"grad_norm": 0.63671875,
"learning_rate": 7.174283834752299e-05,
"loss": 1.588,
"num_input_tokens_seen": 2477260800,
"step": 2100
},
{
"epoch": 0.05910109708623334,
"grad_norm": 0.72265625,
"learning_rate": 7.173334207082453e-05,
"loss": 1.5977,
"num_input_tokens_seen": 2478440448,
"step": 2101
},
{
"epoch": 0.0592190633478825,
"grad_norm": 0.68359375,
"learning_rate": 7.172384096584769e-05,
"loss": 1.4083,
"num_input_tokens_seen": 2479620096,
"step": 2102
},
{
"epoch": 0.059337029609531675,
"grad_norm": 0.63671875,
"learning_rate": 7.171433503403805e-05,
"loss": 1.7229,
"num_input_tokens_seen": 2480799744,
"step": 2103
},
{
"epoch": 0.05945499587118084,
"grad_norm": 0.62890625,
"learning_rate": 7.170482427684199e-05,
"loss": 1.5378,
"num_input_tokens_seen": 2481979392,
"step": 2104
},
{
"epoch": 0.05957296213283001,
"grad_norm": 0.6796875,
"learning_rate": 7.169530869570655e-05,
"loss": 1.6472,
"num_input_tokens_seen": 2483159040,
"step": 2105
},
{
"epoch": 0.059690928394479176,
"grad_norm": 0.71484375,
"learning_rate": 7.168578829207958e-05,
"loss": 1.5256,
"num_input_tokens_seen": 2484338688,
"step": 2106
},
{
"epoch": 0.05980889465612835,
"grad_norm": 0.6171875,
"learning_rate": 7.167626306740961e-05,
"loss": 1.6106,
"num_input_tokens_seen": 2485518336,
"step": 2107
},
{
"epoch": 0.05992686091777751,
"grad_norm": 0.61328125,
"learning_rate": 7.16667330231459e-05,
"loss": 1.6116,
"num_input_tokens_seen": 2486697984,
"step": 2108
},
{
"epoch": 0.060044827179426684,
"grad_norm": 0.67578125,
"learning_rate": 7.165719816073848e-05,
"loss": 1.553,
"num_input_tokens_seen": 2487877632,
"step": 2109
},
{
"epoch": 0.060162793441075856,
"grad_norm": 0.7109375,
"learning_rate": 7.164765848163809e-05,
"loss": 1.5079,
"num_input_tokens_seen": 2489057280,
"step": 2110
},
{
"epoch": 0.06028075970272502,
"grad_norm": 0.6171875,
"learning_rate": 7.16381139872962e-05,
"loss": 1.7166,
"num_input_tokens_seen": 2490236928,
"step": 2111
},
{
"epoch": 0.06039872596437419,
"grad_norm": 0.69140625,
"learning_rate": 7.162856467916504e-05,
"loss": 1.5981,
"num_input_tokens_seen": 2491416576,
"step": 2112
},
{
"epoch": 0.06051669222602336,
"grad_norm": 0.67578125,
"learning_rate": 7.161901055869752e-05,
"loss": 1.4822,
"num_input_tokens_seen": 2492596224,
"step": 2113
},
{
"epoch": 0.06063465848767253,
"grad_norm": 0.65234375,
"learning_rate": 7.160945162734735e-05,
"loss": 1.6089,
"num_input_tokens_seen": 2493775872,
"step": 2114
},
{
"epoch": 0.06075262474932169,
"grad_norm": 0.67578125,
"learning_rate": 7.159988788656892e-05,
"loss": 1.5562,
"num_input_tokens_seen": 2494955520,
"step": 2115
},
{
"epoch": 0.060870591010970865,
"grad_norm": 0.65234375,
"learning_rate": 7.159031933781736e-05,
"loss": 1.5186,
"num_input_tokens_seen": 2496135168,
"step": 2116
},
{
"epoch": 0.06098855727262003,
"grad_norm": 0.6953125,
"learning_rate": 7.158074598254855e-05,
"loss": 1.3422,
"num_input_tokens_seen": 2497314816,
"step": 2117
},
{
"epoch": 0.0611065235342692,
"grad_norm": 0.59765625,
"learning_rate": 7.157116782221909e-05,
"loss": 1.7471,
"num_input_tokens_seen": 2498494464,
"step": 2118
},
{
"epoch": 0.061224489795918366,
"grad_norm": 0.6328125,
"learning_rate": 7.156158485828631e-05,
"loss": 1.6097,
"num_input_tokens_seen": 2499674112,
"step": 2119
},
{
"epoch": 0.06134245605756754,
"grad_norm": 0.63671875,
"learning_rate": 7.155199709220828e-05,
"loss": 1.4928,
"num_input_tokens_seen": 2500853760,
"step": 2120
},
{
"epoch": 0.0614604223192167,
"grad_norm": 0.62890625,
"learning_rate": 7.154240452544378e-05,
"loss": 1.5735,
"num_input_tokens_seen": 2502033408,
"step": 2121
},
{
"epoch": 0.061578388580865874,
"grad_norm": 0.72265625,
"learning_rate": 7.153280715945235e-05,
"loss": 1.5354,
"num_input_tokens_seen": 2503213056,
"step": 2122
},
{
"epoch": 0.06169635484251504,
"grad_norm": 0.6953125,
"learning_rate": 7.152320499569425e-05,
"loss": 1.5084,
"num_input_tokens_seen": 2504392704,
"step": 2123
},
{
"epoch": 0.06181432110416421,
"grad_norm": 0.6328125,
"learning_rate": 7.151359803563042e-05,
"loss": 1.4934,
"num_input_tokens_seen": 2505572352,
"step": 2124
},
{
"epoch": 0.061932287365813375,
"grad_norm": 0.671875,
"learning_rate": 7.150398628072263e-05,
"loss": 1.4994,
"num_input_tokens_seen": 2506752000,
"step": 2125
},
{
"epoch": 0.06205025362746255,
"grad_norm": 0.65234375,
"learning_rate": 7.14943697324333e-05,
"loss": 1.4492,
"num_input_tokens_seen": 2507931648,
"step": 2126
},
{
"epoch": 0.06216821988911171,
"grad_norm": 0.6328125,
"learning_rate": 7.148474839222559e-05,
"loss": 1.7602,
"num_input_tokens_seen": 2509111296,
"step": 2127
},
{
"epoch": 0.06228618615076088,
"grad_norm": 0.6484375,
"learning_rate": 7.147512226156342e-05,
"loss": 1.5096,
"num_input_tokens_seen": 2510290944,
"step": 2128
},
{
"epoch": 0.06240415241241005,
"grad_norm": 0.65234375,
"learning_rate": 7.146549134191142e-05,
"loss": 1.6076,
"num_input_tokens_seen": 2511470592,
"step": 2129
},
{
"epoch": 0.06252211867405921,
"grad_norm": 0.64453125,
"learning_rate": 7.145585563473495e-05,
"loss": 1.449,
"num_input_tokens_seen": 2512650240,
"step": 2130
},
{
"epoch": 0.06264008493570838,
"grad_norm": 0.6484375,
"learning_rate": 7.14462151415001e-05,
"loss": 1.5525,
"num_input_tokens_seen": 2513829888,
"step": 2131
},
{
"epoch": 0.06275805119735756,
"grad_norm": 0.62890625,
"learning_rate": 7.143656986367368e-05,
"loss": 1.4403,
"num_input_tokens_seen": 2515009536,
"step": 2132
},
{
"epoch": 0.06287601745900673,
"grad_norm": 0.62890625,
"learning_rate": 7.142691980272322e-05,
"loss": 1.5139,
"num_input_tokens_seen": 2516189184,
"step": 2133
},
{
"epoch": 0.06299398372065589,
"grad_norm": 0.67578125,
"learning_rate": 7.141726496011701e-05,
"loss": 1.4524,
"num_input_tokens_seen": 2517368832,
"step": 2134
},
{
"epoch": 0.06311194998230506,
"grad_norm": 0.68359375,
"learning_rate": 7.140760533732405e-05,
"loss": 1.4579,
"num_input_tokens_seen": 2518548480,
"step": 2135
},
{
"epoch": 0.06322991624395423,
"grad_norm": 0.61328125,
"learning_rate": 7.139794093581407e-05,
"loss": 1.6199,
"num_input_tokens_seen": 2519728128,
"step": 2136
},
{
"epoch": 0.0633478825056034,
"grad_norm": 0.83203125,
"learning_rate": 7.13882717570575e-05,
"loss": 1.6241,
"num_input_tokens_seen": 2520907776,
"step": 2137
},
{
"epoch": 0.06346584876725257,
"grad_norm": 0.72265625,
"learning_rate": 7.137859780252555e-05,
"loss": 1.6226,
"num_input_tokens_seen": 2522087424,
"step": 2138
},
{
"epoch": 0.06358381502890173,
"grad_norm": 0.78515625,
"learning_rate": 7.136891907369012e-05,
"loss": 1.507,
"num_input_tokens_seen": 2523267072,
"step": 2139
},
{
"epoch": 0.0637017812905509,
"grad_norm": 0.734375,
"learning_rate": 7.135923557202383e-05,
"loss": 1.3917,
"num_input_tokens_seen": 2524446720,
"step": 2140
},
{
"epoch": 0.06381974755220007,
"grad_norm": 0.66796875,
"learning_rate": 7.134954729900006e-05,
"loss": 1.4087,
"num_input_tokens_seen": 2525626368,
"step": 2141
},
{
"epoch": 0.06393771381384925,
"grad_norm": 0.7109375,
"learning_rate": 7.133985425609288e-05,
"loss": 1.5533,
"num_input_tokens_seen": 2526806016,
"step": 2142
},
{
"epoch": 0.0640556800754984,
"grad_norm": 0.73046875,
"learning_rate": 7.13301564447771e-05,
"loss": 1.4244,
"num_input_tokens_seen": 2527985664,
"step": 2143
},
{
"epoch": 0.06417364633714757,
"grad_norm": 0.62109375,
"learning_rate": 7.132045386652829e-05,
"loss": 1.5217,
"num_input_tokens_seen": 2529165312,
"step": 2144
},
{
"epoch": 0.06429161259879675,
"grad_norm": 0.7265625,
"learning_rate": 7.131074652282268e-05,
"loss": 1.4388,
"num_input_tokens_seen": 2530344960,
"step": 2145
},
{
"epoch": 0.06440957886044592,
"grad_norm": 0.6328125,
"learning_rate": 7.130103441513726e-05,
"loss": 1.6643,
"num_input_tokens_seen": 2531524608,
"step": 2146
},
{
"epoch": 0.06452754512209508,
"grad_norm": 0.64453125,
"learning_rate": 7.129131754494975e-05,
"loss": 1.6273,
"num_input_tokens_seen": 2532704256,
"step": 2147
},
{
"epoch": 0.06464551138374425,
"grad_norm": 0.6796875,
"learning_rate": 7.128159591373859e-05,
"loss": 1.5731,
"num_input_tokens_seen": 2533883904,
"step": 2148
},
{
"epoch": 0.06476347764539342,
"grad_norm": 0.6640625,
"learning_rate": 7.127186952298293e-05,
"loss": 1.5956,
"num_input_tokens_seen": 2535063552,
"step": 2149
},
{
"epoch": 0.06488144390704259,
"grad_norm": 0.703125,
"learning_rate": 7.126213837416267e-05,
"loss": 1.5309,
"num_input_tokens_seen": 2536243200,
"step": 2150
},
{
"epoch": 0.06499941016869175,
"grad_norm": 0.6328125,
"learning_rate": 7.125240246875841e-05,
"loss": 1.5006,
"num_input_tokens_seen": 2537422848,
"step": 2151
},
{
"epoch": 0.06511737643034092,
"grad_norm": 0.71484375,
"learning_rate": 7.124266180825148e-05,
"loss": 1.6113,
"num_input_tokens_seen": 2538602496,
"step": 2152
},
{
"epoch": 0.06523534269199009,
"grad_norm": 0.62109375,
"learning_rate": 7.123291639412395e-05,
"loss": 1.4539,
"num_input_tokens_seen": 2539782144,
"step": 2153
},
{
"epoch": 0.06535330895363926,
"grad_norm": 0.7109375,
"learning_rate": 7.12231662278586e-05,
"loss": 1.4511,
"num_input_tokens_seen": 2540961792,
"step": 2154
},
{
"epoch": 0.06547127521528842,
"grad_norm": 0.5625,
"learning_rate": 7.121341131093892e-05,
"loss": 1.5934,
"num_input_tokens_seen": 2542141440,
"step": 2155
},
{
"epoch": 0.0655892414769376,
"grad_norm": 0.625,
"learning_rate": 7.120365164484915e-05,
"loss": 1.5815,
"num_input_tokens_seen": 2543321088,
"step": 2156
},
{
"epoch": 0.06570720773858676,
"grad_norm": 0.73828125,
"learning_rate": 7.119388723107422e-05,
"loss": 1.5041,
"num_input_tokens_seen": 2544500736,
"step": 2157
},
{
"epoch": 0.06582517400023594,
"grad_norm": 0.62109375,
"learning_rate": 7.118411807109983e-05,
"loss": 1.5656,
"num_input_tokens_seen": 2545680384,
"step": 2158
},
{
"epoch": 0.0659431402618851,
"grad_norm": 0.68359375,
"learning_rate": 7.117434416641234e-05,
"loss": 1.5776,
"num_input_tokens_seen": 2546860032,
"step": 2159
},
{
"epoch": 0.06606110652353427,
"grad_norm": 0.64453125,
"learning_rate": 7.116456551849889e-05,
"loss": 1.4988,
"num_input_tokens_seen": 2548039680,
"step": 2160
},
{
"epoch": 0.06617907278518344,
"grad_norm": 0.6484375,
"learning_rate": 7.11547821288473e-05,
"loss": 1.374,
"num_input_tokens_seen": 2549219328,
"step": 2161
},
{
"epoch": 0.06629703904683261,
"grad_norm": 0.59375,
"learning_rate": 7.114499399894614e-05,
"loss": 1.4518,
"num_input_tokens_seen": 2550398976,
"step": 2162
},
{
"epoch": 0.06641500530848178,
"grad_norm": 0.6171875,
"learning_rate": 7.113520113028468e-05,
"loss": 1.6117,
"num_input_tokens_seen": 2551578624,
"step": 2163
},
{
"epoch": 0.06653297157013094,
"grad_norm": 0.6171875,
"learning_rate": 7.112540352435294e-05,
"loss": 1.4237,
"num_input_tokens_seen": 2552758272,
"step": 2164
},
{
"epoch": 0.06665093783178011,
"grad_norm": 0.66015625,
"learning_rate": 7.111560118264162e-05,
"loss": 1.4234,
"num_input_tokens_seen": 2553937920,
"step": 2165
},
{
"epoch": 0.06676890409342928,
"grad_norm": 0.60546875,
"learning_rate": 7.110579410664217e-05,
"loss": 1.5433,
"num_input_tokens_seen": 2555117568,
"step": 2166
},
{
"epoch": 0.06688687035507845,
"grad_norm": 0.6328125,
"learning_rate": 7.109598229784675e-05,
"loss": 1.6265,
"num_input_tokens_seen": 2556297216,
"step": 2167
},
{
"epoch": 0.06700483661672761,
"grad_norm": 0.59765625,
"learning_rate": 7.108616575774824e-05,
"loss": 1.599,
"num_input_tokens_seen": 2557476864,
"step": 2168
},
{
"epoch": 0.06712280287837678,
"grad_norm": 0.6796875,
"learning_rate": 7.107634448784025e-05,
"loss": 1.7717,
"num_input_tokens_seen": 2558656512,
"step": 2169
},
{
"epoch": 0.06724076914002595,
"grad_norm": 0.71484375,
"learning_rate": 7.106651848961711e-05,
"loss": 1.4123,
"num_input_tokens_seen": 2559836160,
"step": 2170
},
{
"epoch": 0.06735873540167513,
"grad_norm": 0.734375,
"learning_rate": 7.105668776457384e-05,
"loss": 1.4984,
"num_input_tokens_seen": 2561015808,
"step": 2171
},
{
"epoch": 0.06747670166332428,
"grad_norm": 0.7734375,
"learning_rate": 7.10468523142062e-05,
"loss": 1.5935,
"num_input_tokens_seen": 2562195456,
"step": 2172
},
{
"epoch": 0.06759466792497346,
"grad_norm": 0.76171875,
"learning_rate": 7.10370121400107e-05,
"loss": 1.4926,
"num_input_tokens_seen": 2563375104,
"step": 2173
},
{
"epoch": 0.06771263418662263,
"grad_norm": 0.8671875,
"learning_rate": 7.102716724348449e-05,
"loss": 1.45,
"num_input_tokens_seen": 2564554752,
"step": 2174
},
{
"epoch": 0.0678306004482718,
"grad_norm": 0.6484375,
"learning_rate": 7.101731762612554e-05,
"loss": 1.6375,
"num_input_tokens_seen": 2565734400,
"step": 2175
},
{
"epoch": 0.06794856670992096,
"grad_norm": 0.88671875,
"learning_rate": 7.100746328943245e-05,
"loss": 1.3712,
"num_input_tokens_seen": 2566914048,
"step": 2176
},
{
"epoch": 0.06806653297157013,
"grad_norm": 0.65625,
"learning_rate": 7.099760423490457e-05,
"loss": 1.5398,
"num_input_tokens_seen": 2568093696,
"step": 2177
},
{
"epoch": 0.0681844992332193,
"grad_norm": 0.83203125,
"learning_rate": 7.098774046404199e-05,
"loss": 1.5061,
"num_input_tokens_seen": 2569273344,
"step": 2178
},
{
"epoch": 0.06830246549486847,
"grad_norm": 0.69140625,
"learning_rate": 7.09778719783455e-05,
"loss": 1.4503,
"num_input_tokens_seen": 2570452992,
"step": 2179
},
{
"epoch": 0.06842043175651763,
"grad_norm": 0.8515625,
"learning_rate": 7.096799877931659e-05,
"loss": 1.6409,
"num_input_tokens_seen": 2571632640,
"step": 2180
},
{
"epoch": 0.0685383980181668,
"grad_norm": 0.671875,
"learning_rate": 7.095812086845749e-05,
"loss": 1.6075,
"num_input_tokens_seen": 2572812288,
"step": 2181
},
{
"epoch": 0.06865636427981597,
"grad_norm": 0.78515625,
"learning_rate": 7.094823824727114e-05,
"loss": 1.4472,
"num_input_tokens_seen": 2573991936,
"step": 2182
},
{
"epoch": 0.06877433054146515,
"grad_norm": 0.6796875,
"learning_rate": 7.09383509172612e-05,
"loss": 1.4211,
"num_input_tokens_seen": 2575171584,
"step": 2183
},
{
"epoch": 0.0688922968031143,
"grad_norm": 0.70703125,
"learning_rate": 7.092845887993201e-05,
"loss": 1.5067,
"num_input_tokens_seen": 2576351232,
"step": 2184
},
{
"epoch": 0.06901026306476347,
"grad_norm": 0.68359375,
"learning_rate": 7.091856213678872e-05,
"loss": 1.6708,
"num_input_tokens_seen": 2577530880,
"step": 2185
},
{
"epoch": 0.06912822932641265,
"grad_norm": 0.69921875,
"learning_rate": 7.09086606893371e-05,
"loss": 1.5261,
"num_input_tokens_seen": 2578710528,
"step": 2186
},
{
"epoch": 0.06924619558806182,
"grad_norm": 0.7109375,
"learning_rate": 7.089875453908366e-05,
"loss": 1.476,
"num_input_tokens_seen": 2579890176,
"step": 2187
},
{
"epoch": 0.06936416184971098,
"grad_norm": 0.8046875,
"learning_rate": 7.088884368753566e-05,
"loss": 1.587,
"num_input_tokens_seen": 2581069824,
"step": 2188
},
{
"epoch": 0.06948212811136015,
"grad_norm": 0.68359375,
"learning_rate": 7.087892813620104e-05,
"loss": 1.52,
"num_input_tokens_seen": 2582249472,
"step": 2189
},
{
"epoch": 0.06960009437300932,
"grad_norm": 0.8515625,
"learning_rate": 7.086900788658848e-05,
"loss": 1.4511,
"num_input_tokens_seen": 2583429120,
"step": 2190
},
{
"epoch": 0.06971806063465849,
"grad_norm": 0.61328125,
"learning_rate": 7.085908294020734e-05,
"loss": 1.5586,
"num_input_tokens_seen": 2584608768,
"step": 2191
},
{
"epoch": 0.06983602689630766,
"grad_norm": 0.71484375,
"learning_rate": 7.084915329856773e-05,
"loss": 1.3686,
"num_input_tokens_seen": 2585788416,
"step": 2192
},
{
"epoch": 0.06995399315795682,
"grad_norm": 0.65625,
"learning_rate": 7.083921896318045e-05,
"loss": 1.5593,
"num_input_tokens_seen": 2586968064,
"step": 2193
},
{
"epoch": 0.07007195941960599,
"grad_norm": 0.65234375,
"learning_rate": 7.082927993555704e-05,
"loss": 1.4251,
"num_input_tokens_seen": 2588147712,
"step": 2194
},
{
"epoch": 0.07018992568125516,
"grad_norm": 0.703125,
"learning_rate": 7.081933621720973e-05,
"loss": 1.5143,
"num_input_tokens_seen": 2589327360,
"step": 2195
},
{
"epoch": 0.07030789194290434,
"grad_norm": 0.6953125,
"learning_rate": 7.080938780965148e-05,
"loss": 1.6101,
"num_input_tokens_seen": 2590507008,
"step": 2196
},
{
"epoch": 0.0704258582045535,
"grad_norm": 0.6640625,
"learning_rate": 7.079943471439593e-05,
"loss": 1.5199,
"num_input_tokens_seen": 2591686656,
"step": 2197
},
{
"epoch": 0.07054382446620266,
"grad_norm": 0.67578125,
"learning_rate": 7.078947693295751e-05,
"loss": 1.6252,
"num_input_tokens_seen": 2592866304,
"step": 2198
},
{
"epoch": 0.07066179072785184,
"grad_norm": 0.73046875,
"learning_rate": 7.077951446685128e-05,
"loss": 1.5631,
"num_input_tokens_seen": 2594045952,
"step": 2199
},
{
"epoch": 0.07077975698950101,
"grad_norm": 0.58984375,
"learning_rate": 7.076954731759302e-05,
"loss": 1.6381,
"num_input_tokens_seen": 2595225600,
"step": 2200
},
{
"epoch": 0.07077975698950101,
"eval_wikipedia_loss": 2.2444777488708496,
"eval_wikipedia_runtime": 162.8868,
"eval_wikipedia_samples_per_second": 4.31,
"eval_wikipedia_steps_per_second": 0.184,
"num_input_tokens_seen": 2595225600,
"step": 2200
},
{
"epoch": 0.07077975698950101,
"eval_toxicity_loss": 4.000207901000977,
"eval_toxicity_runtime": 1.0596,
"eval_toxicity_samples_per_second": 1.887,
"eval_toxicity_steps_per_second": 0.944,
"num_input_tokens_seen": 2595225600,
"step": 2200
}
],
"logging_steps": 1,
"max_steps": 8477,
"num_input_tokens_seen": 2595225600,
"num_train_epochs": 9223372036854775807,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.15244846417707e+19,
"train_batch_size": 6,
"trial_name": null,
"trial_params": null
}