deepspeed_no_offload / trainer_state.json
sedrickkeh's picture
End of training
e57a620 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.980891719745223,
"eval_steps": 500,
"global_step": 156,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01910828025477707,
"grad_norm": 9.240381663351364,
"learning_rate": 6.25e-07,
"loss": 1.3429,
"step": 1
},
{
"epoch": 0.03821656050955414,
"grad_norm": 9.47844028531301,
"learning_rate": 1.25e-06,
"loss": 1.3616,
"step": 2
},
{
"epoch": 0.05732484076433121,
"grad_norm": 8.827171685141213,
"learning_rate": 1.8750000000000003e-06,
"loss": 1.296,
"step": 3
},
{
"epoch": 0.07643312101910828,
"grad_norm": 8.794353970740918,
"learning_rate": 2.5e-06,
"loss": 1.3065,
"step": 4
},
{
"epoch": 0.09554140127388536,
"grad_norm": 8.380185798467306,
"learning_rate": 3.125e-06,
"loss": 1.3084,
"step": 5
},
{
"epoch": 0.11464968152866242,
"grad_norm": 6.300000500843309,
"learning_rate": 3.7500000000000005e-06,
"loss": 1.2615,
"step": 6
},
{
"epoch": 0.1337579617834395,
"grad_norm": 3.91637274750174,
"learning_rate": 4.3750000000000005e-06,
"loss": 1.1771,
"step": 7
},
{
"epoch": 0.15286624203821655,
"grad_norm": 3.4401867651997944,
"learning_rate": 5e-06,
"loss": 1.2099,
"step": 8
},
{
"epoch": 0.17197452229299362,
"grad_norm": 4.043563846285618,
"learning_rate": 5.625e-06,
"loss": 1.1225,
"step": 9
},
{
"epoch": 0.1910828025477707,
"grad_norm": 4.249618394544418,
"learning_rate": 6.25e-06,
"loss": 1.0774,
"step": 10
},
{
"epoch": 0.21019108280254778,
"grad_norm": 4.073391370064329,
"learning_rate": 6.875e-06,
"loss": 1.0759,
"step": 11
},
{
"epoch": 0.22929936305732485,
"grad_norm": 3.435634336711639,
"learning_rate": 7.500000000000001e-06,
"loss": 1.0191,
"step": 12
},
{
"epoch": 0.2484076433121019,
"grad_norm": 3.8282958871309263,
"learning_rate": 8.125000000000001e-06,
"loss": 1.0475,
"step": 13
},
{
"epoch": 0.267515923566879,
"grad_norm": 3.067578210234905,
"learning_rate": 8.750000000000001e-06,
"loss": 1.0031,
"step": 14
},
{
"epoch": 0.28662420382165604,
"grad_norm": 2.199676630480871,
"learning_rate": 9.375000000000001e-06,
"loss": 0.9644,
"step": 15
},
{
"epoch": 0.3057324840764331,
"grad_norm": 2.0670790352065835,
"learning_rate": 1e-05,
"loss": 0.958,
"step": 16
},
{
"epoch": 0.3248407643312102,
"grad_norm": 2.5313063928601958,
"learning_rate": 9.998741174712534e-06,
"loss": 0.9513,
"step": 17
},
{
"epoch": 0.34394904458598724,
"grad_norm": 2.3988232851885085,
"learning_rate": 9.994965332706574e-06,
"loss": 0.9529,
"step": 18
},
{
"epoch": 0.3630573248407643,
"grad_norm": 1.8319570038928779,
"learning_rate": 9.98867437523228e-06,
"loss": 0.9513,
"step": 19
},
{
"epoch": 0.3821656050955414,
"grad_norm": 1.4107983554325916,
"learning_rate": 9.979871469976197e-06,
"loss": 0.9268,
"step": 20
},
{
"epoch": 0.4012738853503185,
"grad_norm": 1.4418242532246308,
"learning_rate": 9.968561049466214e-06,
"loss": 0.9387,
"step": 21
},
{
"epoch": 0.42038216560509556,
"grad_norm": 1.4778212495333354,
"learning_rate": 9.954748808839675e-06,
"loss": 0.8954,
"step": 22
},
{
"epoch": 0.4394904458598726,
"grad_norm": 1.1698382833995071,
"learning_rate": 9.938441702975689e-06,
"loss": 0.8937,
"step": 23
},
{
"epoch": 0.4585987261146497,
"grad_norm": 1.0143850225956845,
"learning_rate": 9.91964794299315e-06,
"loss": 0.9142,
"step": 24
},
{
"epoch": 0.47770700636942676,
"grad_norm": 1.1223442487912831,
"learning_rate": 9.898376992116179e-06,
"loss": 0.8878,
"step": 25
},
{
"epoch": 0.4968152866242038,
"grad_norm": 1.1054280647625254,
"learning_rate": 9.874639560909118e-06,
"loss": 0.8914,
"step": 26
},
{
"epoch": 0.5159235668789809,
"grad_norm": 1.1520520690955263,
"learning_rate": 9.848447601883436e-06,
"loss": 0.918,
"step": 27
},
{
"epoch": 0.535031847133758,
"grad_norm": 1.0410238584436573,
"learning_rate": 9.819814303479268e-06,
"loss": 0.9225,
"step": 28
},
{
"epoch": 0.554140127388535,
"grad_norm": 1.0490555259773153,
"learning_rate": 9.788754083424654e-06,
"loss": 0.8652,
"step": 29
},
{
"epoch": 0.5732484076433121,
"grad_norm": 0.9630257990162795,
"learning_rate": 9.755282581475769e-06,
"loss": 0.8733,
"step": 30
},
{
"epoch": 0.5923566878980892,
"grad_norm": 0.9430620751151452,
"learning_rate": 9.719416651541839e-06,
"loss": 0.8964,
"step": 31
},
{
"epoch": 0.6114649681528662,
"grad_norm": 1.0245211883752072,
"learning_rate": 9.681174353198687e-06,
"loss": 0.896,
"step": 32
},
{
"epoch": 0.6305732484076433,
"grad_norm": 0.983726807960442,
"learning_rate": 9.640574942595195e-06,
"loss": 0.8807,
"step": 33
},
{
"epoch": 0.6496815286624203,
"grad_norm": 0.974877189058907,
"learning_rate": 9.597638862757255e-06,
"loss": 0.8396,
"step": 34
},
{
"epoch": 0.6687898089171974,
"grad_norm": 0.9870915751264683,
"learning_rate": 9.552387733294081e-06,
"loss": 0.8644,
"step": 35
},
{
"epoch": 0.6878980891719745,
"grad_norm": 1.007133010597439,
"learning_rate": 9.504844339512096e-06,
"loss": 0.8791,
"step": 36
},
{
"epoch": 0.7070063694267515,
"grad_norm": 0.9569381973911784,
"learning_rate": 9.45503262094184e-06,
"loss": 0.8776,
"step": 37
},
{
"epoch": 0.7261146496815286,
"grad_norm": 1.0165148690890629,
"learning_rate": 9.40297765928369e-06,
"loss": 0.8911,
"step": 38
},
{
"epoch": 0.7452229299363057,
"grad_norm": 0.9513014018799283,
"learning_rate": 9.348705665778479e-06,
"loss": 0.8545,
"step": 39
},
{
"epoch": 0.7643312101910829,
"grad_norm": 0.9560755511481084,
"learning_rate": 9.292243968009332e-06,
"loss": 0.8032,
"step": 40
},
{
"epoch": 0.7834394904458599,
"grad_norm": 1.0110726824374339,
"learning_rate": 9.233620996141421e-06,
"loss": 0.8763,
"step": 41
},
{
"epoch": 0.802547770700637,
"grad_norm": 0.927700813126278,
"learning_rate": 9.172866268606514e-06,
"loss": 0.8522,
"step": 42
},
{
"epoch": 0.821656050955414,
"grad_norm": 0.9883917413731951,
"learning_rate": 9.110010377239552e-06,
"loss": 0.8872,
"step": 43
},
{
"epoch": 0.8407643312101911,
"grad_norm": 0.9665141899298249,
"learning_rate": 9.045084971874738e-06,
"loss": 0.8448,
"step": 44
},
{
"epoch": 0.8598726114649682,
"grad_norm": 0.9915493339603894,
"learning_rate": 8.978122744408905e-06,
"loss": 0.8585,
"step": 45
},
{
"epoch": 0.8789808917197452,
"grad_norm": 1.0555592931701923,
"learning_rate": 8.90915741234015e-06,
"loss": 0.8294,
"step": 46
},
{
"epoch": 0.8980891719745223,
"grad_norm": 0.9701224175594678,
"learning_rate": 8.838223701790057e-06,
"loss": 0.8258,
"step": 47
},
{
"epoch": 0.9171974522292994,
"grad_norm": 1.0295314670496143,
"learning_rate": 8.765357330018056e-06,
"loss": 0.823,
"step": 48
},
{
"epoch": 0.9363057324840764,
"grad_norm": 1.0462850419970708,
"learning_rate": 8.690594987436705e-06,
"loss": 0.8533,
"step": 49
},
{
"epoch": 0.9554140127388535,
"grad_norm": 1.0071857259726584,
"learning_rate": 8.613974319136959e-06,
"loss": 0.8479,
"step": 50
},
{
"epoch": 0.9745222929936306,
"grad_norm": 1.0900740999376952,
"learning_rate": 8.535533905932739e-06,
"loss": 0.871,
"step": 51
},
{
"epoch": 0.9936305732484076,
"grad_norm": 0.9679556144678486,
"learning_rate": 8.455313244934324e-06,
"loss": 0.8503,
"step": 52
},
{
"epoch": 1.0127388535031847,
"grad_norm": 1.8195574679031417,
"learning_rate": 8.373352729660373e-06,
"loss": 1.2777,
"step": 53
},
{
"epoch": 1.0318471337579618,
"grad_norm": 1.1641905698090835,
"learning_rate": 8.289693629698564e-06,
"loss": 0.7338,
"step": 54
},
{
"epoch": 1.0509554140127388,
"grad_norm": 0.9459142079486664,
"learning_rate": 8.204378069925121e-06,
"loss": 0.7477,
"step": 55
},
{
"epoch": 1.070063694267516,
"grad_norm": 0.9602565324416047,
"learning_rate": 8.117449009293668e-06,
"loss": 0.7373,
"step": 56
},
{
"epoch": 1.089171974522293,
"grad_norm": 0.9187451347193307,
"learning_rate": 8.0289502192041e-06,
"loss": 0.6939,
"step": 57
},
{
"epoch": 1.10828025477707,
"grad_norm": 1.1085993724743657,
"learning_rate": 7.938926261462366e-06,
"loss": 0.7188,
"step": 58
},
{
"epoch": 1.127388535031847,
"grad_norm": 0.8986379078032704,
"learning_rate": 7.84742246584226e-06,
"loss": 0.7351,
"step": 59
},
{
"epoch": 1.1464968152866242,
"grad_norm": 0.886161757685445,
"learning_rate": 7.754484907260513e-06,
"loss": 0.6817,
"step": 60
},
{
"epoch": 1.1656050955414012,
"grad_norm": 0.9138244534139465,
"learning_rate": 7.660160382576683e-06,
"loss": 0.7569,
"step": 61
},
{
"epoch": 1.1847133757961783,
"grad_norm": 0.9459351084860834,
"learning_rate": 7.564496387029532e-06,
"loss": 0.6861,
"step": 62
},
{
"epoch": 1.2038216560509554,
"grad_norm": 0.8915088341422834,
"learning_rate": 7.467541090321735e-06,
"loss": 0.7372,
"step": 63
},
{
"epoch": 1.2229299363057324,
"grad_norm": 0.9663235478219123,
"learning_rate": 7.369343312364994e-06,
"loss": 0.6683,
"step": 64
},
{
"epoch": 1.2420382165605095,
"grad_norm": 0.9732340633103796,
"learning_rate": 7.269952498697734e-06,
"loss": 0.7442,
"step": 65
},
{
"epoch": 1.2611464968152866,
"grad_norm": 0.9390914186947825,
"learning_rate": 7.169418695587791e-06,
"loss": 0.721,
"step": 66
},
{
"epoch": 1.2802547770700636,
"grad_norm": 0.8512256422683987,
"learning_rate": 7.067792524832604e-06,
"loss": 0.7401,
"step": 67
},
{
"epoch": 1.2993630573248407,
"grad_norm": 0.8768301826222742,
"learning_rate": 6.965125158269619e-06,
"loss": 0.689,
"step": 68
},
{
"epoch": 1.3184713375796178,
"grad_norm": 0.932801688047002,
"learning_rate": 6.8614682920097265e-06,
"loss": 0.7508,
"step": 69
},
{
"epoch": 1.3375796178343948,
"grad_norm": 0.7967853762224638,
"learning_rate": 6.7568741204067145e-06,
"loss": 0.6699,
"step": 70
},
{
"epoch": 1.356687898089172,
"grad_norm": 0.9103430262001018,
"learning_rate": 6.651395309775837e-06,
"loss": 0.7125,
"step": 71
},
{
"epoch": 1.3757961783439492,
"grad_norm": 0.7578359353459057,
"learning_rate": 6.545084971874738e-06,
"loss": 0.6811,
"step": 72
},
{
"epoch": 1.394904458598726,
"grad_norm": 0.8188167405076486,
"learning_rate": 6.437996637160086e-06,
"loss": 0.7014,
"step": 73
},
{
"epoch": 1.4140127388535033,
"grad_norm": 0.8195141528906517,
"learning_rate": 6.330184227833376e-06,
"loss": 0.6723,
"step": 74
},
{
"epoch": 1.4331210191082802,
"grad_norm": 0.8171369318549622,
"learning_rate": 6.2217020306894705e-06,
"loss": 0.6753,
"step": 75
},
{
"epoch": 1.4522292993630574,
"grad_norm": 0.7577464284493493,
"learning_rate": 6.112604669781572e-06,
"loss": 0.703,
"step": 76
},
{
"epoch": 1.4713375796178343,
"grad_norm": 0.8125996234129923,
"learning_rate": 6.002947078916365e-06,
"loss": 0.7443,
"step": 77
},
{
"epoch": 1.4904458598726116,
"grad_norm": 0.7728841086379103,
"learning_rate": 5.892784473993184e-06,
"loss": 0.6685,
"step": 78
},
{
"epoch": 1.5095541401273884,
"grad_norm": 0.7612625352676844,
"learning_rate": 5.782172325201155e-06,
"loss": 0.6591,
"step": 79
},
{
"epoch": 1.5286624203821657,
"grad_norm": 0.8001204845967484,
"learning_rate": 5.671166329088278e-06,
"loss": 0.6936,
"step": 80
},
{
"epoch": 1.5477707006369426,
"grad_norm": 0.7912008740596617,
"learning_rate": 5.559822380516539e-06,
"loss": 0.6929,
"step": 81
},
{
"epoch": 1.5668789808917198,
"grad_norm": 0.7923380310923759,
"learning_rate": 5.448196544517168e-06,
"loss": 0.7198,
"step": 82
},
{
"epoch": 1.5859872611464967,
"grad_norm": 0.8423688254953324,
"learning_rate": 5.336345028060199e-06,
"loss": 0.7355,
"step": 83
},
{
"epoch": 1.605095541401274,
"grad_norm": 0.7791539359416605,
"learning_rate": 5.224324151752575e-06,
"loss": 0.6984,
"step": 84
},
{
"epoch": 1.6242038216560508,
"grad_norm": 0.823125420938912,
"learning_rate": 5.112190321479026e-06,
"loss": 0.7028,
"step": 85
},
{
"epoch": 1.643312101910828,
"grad_norm": 0.752691830825963,
"learning_rate": 5e-06,
"loss": 0.666,
"step": 86
},
{
"epoch": 1.662420382165605,
"grad_norm": 0.8221262952489863,
"learning_rate": 4.887809678520976e-06,
"loss": 0.7537,
"step": 87
},
{
"epoch": 1.6815286624203822,
"grad_norm": 0.8520082611749908,
"learning_rate": 4.775675848247427e-06,
"loss": 0.7102,
"step": 88
},
{
"epoch": 1.700636942675159,
"grad_norm": 0.847120677130926,
"learning_rate": 4.663654971939802e-06,
"loss": 0.6781,
"step": 89
},
{
"epoch": 1.7197452229299364,
"grad_norm": 0.7545040428235671,
"learning_rate": 4.551803455482833e-06,
"loss": 0.7141,
"step": 90
},
{
"epoch": 1.7388535031847132,
"grad_norm": 0.8058099820461135,
"learning_rate": 4.4401776194834615e-06,
"loss": 0.7032,
"step": 91
},
{
"epoch": 1.7579617834394905,
"grad_norm": 0.8249735439437605,
"learning_rate": 4.3288336709117246e-06,
"loss": 0.7434,
"step": 92
},
{
"epoch": 1.7770700636942676,
"grad_norm": 0.7799760379329518,
"learning_rate": 4.217827674798845e-06,
"loss": 0.68,
"step": 93
},
{
"epoch": 1.7961783439490446,
"grad_norm": 0.7469859435072883,
"learning_rate": 4.107215526006818e-06,
"loss": 0.6568,
"step": 94
},
{
"epoch": 1.8152866242038217,
"grad_norm": 0.7529913774470228,
"learning_rate": 3.997052921083637e-06,
"loss": 0.7393,
"step": 95
},
{
"epoch": 1.8343949044585988,
"grad_norm": 0.7899361356445419,
"learning_rate": 3.887395330218429e-06,
"loss": 0.6979,
"step": 96
},
{
"epoch": 1.8535031847133758,
"grad_norm": 0.7747195837460693,
"learning_rate": 3.778297969310529e-06,
"loss": 0.7,
"step": 97
},
{
"epoch": 1.872611464968153,
"grad_norm": 0.8698671177115019,
"learning_rate": 3.669815772166625e-06,
"loss": 0.7173,
"step": 98
},
{
"epoch": 1.89171974522293,
"grad_norm": 0.7846154352930541,
"learning_rate": 3.562003362839914e-06,
"loss": 0.6773,
"step": 99
},
{
"epoch": 1.910828025477707,
"grad_norm": 0.8460944964935945,
"learning_rate": 3.4549150281252635e-06,
"loss": 0.6369,
"step": 100
},
{
"epoch": 1.929936305732484,
"grad_norm": 0.8438802655163977,
"learning_rate": 3.3486046902241663e-06,
"loss": 0.7553,
"step": 101
},
{
"epoch": 1.9490445859872612,
"grad_norm": 0.7832498480226588,
"learning_rate": 3.2431258795932863e-06,
"loss": 0.6749,
"step": 102
},
{
"epoch": 1.9681528662420382,
"grad_norm": 0.7006878581759661,
"learning_rate": 3.1385317079902743e-06,
"loss": 0.6918,
"step": 103
},
{
"epoch": 1.9872611464968153,
"grad_norm": 0.8106734791225854,
"learning_rate": 3.0348748417303826e-06,
"loss": 0.6841,
"step": 104
},
{
"epoch": 2.0063694267515926,
"grad_norm": 1.6133976736380617,
"learning_rate": 2.932207475167398e-06,
"loss": 1.1013,
"step": 105
},
{
"epoch": 2.0254777070063694,
"grad_norm": 0.8470823367059515,
"learning_rate": 2.83058130441221e-06,
"loss": 0.6537,
"step": 106
},
{
"epoch": 2.0445859872611467,
"grad_norm": 0.8385507154200827,
"learning_rate": 2.7300475013022666e-06,
"loss": 0.5645,
"step": 107
},
{
"epoch": 2.0636942675159236,
"grad_norm": 0.7303000642105223,
"learning_rate": 2.6306566876350072e-06,
"loss": 0.5989,
"step": 108
},
{
"epoch": 2.082802547770701,
"grad_norm": 0.6857970192104155,
"learning_rate": 2.532458909678266e-06,
"loss": 0.574,
"step": 109
},
{
"epoch": 2.1019108280254777,
"grad_norm": 0.7250126564429323,
"learning_rate": 2.43550361297047e-06,
"loss": 0.6267,
"step": 110
},
{
"epoch": 2.121019108280255,
"grad_norm": 0.7792904891391182,
"learning_rate": 2.339839617423318e-06,
"loss": 0.635,
"step": 111
},
{
"epoch": 2.140127388535032,
"grad_norm": 0.7485612751031683,
"learning_rate": 2.245515092739488e-06,
"loss": 0.6011,
"step": 112
},
{
"epoch": 2.159235668789809,
"grad_norm": 0.8315785716764414,
"learning_rate": 2.1525775341577404e-06,
"loss": 0.6377,
"step": 113
},
{
"epoch": 2.178343949044586,
"grad_norm": 0.724833434775396,
"learning_rate": 2.061073738537635e-06,
"loss": 0.5621,
"step": 114
},
{
"epoch": 2.1974522292993632,
"grad_norm": 0.69070701526373,
"learning_rate": 1.971049780795901e-06,
"loss": 0.5837,
"step": 115
},
{
"epoch": 2.21656050955414,
"grad_norm": 0.7007382760550266,
"learning_rate": 1.8825509907063328e-06,
"loss": 0.5976,
"step": 116
},
{
"epoch": 2.2356687898089174,
"grad_norm": 0.7330313492544757,
"learning_rate": 1.7956219300748796e-06,
"loss": 0.6231,
"step": 117
},
{
"epoch": 2.254777070063694,
"grad_norm": 0.6790323737031737,
"learning_rate": 1.7103063703014372e-06,
"loss": 0.5705,
"step": 118
},
{
"epoch": 2.2738853503184715,
"grad_norm": 0.719926750707592,
"learning_rate": 1.6266472703396286e-06,
"loss": 0.6073,
"step": 119
},
{
"epoch": 2.2929936305732483,
"grad_norm": 0.7465439468628049,
"learning_rate": 1.544686755065677e-06,
"loss": 0.6255,
"step": 120
},
{
"epoch": 2.3121019108280256,
"grad_norm": 0.739535044631204,
"learning_rate": 1.4644660940672628e-06,
"loss": 0.5905,
"step": 121
},
{
"epoch": 2.3312101910828025,
"grad_norm": 0.7475970852807615,
"learning_rate": 1.3860256808630429e-06,
"loss": 0.5853,
"step": 122
},
{
"epoch": 2.3503184713375798,
"grad_norm": 0.715061032596014,
"learning_rate": 1.3094050125632973e-06,
"loss": 0.6085,
"step": 123
},
{
"epoch": 2.3694267515923566,
"grad_norm": 0.6573301240311847,
"learning_rate": 1.234642669981946e-06,
"loss": 0.5548,
"step": 124
},
{
"epoch": 2.388535031847134,
"grad_norm": 0.8085276652639457,
"learning_rate": 1.1617762982099446e-06,
"loss": 0.6369,
"step": 125
},
{
"epoch": 2.4076433121019107,
"grad_norm": 0.6833172235104982,
"learning_rate": 1.0908425876598512e-06,
"loss": 0.5646,
"step": 126
},
{
"epoch": 2.426751592356688,
"grad_norm": 0.7022677730268799,
"learning_rate": 1.0218772555910955e-06,
"loss": 0.6004,
"step": 127
},
{
"epoch": 2.445859872611465,
"grad_norm": 0.7343299314903051,
"learning_rate": 9.549150281252633e-07,
"loss": 0.6111,
"step": 128
},
{
"epoch": 2.464968152866242,
"grad_norm": 0.7391768076193667,
"learning_rate": 8.899896227604509e-07,
"loss": 0.6004,
"step": 129
},
{
"epoch": 2.484076433121019,
"grad_norm": 0.6750818571924504,
"learning_rate": 8.271337313934869e-07,
"loss": 0.5499,
"step": 130
},
{
"epoch": 2.5031847133757963,
"grad_norm": 0.7346910621114229,
"learning_rate": 7.663790038585794e-07,
"loss": 0.6805,
"step": 131
},
{
"epoch": 2.522292993630573,
"grad_norm": 0.6669176805767499,
"learning_rate": 7.077560319906696e-07,
"loss": 0.5708,
"step": 132
},
{
"epoch": 2.5414012738853504,
"grad_norm": 0.7225129844356113,
"learning_rate": 6.512943342215234e-07,
"loss": 0.6088,
"step": 133
},
{
"epoch": 2.5605095541401273,
"grad_norm": 0.6546072693274814,
"learning_rate": 5.9702234071631e-07,
"loss": 0.558,
"step": 134
},
{
"epoch": 2.5796178343949046,
"grad_norm": 0.6814514799540362,
"learning_rate": 5.449673790581611e-07,
"loss": 0.5846,
"step": 135
},
{
"epoch": 2.5987261146496814,
"grad_norm": 0.6780900122530871,
"learning_rate": 4.951556604879049e-07,
"loss": 0.5835,
"step": 136
},
{
"epoch": 2.6178343949044587,
"grad_norm": 0.7002320873890087,
"learning_rate": 4.4761226670592074e-07,
"loss": 0.5937,
"step": 137
},
{
"epoch": 2.6369426751592355,
"grad_norm": 0.6874340650991541,
"learning_rate": 4.0236113724274716e-07,
"loss": 0.5975,
"step": 138
},
{
"epoch": 2.656050955414013,
"grad_norm": 0.6641836250975586,
"learning_rate": 3.5942505740480583e-07,
"loss": 0.5917,
"step": 139
},
{
"epoch": 2.6751592356687897,
"grad_norm": 0.6883775490890136,
"learning_rate": 3.18825646801314e-07,
"loss": 0.58,
"step": 140
},
{
"epoch": 2.694267515923567,
"grad_norm": 0.6879467238286184,
"learning_rate": 2.8058334845816214e-07,
"loss": 0.6181,
"step": 141
},
{
"epoch": 2.713375796178344,
"grad_norm": 0.6844028205746674,
"learning_rate": 2.447174185242324e-07,
"loss": 0.6212,
"step": 142
},
{
"epoch": 2.732484076433121,
"grad_norm": 0.691436740082704,
"learning_rate": 2.1124591657534776e-07,
"loss": 0.6067,
"step": 143
},
{
"epoch": 2.7515923566878984,
"grad_norm": 0.671783754173785,
"learning_rate": 1.801856965207338e-07,
"loss": 0.545,
"step": 144
},
{
"epoch": 2.770700636942675,
"grad_norm": 0.6625200282153909,
"learning_rate": 1.5155239811656562e-07,
"loss": 0.5746,
"step": 145
},
{
"epoch": 2.789808917197452,
"grad_norm": 0.7006220896747084,
"learning_rate": 1.253604390908819e-07,
"loss": 0.6175,
"step": 146
},
{
"epoch": 2.8089171974522293,
"grad_norm": 0.7147369953238405,
"learning_rate": 1.0162300788382263e-07,
"loss": 0.6413,
"step": 147
},
{
"epoch": 2.8280254777070066,
"grad_norm": 0.6489688568703844,
"learning_rate": 8.035205700685167e-08,
"loss": 0.5595,
"step": 148
},
{
"epoch": 2.8471337579617835,
"grad_norm": 0.6268863182240232,
"learning_rate": 6.15582970243117e-08,
"loss": 0.5932,
"step": 149
},
{
"epoch": 2.8662420382165603,
"grad_norm": 0.666115132279766,
"learning_rate": 4.52511911603265e-08,
"loss": 0.583,
"step": 150
},
{
"epoch": 2.8853503184713376,
"grad_norm": 0.6576241181862866,
"learning_rate": 3.143895053378698e-08,
"loss": 0.5961,
"step": 151
},
{
"epoch": 2.904458598726115,
"grad_norm": 0.7017623706064487,
"learning_rate": 2.012853002380466e-08,
"loss": 0.5726,
"step": 152
},
{
"epoch": 2.9235668789808917,
"grad_norm": 0.7106099205481908,
"learning_rate": 1.132562476771959e-08,
"loss": 0.6533,
"step": 153
},
{
"epoch": 2.9426751592356686,
"grad_norm": 0.6652256303204708,
"learning_rate": 5.034667293427053e-09,
"loss": 0.5776,
"step": 154
},
{
"epoch": 2.961783439490446,
"grad_norm": 0.6786592086257924,
"learning_rate": 1.2588252874673469e-09,
"loss": 0.5784,
"step": 155
},
{
"epoch": 2.980891719745223,
"grad_norm": 0.6545650863305297,
"learning_rate": 0.0,
"loss": 0.6064,
"step": 156
},
{
"epoch": 2.980891719745223,
"step": 156,
"total_flos": 55447011295232.0,
"train_loss": 0.7628553406550334,
"train_runtime": 1911.8295,
"train_samples_per_second": 7.847,
"train_steps_per_second": 0.082
}
],
"logging_steps": 1.0,
"max_steps": 156,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 55447011295232.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}