nttx's picture
Training in progress, step 200, checkpoint
d83aea1 verified
{
"best_metric": 0.427767276763916,
"best_model_checkpoint": "miner_id_24/checkpoint-200",
"epoch": 0.3004130679684566,
"eval_steps": 50,
"global_step": 200,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0015020653398422831,
"grad_norm": 0.48013806343078613,
"learning_rate": 1e-05,
"loss": 1.1844,
"step": 1
},
{
"epoch": 0.0015020653398422831,
"eval_loss": 1.4352283477783203,
"eval_runtime": 33.3798,
"eval_samples_per_second": 33.613,
"eval_steps_per_second": 8.418,
"step": 1
},
{
"epoch": 0.0030041306796845663,
"grad_norm": 0.42095404863357544,
"learning_rate": 2e-05,
"loss": 1.2759,
"step": 2
},
{
"epoch": 0.004506196019526849,
"grad_norm": 0.40659624338150024,
"learning_rate": 3e-05,
"loss": 1.3025,
"step": 3
},
{
"epoch": 0.0060082613593691325,
"grad_norm": 0.41802921891212463,
"learning_rate": 4e-05,
"loss": 1.3053,
"step": 4
},
{
"epoch": 0.007510326699211416,
"grad_norm": 0.4731360673904419,
"learning_rate": 5e-05,
"loss": 1.305,
"step": 5
},
{
"epoch": 0.009012392039053699,
"grad_norm": 0.4001414179801941,
"learning_rate": 6e-05,
"loss": 1.301,
"step": 6
},
{
"epoch": 0.010514457378895982,
"grad_norm": 0.3494016230106354,
"learning_rate": 7e-05,
"loss": 1.2796,
"step": 7
},
{
"epoch": 0.012016522718738265,
"grad_norm": 0.25853636860847473,
"learning_rate": 8e-05,
"loss": 1.2454,
"step": 8
},
{
"epoch": 0.013518588058580548,
"grad_norm": 0.36195915937423706,
"learning_rate": 9e-05,
"loss": 1.2594,
"step": 9
},
{
"epoch": 0.015020653398422831,
"grad_norm": 0.41546186804771423,
"learning_rate": 0.0001,
"loss": 1.2273,
"step": 10
},
{
"epoch": 0.016522718738265114,
"grad_norm": 0.38150355219841003,
"learning_rate": 9.999316524962345e-05,
"loss": 1.1949,
"step": 11
},
{
"epoch": 0.018024784078107398,
"grad_norm": 0.3044074773788452,
"learning_rate": 9.997266286704631e-05,
"loss": 1.1619,
"step": 12
},
{
"epoch": 0.01952684941794968,
"grad_norm": 0.23984934389591217,
"learning_rate": 9.993849845741524e-05,
"loss": 1.1339,
"step": 13
},
{
"epoch": 0.021028914757791964,
"grad_norm": 0.24785546958446503,
"learning_rate": 9.989068136093873e-05,
"loss": 1.0967,
"step": 14
},
{
"epoch": 0.022530980097634247,
"grad_norm": 0.25208961963653564,
"learning_rate": 9.98292246503335e-05,
"loss": 1.074,
"step": 15
},
{
"epoch": 0.02403304543747653,
"grad_norm": 0.26858824491500854,
"learning_rate": 9.975414512725057e-05,
"loss": 1.0323,
"step": 16
},
{
"epoch": 0.025535110777318813,
"grad_norm": 0.2699333131313324,
"learning_rate": 9.966546331768191e-05,
"loss": 1.0193,
"step": 17
},
{
"epoch": 0.027037176117161096,
"grad_norm": 0.29868969321250916,
"learning_rate": 9.956320346634876e-05,
"loss": 0.9937,
"step": 18
},
{
"epoch": 0.02853924145700338,
"grad_norm": 0.2335326373577118,
"learning_rate": 9.944739353007344e-05,
"loss": 0.9666,
"step": 19
},
{
"epoch": 0.030041306796845663,
"grad_norm": 0.25925031304359436,
"learning_rate": 9.931806517013612e-05,
"loss": 0.9465,
"step": 20
},
{
"epoch": 0.031543372136687946,
"grad_norm": 0.3260612189769745,
"learning_rate": 9.917525374361912e-05,
"loss": 0.9378,
"step": 21
},
{
"epoch": 0.03304543747653023,
"grad_norm": 0.2550596594810486,
"learning_rate": 9.901899829374047e-05,
"loss": 0.9038,
"step": 22
},
{
"epoch": 0.03454750281637251,
"grad_norm": 0.28878340125083923,
"learning_rate": 9.884934153917997e-05,
"loss": 0.8962,
"step": 23
},
{
"epoch": 0.036049568156214795,
"grad_norm": 0.27984026074409485,
"learning_rate": 9.86663298624003e-05,
"loss": 0.8538,
"step": 24
},
{
"epoch": 0.03755163349605708,
"grad_norm": 0.2650708556175232,
"learning_rate": 9.847001329696653e-05,
"loss": 0.8394,
"step": 25
},
{
"epoch": 0.03905369883589936,
"grad_norm": 0.24351182579994202,
"learning_rate": 9.826044551386744e-05,
"loss": 0.8104,
"step": 26
},
{
"epoch": 0.040555764175741645,
"grad_norm": 0.3326224088668823,
"learning_rate": 9.803768380684242e-05,
"loss": 0.794,
"step": 27
},
{
"epoch": 0.04205782951558393,
"grad_norm": 0.298579603433609,
"learning_rate": 9.780178907671789e-05,
"loss": 0.7783,
"step": 28
},
{
"epoch": 0.04355989485542621,
"grad_norm": 0.2561696469783783,
"learning_rate": 9.755282581475769e-05,
"loss": 0.7657,
"step": 29
},
{
"epoch": 0.045061960195268494,
"grad_norm": 0.2900291383266449,
"learning_rate": 9.729086208503174e-05,
"loss": 0.7382,
"step": 30
},
{
"epoch": 0.04656402553511078,
"grad_norm": 0.2575244605541229,
"learning_rate": 9.701596950580806e-05,
"loss": 0.723,
"step": 31
},
{
"epoch": 0.04806609087495306,
"grad_norm": 0.34842997789382935,
"learning_rate": 9.672822322997305e-05,
"loss": 0.7148,
"step": 32
},
{
"epoch": 0.04956815621479534,
"grad_norm": 0.2418001890182495,
"learning_rate": 9.642770192448536e-05,
"loss": 0.6728,
"step": 33
},
{
"epoch": 0.051070221554637626,
"grad_norm": 0.37016379833221436,
"learning_rate": 9.611448774886924e-05,
"loss": 0.6626,
"step": 34
},
{
"epoch": 0.05257228689447991,
"grad_norm": 0.3161463439464569,
"learning_rate": 9.578866633275288e-05,
"loss": 0.6567,
"step": 35
},
{
"epoch": 0.05407435223432219,
"grad_norm": 0.4346078038215637,
"learning_rate": 9.545032675245813e-05,
"loss": 0.6694,
"step": 36
},
{
"epoch": 0.055576417574164476,
"grad_norm": 0.435017853975296,
"learning_rate": 9.509956150664796e-05,
"loss": 0.6299,
"step": 37
},
{
"epoch": 0.05707848291400676,
"grad_norm": 0.3490477204322815,
"learning_rate": 9.473646649103818e-05,
"loss": 0.6261,
"step": 38
},
{
"epoch": 0.05858054825384904,
"grad_norm": 0.5194304585456848,
"learning_rate": 9.43611409721806e-05,
"loss": 0.5896,
"step": 39
},
{
"epoch": 0.060082613593691325,
"grad_norm": 0.2708154618740082,
"learning_rate": 9.397368756032445e-05,
"loss": 0.6089,
"step": 40
},
{
"epoch": 0.06158467893353361,
"grad_norm": 0.4544765055179596,
"learning_rate": 9.357421218136386e-05,
"loss": 0.5747,
"step": 41
},
{
"epoch": 0.06308674427337589,
"grad_norm": 0.3197058439254761,
"learning_rate": 9.316282404787871e-05,
"loss": 0.5633,
"step": 42
},
{
"epoch": 0.06458880961321818,
"grad_norm": 0.49233025312423706,
"learning_rate": 9.273963562927695e-05,
"loss": 0.5625,
"step": 43
},
{
"epoch": 0.06609087495306046,
"grad_norm": 0.31287699937820435,
"learning_rate": 9.230476262104677e-05,
"loss": 0.5604,
"step": 44
},
{
"epoch": 0.06759294029290275,
"grad_norm": 0.4544554054737091,
"learning_rate": 9.185832391312644e-05,
"loss": 0.5784,
"step": 45
},
{
"epoch": 0.06909500563274502,
"grad_norm": 0.4347470998764038,
"learning_rate": 9.140044155740101e-05,
"loss": 0.5394,
"step": 46
},
{
"epoch": 0.07059707097258731,
"grad_norm": 0.4225442707538605,
"learning_rate": 9.093124073433463e-05,
"loss": 0.55,
"step": 47
},
{
"epoch": 0.07209913631242959,
"grad_norm": 0.4641411602497101,
"learning_rate": 9.045084971874738e-05,
"loss": 0.5168,
"step": 48
},
{
"epoch": 0.07360120165227188,
"grad_norm": 0.5490090847015381,
"learning_rate": 8.995939984474624e-05,
"loss": 0.545,
"step": 49
},
{
"epoch": 0.07510326699211416,
"grad_norm": 0.5186936855316162,
"learning_rate": 8.945702546981969e-05,
"loss": 0.4797,
"step": 50
},
{
"epoch": 0.07510326699211416,
"eval_loss": 0.9435171484947205,
"eval_runtime": 33.2926,
"eval_samples_per_second": 33.701,
"eval_steps_per_second": 8.44,
"step": 50
},
{
"epoch": 0.07660533233195645,
"grad_norm": 5.346027851104736,
"learning_rate": 8.894386393810563e-05,
"loss": 1.3858,
"step": 51
},
{
"epoch": 0.07810739767179872,
"grad_norm": 4.529382228851318,
"learning_rate": 8.842005554284296e-05,
"loss": 1.1019,
"step": 52
},
{
"epoch": 0.07960946301164101,
"grad_norm": 2.0946102142333984,
"learning_rate": 8.788574348801675e-05,
"loss": 0.7455,
"step": 53
},
{
"epoch": 0.08111152835148329,
"grad_norm": 0.4784291386604309,
"learning_rate": 8.73410738492077e-05,
"loss": 0.6575,
"step": 54
},
{
"epoch": 0.08261359369132558,
"grad_norm": 0.43049320578575134,
"learning_rate": 8.678619553365659e-05,
"loss": 0.6418,
"step": 55
},
{
"epoch": 0.08411565903116786,
"grad_norm": 0.3956131637096405,
"learning_rate": 8.622126023955446e-05,
"loss": 0.6338,
"step": 56
},
{
"epoch": 0.08561772437101015,
"grad_norm": 0.3648943603038788,
"learning_rate": 8.564642241456986e-05,
"loss": 0.6281,
"step": 57
},
{
"epoch": 0.08711978971085242,
"grad_norm": 0.265221506357193,
"learning_rate": 8.506183921362443e-05,
"loss": 0.6154,
"step": 58
},
{
"epoch": 0.08862185505069471,
"grad_norm": 0.3286784887313843,
"learning_rate": 8.44676704559283e-05,
"loss": 0.6275,
"step": 59
},
{
"epoch": 0.09012392039053699,
"grad_norm": 0.36223769187927246,
"learning_rate": 8.386407858128706e-05,
"loss": 0.6223,
"step": 60
},
{
"epoch": 0.09162598573037928,
"grad_norm": 0.2875317931175232,
"learning_rate": 8.32512286056924e-05,
"loss": 0.614,
"step": 61
},
{
"epoch": 0.09312805107022155,
"grad_norm": 0.24351203441619873,
"learning_rate": 8.262928807620843e-05,
"loss": 0.5952,
"step": 62
},
{
"epoch": 0.09463011641006384,
"grad_norm": 0.2812121510505676,
"learning_rate": 8.199842702516583e-05,
"loss": 0.6036,
"step": 63
},
{
"epoch": 0.09613218174990612,
"grad_norm": 0.2832166850566864,
"learning_rate": 8.135881792367686e-05,
"loss": 0.5976,
"step": 64
},
{
"epoch": 0.09763424708974841,
"grad_norm": 0.2672559916973114,
"learning_rate": 8.07106356344834e-05,
"loss": 0.5743,
"step": 65
},
{
"epoch": 0.09913631242959069,
"grad_norm": 0.25090304017066956,
"learning_rate": 8.005405736415126e-05,
"loss": 0.5795,
"step": 66
},
{
"epoch": 0.10063837776943298,
"grad_norm": 0.3040817677974701,
"learning_rate": 7.938926261462366e-05,
"loss": 0.5613,
"step": 67
},
{
"epoch": 0.10214044310927525,
"grad_norm": 0.23814338445663452,
"learning_rate": 7.871643313414718e-05,
"loss": 0.5521,
"step": 68
},
{
"epoch": 0.10364250844911754,
"grad_norm": 0.2542641758918762,
"learning_rate": 7.803575286758364e-05,
"loss": 0.5535,
"step": 69
},
{
"epoch": 0.10514457378895982,
"grad_norm": 0.2717762887477875,
"learning_rate": 7.734740790612136e-05,
"loss": 0.5361,
"step": 70
},
{
"epoch": 0.10664663912880211,
"grad_norm": 0.2743045389652252,
"learning_rate": 7.66515864363997e-05,
"loss": 0.5338,
"step": 71
},
{
"epoch": 0.10814870446864439,
"grad_norm": 0.22834768891334534,
"learning_rate": 7.594847868906076e-05,
"loss": 0.5238,
"step": 72
},
{
"epoch": 0.10965076980848668,
"grad_norm": 0.3098483085632324,
"learning_rate": 7.52382768867422e-05,
"loss": 0.5278,
"step": 73
},
{
"epoch": 0.11115283514832895,
"grad_norm": 0.2830328941345215,
"learning_rate": 7.452117519152542e-05,
"loss": 0.5089,
"step": 74
},
{
"epoch": 0.11265490048817124,
"grad_norm": 0.2570495009422302,
"learning_rate": 7.379736965185368e-05,
"loss": 0.4981,
"step": 75
},
{
"epoch": 0.11415696582801352,
"grad_norm": 0.27931836247444153,
"learning_rate": 7.30670581489344e-05,
"loss": 0.4863,
"step": 76
},
{
"epoch": 0.11565903116785581,
"grad_norm": 0.2561793625354767,
"learning_rate": 7.233044034264034e-05,
"loss": 0.4959,
"step": 77
},
{
"epoch": 0.11716109650769808,
"grad_norm": 0.29976847767829895,
"learning_rate": 7.158771761692464e-05,
"loss": 0.4882,
"step": 78
},
{
"epoch": 0.11866316184754037,
"grad_norm": 0.27593934535980225,
"learning_rate": 7.083909302476453e-05,
"loss": 0.4959,
"step": 79
},
{
"epoch": 0.12016522718738265,
"grad_norm": 0.2488517165184021,
"learning_rate": 7.008477123264848e-05,
"loss": 0.4699,
"step": 80
},
{
"epoch": 0.12166729252722494,
"grad_norm": 0.19852301478385925,
"learning_rate": 6.932495846462261e-05,
"loss": 0.4576,
"step": 81
},
{
"epoch": 0.12316935786706722,
"grad_norm": 0.2308630794286728,
"learning_rate": 6.855986244591104e-05,
"loss": 0.4519,
"step": 82
},
{
"epoch": 0.1246714232069095,
"grad_norm": 0.28317439556121826,
"learning_rate": 6.778969234612584e-05,
"loss": 0.4773,
"step": 83
},
{
"epoch": 0.12617348854675178,
"grad_norm": 0.280638188123703,
"learning_rate": 6.701465872208216e-05,
"loss": 0.4727,
"step": 84
},
{
"epoch": 0.12767555388659407,
"grad_norm": 0.29901838302612305,
"learning_rate": 6.623497346023418e-05,
"loss": 0.463,
"step": 85
},
{
"epoch": 0.12917761922643636,
"grad_norm": 0.32130149006843567,
"learning_rate": 6.545084971874738e-05,
"loss": 0.4546,
"step": 86
},
{
"epoch": 0.13067968456627863,
"grad_norm": 0.24742907285690308,
"learning_rate": 6.466250186922325e-05,
"loss": 0.4519,
"step": 87
},
{
"epoch": 0.13218174990612092,
"grad_norm": 0.2867355942726135,
"learning_rate": 6.387014543809223e-05,
"loss": 0.4445,
"step": 88
},
{
"epoch": 0.1336838152459632,
"grad_norm": 0.27279844880104065,
"learning_rate": 6.307399704769099e-05,
"loss": 0.4328,
"step": 89
},
{
"epoch": 0.1351858805858055,
"grad_norm": 0.25699475407600403,
"learning_rate": 6.227427435703997e-05,
"loss": 0.4397,
"step": 90
},
{
"epoch": 0.13668794592564776,
"grad_norm": 0.2791455090045929,
"learning_rate": 6.147119600233758e-05,
"loss": 0.4523,
"step": 91
},
{
"epoch": 0.13819001126549005,
"grad_norm": 0.3582814335823059,
"learning_rate": 6.066498153718735e-05,
"loss": 0.4553,
"step": 92
},
{
"epoch": 0.13969207660533234,
"grad_norm": 0.28347545862197876,
"learning_rate": 5.985585137257401e-05,
"loss": 0.4314,
"step": 93
},
{
"epoch": 0.14119414194517463,
"grad_norm": 0.2645236551761627,
"learning_rate": 5.90440267166055e-05,
"loss": 0.4493,
"step": 94
},
{
"epoch": 0.1426962072850169,
"grad_norm": 0.308789998292923,
"learning_rate": 5.8229729514036705e-05,
"loss": 0.4487,
"step": 95
},
{
"epoch": 0.14419827262485918,
"grad_norm": 0.27754440903663635,
"learning_rate": 5.74131823855921e-05,
"loss": 0.4236,
"step": 96
},
{
"epoch": 0.14570033796470147,
"grad_norm": 0.29128730297088623,
"learning_rate": 5.6594608567103456e-05,
"loss": 0.4523,
"step": 97
},
{
"epoch": 0.14720240330454376,
"grad_norm": 0.4283706247806549,
"learning_rate": 5.577423184847932e-05,
"loss": 0.4624,
"step": 98
},
{
"epoch": 0.14870446864438602,
"grad_norm": 0.3860466480255127,
"learning_rate": 5.495227651252315e-05,
"loss": 0.4384,
"step": 99
},
{
"epoch": 0.1502065339842283,
"grad_norm": 0.4692773222923279,
"learning_rate": 5.4128967273616625e-05,
"loss": 0.4551,
"step": 100
},
{
"epoch": 0.1502065339842283,
"eval_loss": 0.5659072399139404,
"eval_runtime": 33.2992,
"eval_samples_per_second": 33.694,
"eval_steps_per_second": 8.439,
"step": 100
},
{
"epoch": 0.1517085993240706,
"grad_norm": 2.895529270172119,
"learning_rate": 5.330452921628497e-05,
"loss": 0.7982,
"step": 101
},
{
"epoch": 0.1532106646639129,
"grad_norm": 1.8724274635314941,
"learning_rate": 5.247918773366112e-05,
"loss": 0.6634,
"step": 102
},
{
"epoch": 0.15471273000375516,
"grad_norm": 0.6110711097717285,
"learning_rate": 5.165316846586541e-05,
"loss": 0.5691,
"step": 103
},
{
"epoch": 0.15621479534359745,
"grad_norm": 0.33346492052078247,
"learning_rate": 5.0826697238317935e-05,
"loss": 0.5415,
"step": 104
},
{
"epoch": 0.15771686068343974,
"grad_norm": 0.3243321180343628,
"learning_rate": 5e-05,
"loss": 0.535,
"step": 105
},
{
"epoch": 0.15921892602328203,
"grad_norm": 0.30207017064094543,
"learning_rate": 4.917330276168208e-05,
"loss": 0.5076,
"step": 106
},
{
"epoch": 0.1607209913631243,
"grad_norm": 0.2547878921031952,
"learning_rate": 4.834683153413459e-05,
"loss": 0.5209,
"step": 107
},
{
"epoch": 0.16222305670296658,
"grad_norm": 0.2538004219532013,
"learning_rate": 4.7520812266338885e-05,
"loss": 0.5153,
"step": 108
},
{
"epoch": 0.16372512204280887,
"grad_norm": 0.310922235250473,
"learning_rate": 4.669547078371504e-05,
"loss": 0.5151,
"step": 109
},
{
"epoch": 0.16522718738265116,
"grad_norm": 0.2497279793024063,
"learning_rate": 4.5871032726383386e-05,
"loss": 0.496,
"step": 110
},
{
"epoch": 0.16672925272249342,
"grad_norm": 0.2196495532989502,
"learning_rate": 4.504772348747687e-05,
"loss": 0.4956,
"step": 111
},
{
"epoch": 0.1682313180623357,
"grad_norm": 0.24011313915252686,
"learning_rate": 4.4225768151520694e-05,
"loss": 0.5075,
"step": 112
},
{
"epoch": 0.169733383402178,
"grad_norm": 0.2906111776828766,
"learning_rate": 4.3405391432896555e-05,
"loss": 0.5019,
"step": 113
},
{
"epoch": 0.1712354487420203,
"grad_norm": 0.250466525554657,
"learning_rate": 4.2586817614407895e-05,
"loss": 0.5044,
"step": 114
},
{
"epoch": 0.17273751408186255,
"grad_norm": 0.2569282352924347,
"learning_rate": 4.17702704859633e-05,
"loss": 0.4807,
"step": 115
},
{
"epoch": 0.17423957942170484,
"grad_norm": 0.21970154345035553,
"learning_rate": 4.095597328339452e-05,
"loss": 0.475,
"step": 116
},
{
"epoch": 0.17574164476154713,
"grad_norm": 0.27862656116485596,
"learning_rate": 4.0144148627425993e-05,
"loss": 0.4507,
"step": 117
},
{
"epoch": 0.17724371010138942,
"grad_norm": 0.2613980174064636,
"learning_rate": 3.933501846281267e-05,
"loss": 0.4843,
"step": 118
},
{
"epoch": 0.17874577544123169,
"grad_norm": 0.2648200988769531,
"learning_rate": 3.852880399766243e-05,
"loss": 0.4803,
"step": 119
},
{
"epoch": 0.18024784078107398,
"grad_norm": 0.21359142661094666,
"learning_rate": 3.772572564296005e-05,
"loss": 0.4547,
"step": 120
},
{
"epoch": 0.18174990612091627,
"grad_norm": 0.3049919605255127,
"learning_rate": 3.6926002952309016e-05,
"loss": 0.4756,
"step": 121
},
{
"epoch": 0.18325197146075856,
"grad_norm": 0.2409057766199112,
"learning_rate": 3.612985456190778e-05,
"loss": 0.4482,
"step": 122
},
{
"epoch": 0.18475403680060082,
"grad_norm": 0.2844046652317047,
"learning_rate": 3.533749813077677e-05,
"loss": 0.4584,
"step": 123
},
{
"epoch": 0.1862561021404431,
"grad_norm": 0.25576141476631165,
"learning_rate": 3.4549150281252636e-05,
"loss": 0.4534,
"step": 124
},
{
"epoch": 0.1877581674802854,
"grad_norm": 0.2949036657810211,
"learning_rate": 3.3765026539765834e-05,
"loss": 0.4317,
"step": 125
},
{
"epoch": 0.1892602328201277,
"grad_norm": 0.26583975553512573,
"learning_rate": 3.298534127791785e-05,
"loss": 0.4551,
"step": 126
},
{
"epoch": 0.19076229815996995,
"grad_norm": 0.23787036538124084,
"learning_rate": 3.221030765387417e-05,
"loss": 0.4281,
"step": 127
},
{
"epoch": 0.19226436349981224,
"grad_norm": 0.2489592432975769,
"learning_rate": 3.144013755408895e-05,
"loss": 0.4411,
"step": 128
},
{
"epoch": 0.19376642883965453,
"grad_norm": 0.24158670008182526,
"learning_rate": 3.0675041535377405e-05,
"loss": 0.426,
"step": 129
},
{
"epoch": 0.19526849417949682,
"grad_norm": 0.26309508085250854,
"learning_rate": 2.991522876735154e-05,
"loss": 0.4176,
"step": 130
},
{
"epoch": 0.19677055951933908,
"grad_norm": 0.30277812480926514,
"learning_rate": 2.916090697523549e-05,
"loss": 0.4447,
"step": 131
},
{
"epoch": 0.19827262485918137,
"grad_norm": 0.2779122292995453,
"learning_rate": 2.8412282383075363e-05,
"loss": 0.4184,
"step": 132
},
{
"epoch": 0.19977469019902366,
"grad_norm": 0.27596765756607056,
"learning_rate": 2.766955965735968e-05,
"loss": 0.4194,
"step": 133
},
{
"epoch": 0.20127675553886595,
"grad_norm": 0.29626959562301636,
"learning_rate": 2.693294185106562e-05,
"loss": 0.4133,
"step": 134
},
{
"epoch": 0.20277882087870822,
"grad_norm": 0.28480854630470276,
"learning_rate": 2.6202630348146324e-05,
"loss": 0.4135,
"step": 135
},
{
"epoch": 0.2042808862185505,
"grad_norm": 0.25113052129745483,
"learning_rate": 2.547882480847461e-05,
"loss": 0.4278,
"step": 136
},
{
"epoch": 0.2057829515583928,
"grad_norm": 0.2552507519721985,
"learning_rate": 2.476172311325783e-05,
"loss": 0.4119,
"step": 137
},
{
"epoch": 0.20728501689823509,
"grad_norm": 0.30283552408218384,
"learning_rate": 2.405152131093926e-05,
"loss": 0.4133,
"step": 138
},
{
"epoch": 0.20878708223807735,
"grad_norm": 0.2749648988246918,
"learning_rate": 2.3348413563600325e-05,
"loss": 0.3889,
"step": 139
},
{
"epoch": 0.21028914757791964,
"grad_norm": 0.2759481370449066,
"learning_rate": 2.2652592093878666e-05,
"loss": 0.4072,
"step": 140
},
{
"epoch": 0.21179121291776193,
"grad_norm": 0.30377233028411865,
"learning_rate": 2.196424713241637e-05,
"loss": 0.414,
"step": 141
},
{
"epoch": 0.21329327825760422,
"grad_norm": 0.28212490677833557,
"learning_rate": 2.128356686585282e-05,
"loss": 0.4027,
"step": 142
},
{
"epoch": 0.21479534359744648,
"grad_norm": 0.2726624608039856,
"learning_rate": 2.061073738537635e-05,
"loss": 0.4062,
"step": 143
},
{
"epoch": 0.21629740893728877,
"grad_norm": 0.3503369987010956,
"learning_rate": 1.9945942635848748e-05,
"loss": 0.4023,
"step": 144
},
{
"epoch": 0.21779947427713106,
"grad_norm": 0.3181203603744507,
"learning_rate": 1.928936436551661e-05,
"loss": 0.4119,
"step": 145
},
{
"epoch": 0.21930153961697335,
"grad_norm": 0.28087958693504333,
"learning_rate": 1.8641182076323148e-05,
"loss": 0.3891,
"step": 146
},
{
"epoch": 0.2208036049568156,
"grad_norm": 0.3313179910182953,
"learning_rate": 1.800157297483417e-05,
"loss": 0.4068,
"step": 147
},
{
"epoch": 0.2223056702966579,
"grad_norm": 0.3119460940361023,
"learning_rate": 1.7370711923791567e-05,
"loss": 0.4139,
"step": 148
},
{
"epoch": 0.2238077356365002,
"grad_norm": 0.3136446177959442,
"learning_rate": 1.6748771394307585e-05,
"loss": 0.3918,
"step": 149
},
{
"epoch": 0.22530980097634248,
"grad_norm": 0.4063158333301544,
"learning_rate": 1.6135921418712956e-05,
"loss": 0.3947,
"step": 150
},
{
"epoch": 0.22530980097634248,
"eval_loss": 0.44630166888237,
"eval_runtime": 33.3195,
"eval_samples_per_second": 33.674,
"eval_steps_per_second": 8.433,
"step": 150
},
{
"epoch": 0.22681186631618475,
"grad_norm": 0.6761639714241028,
"learning_rate": 1.553232954407171e-05,
"loss": 0.556,
"step": 151
},
{
"epoch": 0.22831393165602704,
"grad_norm": 0.5771117806434631,
"learning_rate": 1.4938160786375572e-05,
"loss": 0.5283,
"step": 152
},
{
"epoch": 0.22981599699586933,
"grad_norm": 0.4977632462978363,
"learning_rate": 1.435357758543015e-05,
"loss": 0.5066,
"step": 153
},
{
"epoch": 0.23131806233571162,
"grad_norm": 0.4167040288448334,
"learning_rate": 1.3778739760445552e-05,
"loss": 0.4982,
"step": 154
},
{
"epoch": 0.23282012767555388,
"grad_norm": 0.37243950366973877,
"learning_rate": 1.3213804466343421e-05,
"loss": 0.4843,
"step": 155
},
{
"epoch": 0.23432219301539617,
"grad_norm": 0.3721346855163574,
"learning_rate": 1.2658926150792322e-05,
"loss": 0.4715,
"step": 156
},
{
"epoch": 0.23582425835523846,
"grad_norm": 0.3776877522468567,
"learning_rate": 1.2114256511983274e-05,
"loss": 0.4811,
"step": 157
},
{
"epoch": 0.23732632369508075,
"grad_norm": 0.28867435455322266,
"learning_rate": 1.157994445715706e-05,
"loss": 0.4688,
"step": 158
},
{
"epoch": 0.238828389034923,
"grad_norm": 0.24486391246318817,
"learning_rate": 1.1056136061894384e-05,
"loss": 0.4596,
"step": 159
},
{
"epoch": 0.2403304543747653,
"grad_norm": 0.2534550130367279,
"learning_rate": 1.0542974530180327e-05,
"loss": 0.4498,
"step": 160
},
{
"epoch": 0.2418325197146076,
"grad_norm": 0.2360064685344696,
"learning_rate": 1.0040600155253765e-05,
"loss": 0.4588,
"step": 161
},
{
"epoch": 0.24333458505444988,
"grad_norm": 0.2181614190340042,
"learning_rate": 9.549150281252633e-06,
"loss": 0.448,
"step": 162
},
{
"epoch": 0.24483665039429214,
"grad_norm": 0.23044590651988983,
"learning_rate": 9.068759265665384e-06,
"loss": 0.4488,
"step": 163
},
{
"epoch": 0.24633871573413443,
"grad_norm": 0.23173829913139343,
"learning_rate": 8.599558442598998e-06,
"loss": 0.4436,
"step": 164
},
{
"epoch": 0.24784078107397672,
"grad_norm": 0.2390686273574829,
"learning_rate": 8.141676086873572e-06,
"loss": 0.4539,
"step": 165
},
{
"epoch": 0.249342846413819,
"grad_norm": 0.24815574288368225,
"learning_rate": 7.695237378953223e-06,
"loss": 0.4399,
"step": 166
},
{
"epoch": 0.2508449117536613,
"grad_norm": 0.22618462145328522,
"learning_rate": 7.260364370723044e-06,
"loss": 0.4393,
"step": 167
},
{
"epoch": 0.25234697709350357,
"grad_norm": 0.27694129943847656,
"learning_rate": 6.837175952121306e-06,
"loss": 0.4605,
"step": 168
},
{
"epoch": 0.25384904243334583,
"grad_norm": 0.2610919773578644,
"learning_rate": 6.425787818636131e-06,
"loss": 0.4299,
"step": 169
},
{
"epoch": 0.25535110777318815,
"grad_norm": 0.2240491807460785,
"learning_rate": 6.026312439675552e-06,
"loss": 0.4194,
"step": 170
},
{
"epoch": 0.2568531731130304,
"grad_norm": 0.22954358160495758,
"learning_rate": 5.6388590278194096e-06,
"loss": 0.4403,
"step": 171
},
{
"epoch": 0.2583552384528727,
"grad_norm": 0.2217157781124115,
"learning_rate": 5.263533508961827e-06,
"loss": 0.4206,
"step": 172
},
{
"epoch": 0.259857303792715,
"grad_norm": 0.232159823179245,
"learning_rate": 4.900438493352055e-06,
"loss": 0.4164,
"step": 173
},
{
"epoch": 0.26135936913255725,
"grad_norm": 0.19874323904514313,
"learning_rate": 4.549673247541875e-06,
"loss": 0.4237,
"step": 174
},
{
"epoch": 0.26286143447239957,
"grad_norm": 0.22404231131076813,
"learning_rate": 4.2113336672471245e-06,
"loss": 0.4112,
"step": 175
},
{
"epoch": 0.26436349981224183,
"grad_norm": 0.23319314420223236,
"learning_rate": 3.885512251130763e-06,
"loss": 0.4042,
"step": 176
},
{
"epoch": 0.2658655651520841,
"grad_norm": 0.21007098257541656,
"learning_rate": 3.5722980755146517e-06,
"loss": 0.4308,
"step": 177
},
{
"epoch": 0.2673676304919264,
"grad_norm": 0.22640475630760193,
"learning_rate": 3.271776770026963e-06,
"loss": 0.4037,
"step": 178
},
{
"epoch": 0.2688696958317687,
"grad_norm": 0.22629423439502716,
"learning_rate": 2.9840304941919415e-06,
"loss": 0.4074,
"step": 179
},
{
"epoch": 0.270371761171611,
"grad_norm": 0.2365429401397705,
"learning_rate": 2.7091379149682685e-06,
"loss": 0.4236,
"step": 180
},
{
"epoch": 0.27187382651145325,
"grad_norm": 0.2148093730211258,
"learning_rate": 2.4471741852423237e-06,
"loss": 0.4224,
"step": 181
},
{
"epoch": 0.2733758918512955,
"grad_norm": 0.24300482869148254,
"learning_rate": 2.1982109232821178e-06,
"loss": 0.412,
"step": 182
},
{
"epoch": 0.27487795719113783,
"grad_norm": 0.24461029469966888,
"learning_rate": 1.962316193157593e-06,
"loss": 0.4206,
"step": 183
},
{
"epoch": 0.2763800225309801,
"grad_norm": 0.23772835731506348,
"learning_rate": 1.7395544861325718e-06,
"loss": 0.3985,
"step": 184
},
{
"epoch": 0.27788208787082236,
"grad_norm": 0.22865405678749084,
"learning_rate": 1.5299867030334814e-06,
"loss": 0.3981,
"step": 185
},
{
"epoch": 0.2793841532106647,
"grad_norm": 0.23448914289474487,
"learning_rate": 1.333670137599713e-06,
"loss": 0.4014,
"step": 186
},
{
"epoch": 0.28088621855050694,
"grad_norm": 0.2333112210035324,
"learning_rate": 1.1506584608200367e-06,
"loss": 0.3981,
"step": 187
},
{
"epoch": 0.28238828389034926,
"grad_norm": 0.23065517842769623,
"learning_rate": 9.810017062595322e-07,
"loss": 0.4051,
"step": 188
},
{
"epoch": 0.2838903492301915,
"grad_norm": 0.2554570138454437,
"learning_rate": 8.247462563808817e-07,
"loss": 0.4156,
"step": 189
},
{
"epoch": 0.2853924145700338,
"grad_norm": 0.24712638556957245,
"learning_rate": 6.819348298638839e-07,
"loss": 0.4041,
"step": 190
},
{
"epoch": 0.2868944799098761,
"grad_norm": 0.25269147753715515,
"learning_rate": 5.526064699265753e-07,
"loss": 0.4001,
"step": 191
},
{
"epoch": 0.28839654524971836,
"grad_norm": 0.27400457859039307,
"learning_rate": 4.367965336512403e-07,
"loss": 0.4021,
"step": 192
},
{
"epoch": 0.2898986105895606,
"grad_norm": 0.28076010942459106,
"learning_rate": 3.3453668231809286e-07,
"loss": 0.3883,
"step": 193
},
{
"epoch": 0.29140067592940294,
"grad_norm": 0.2610340714454651,
"learning_rate": 2.458548727494292e-07,
"loss": 0.407,
"step": 194
},
{
"epoch": 0.2929027412692452,
"grad_norm": 0.28357601165771484,
"learning_rate": 1.7077534966650766e-07,
"loss": 0.3873,
"step": 195
},
{
"epoch": 0.2944048066090875,
"grad_norm": 0.3212686777114868,
"learning_rate": 1.0931863906127327e-07,
"loss": 0.4189,
"step": 196
},
{
"epoch": 0.2959068719489298,
"grad_norm": 0.34003758430480957,
"learning_rate": 6.150154258476315e-08,
"loss": 0.426,
"step": 197
},
{
"epoch": 0.29740893728877205,
"grad_norm": 0.3577791154384613,
"learning_rate": 2.7337132953697554e-08,
"loss": 0.4195,
"step": 198
},
{
"epoch": 0.29891100262861436,
"grad_norm": 0.3521157503128052,
"learning_rate": 6.834750376549792e-09,
"loss": 0.4127,
"step": 199
},
{
"epoch": 0.3004130679684566,
"grad_norm": 0.45909836888313293,
"learning_rate": 0.0,
"loss": 0.4286,
"step": 200
},
{
"epoch": 0.3004130679684566,
"eval_loss": 0.427767276763916,
"eval_runtime": 33.3252,
"eval_samples_per_second": 33.668,
"eval_steps_per_second": 8.432,
"step": 200
}
],
"logging_steps": 1,
"max_steps": 200,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 50,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 5,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 6.634625410321613e+16,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}