rhdeng's picture
Initial model upload
29e12ec verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.089058524173028,
"eval_steps": 500,
"global_step": 500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.05089058524173028,
"grad_norm": 2.2103662490844727,
"learning_rate": 0.0001999771640780308,
"loss": 3.0938,
"num_input_tokens_seen": 190176,
"step": 5
},
{
"epoch": 0.10178117048346055,
"grad_norm": 0.8035047054290771,
"learning_rate": 0.0001998844110196681,
"loss": 1.471,
"num_input_tokens_seen": 378656,
"step": 10
},
{
"epoch": 0.15267175572519084,
"grad_norm": 0.44624239206314087,
"learning_rate": 0.00019972037971811802,
"loss": 1.2452,
"num_input_tokens_seen": 570208,
"step": 15
},
{
"epoch": 0.2035623409669211,
"grad_norm": 0.41621240973472595,
"learning_rate": 0.00019948518722731206,
"loss": 1.1957,
"num_input_tokens_seen": 763040,
"step": 20
},
{
"epoch": 0.2544529262086514,
"grad_norm": 0.3875686824321747,
"learning_rate": 0.0001991790013823246,
"loss": 1.1319,
"num_input_tokens_seen": 959200,
"step": 25
},
{
"epoch": 0.3053435114503817,
"grad_norm": 0.4382984936237335,
"learning_rate": 0.00019880204067960472,
"loss": 1.1209,
"num_input_tokens_seen": 1153728,
"step": 30
},
{
"epoch": 0.356234096692112,
"grad_norm": 0.45683977007865906,
"learning_rate": 0.00019835457412105528,
"loss": 1.123,
"num_input_tokens_seen": 1351616,
"step": 35
},
{
"epoch": 0.4071246819338422,
"grad_norm": 0.4628298580646515,
"learning_rate": 0.00019783692102207155,
"loss": 1.0532,
"num_input_tokens_seen": 1537696,
"step": 40
},
{
"epoch": 0.4580152671755725,
"grad_norm": 0.5620412826538086,
"learning_rate": 0.00019724945078367513,
"loss": 1.0805,
"num_input_tokens_seen": 1730144,
"step": 45
},
{
"epoch": 0.5089058524173028,
"grad_norm": 0.4734291732311249,
"learning_rate": 0.00019659258262890683,
"loss": 1.0572,
"num_input_tokens_seen": 1916608,
"step": 50
},
{
"epoch": 0.5597964376590331,
"grad_norm": 0.46405333280563354,
"learning_rate": 0.00019586678530366606,
"loss": 1.0069,
"num_input_tokens_seen": 2103712,
"step": 55
},
{
"epoch": 0.6106870229007634,
"grad_norm": 0.4856356084346771,
"learning_rate": 0.00019507257674221027,
"loss": 1.0293,
"num_input_tokens_seen": 2289888,
"step": 60
},
{
"epoch": 0.6615776081424937,
"grad_norm": 0.4761466085910797,
"learning_rate": 0.00019421052369755334,
"loss": 0.9994,
"num_input_tokens_seen": 2477888,
"step": 65
},
{
"epoch": 0.712468193384224,
"grad_norm": 0.5239170789718628,
"learning_rate": 0.0001932812413370265,
"loss": 1.0159,
"num_input_tokens_seen": 2662848,
"step": 70
},
{
"epoch": 0.7633587786259542,
"grad_norm": 0.4161457419395447,
"learning_rate": 0.00019228539280329038,
"loss": 1.0233,
"num_input_tokens_seen": 2861024,
"step": 75
},
{
"epoch": 0.8142493638676844,
"grad_norm": 0.5579718351364136,
"learning_rate": 0.00019122368874111172,
"loss": 1.0109,
"num_input_tokens_seen": 3043712,
"step": 80
},
{
"epoch": 0.8651399491094147,
"grad_norm": 0.4595694839954376,
"learning_rate": 0.0001900968867902419,
"loss": 1.0132,
"num_input_tokens_seen": 3231840,
"step": 85
},
{
"epoch": 0.916030534351145,
"grad_norm": 0.49271535873413086,
"learning_rate": 0.00018890579104475995,
"loss": 1.0079,
"num_input_tokens_seen": 3429632,
"step": 90
},
{
"epoch": 0.9669211195928753,
"grad_norm": 0.5259862542152405,
"learning_rate": 0.00018765125147926476,
"loss": 0.9515,
"num_input_tokens_seen": 3620000,
"step": 95
},
{
"epoch": 1.0178117048346056,
"grad_norm": 0.444477379322052,
"learning_rate": 0.00018633416334232753,
"loss": 0.93,
"num_input_tokens_seen": 3809824,
"step": 100
},
{
"epoch": 1.0687022900763359,
"grad_norm": 0.42680156230926514,
"learning_rate": 0.0001849554665176354,
"loss": 0.8925,
"num_input_tokens_seen": 4005472,
"step": 105
},
{
"epoch": 1.1195928753180662,
"grad_norm": 0.6088462471961975,
"learning_rate": 0.00018351614485328388,
"loss": 0.8715,
"num_input_tokens_seen": 4192480,
"step": 110
},
{
"epoch": 1.1704834605597965,
"grad_norm": 0.5572230815887451,
"learning_rate": 0.0001820172254596956,
"loss": 0.8569,
"num_input_tokens_seen": 4379296,
"step": 115
},
{
"epoch": 1.2213740458015268,
"grad_norm": 0.5509177446365356,
"learning_rate": 0.00018045977797666684,
"loss": 0.8269,
"num_input_tokens_seen": 4568352,
"step": 120
},
{
"epoch": 1.272264631043257,
"grad_norm": 0.67293781042099,
"learning_rate": 0.00017884491381006478,
"loss": 0.84,
"num_input_tokens_seen": 4752928,
"step": 125
},
{
"epoch": 1.3231552162849873,
"grad_norm": 0.5845524668693542,
"learning_rate": 0.00017717378533872017,
"loss": 0.8243,
"num_input_tokens_seen": 4939008,
"step": 130
},
{
"epoch": 1.3740458015267176,
"grad_norm": 0.5623295307159424,
"learning_rate": 0.00017544758509208146,
"loss": 0.8624,
"num_input_tokens_seen": 5133664,
"step": 135
},
{
"epoch": 1.424936386768448,
"grad_norm": 0.5870885848999023,
"learning_rate": 0.00017366754489921694,
"loss": 0.8543,
"num_input_tokens_seen": 5332000,
"step": 140
},
{
"epoch": 1.4758269720101782,
"grad_norm": 0.5874502062797546,
"learning_rate": 0.00017183493500977278,
"loss": 0.842,
"num_input_tokens_seen": 5520640,
"step": 145
},
{
"epoch": 1.5267175572519083,
"grad_norm": 0.5786451101303101,
"learning_rate": 0.0001699510631875134,
"loss": 0.8268,
"num_input_tokens_seen": 5710528,
"step": 150
},
{
"epoch": 1.5776081424936388,
"grad_norm": 0.5929097533226013,
"learning_rate": 0.00016801727377709194,
"loss": 0.8238,
"num_input_tokens_seen": 5907264,
"step": 155
},
{
"epoch": 1.6284987277353689,
"grad_norm": 0.570965051651001,
"learning_rate": 0.00016603494674471593,
"loss": 0.8324,
"num_input_tokens_seen": 6095776,
"step": 160
},
{
"epoch": 1.6793893129770994,
"grad_norm": 0.6001121401786804,
"learning_rate": 0.0001640054966933935,
"loss": 0.8431,
"num_input_tokens_seen": 6281280,
"step": 165
},
{
"epoch": 1.7302798982188294,
"grad_norm": 0.5865895748138428,
"learning_rate": 0.00016193037185346224,
"loss": 0.8054,
"num_input_tokens_seen": 6470368,
"step": 170
},
{
"epoch": 1.78117048346056,
"grad_norm": 0.6706877946853638,
"learning_rate": 0.00015981105304912162,
"loss": 0.8255,
"num_input_tokens_seen": 6658944,
"step": 175
},
{
"epoch": 1.83206106870229,
"grad_norm": 0.6549966931343079,
"learning_rate": 0.0001576490526417059,
"loss": 0.8334,
"num_input_tokens_seen": 6842880,
"step": 180
},
{
"epoch": 1.8829516539440203,
"grad_norm": 0.6813002228736877,
"learning_rate": 0.0001554459134504523,
"loss": 0.8354,
"num_input_tokens_seen": 7034432,
"step": 185
},
{
"epoch": 1.9338422391857506,
"grad_norm": 0.6717746257781982,
"learning_rate": 0.00015320320765153367,
"loss": 0.8294,
"num_input_tokens_seen": 7228672,
"step": 190
},
{
"epoch": 1.984732824427481,
"grad_norm": 0.6509274840354919,
"learning_rate": 0.00015092253565614233,
"loss": 0.8561,
"num_input_tokens_seen": 7414144,
"step": 195
},
{
"epoch": 2.035623409669211,
"grad_norm": 0.6088241338729858,
"learning_rate": 0.00014860552496842494,
"loss": 0.6968,
"num_input_tokens_seen": 7603104,
"step": 200
},
{
"epoch": 2.0865139949109412,
"grad_norm": 0.7045843601226807,
"learning_rate": 0.00014625382902408356,
"loss": 0.6587,
"num_input_tokens_seen": 7794016,
"step": 205
},
{
"epoch": 2.1374045801526718,
"grad_norm": 0.6975011825561523,
"learning_rate": 0.00014386912601047213,
"loss": 0.6276,
"num_input_tokens_seen": 7987168,
"step": 210
},
{
"epoch": 2.188295165394402,
"grad_norm": 0.7472736835479736,
"learning_rate": 0.00014145311766902957,
"loss": 0.6451,
"num_input_tokens_seen": 8180224,
"step": 215
},
{
"epoch": 2.2391857506361323,
"grad_norm": 0.7247308492660522,
"learning_rate": 0.00013900752808090468,
"loss": 0.6776,
"num_input_tokens_seen": 8371136,
"step": 220
},
{
"epoch": 2.2900763358778624,
"grad_norm": 0.8300191164016724,
"learning_rate": 0.00013653410243663952,
"loss": 0.6307,
"num_input_tokens_seen": 8552064,
"step": 225
},
{
"epoch": 2.340966921119593,
"grad_norm": 0.791684627532959,
"learning_rate": 0.00013403460579078833,
"loss": 0.6623,
"num_input_tokens_seen": 8737120,
"step": 230
},
{
"epoch": 2.391857506361323,
"grad_norm": 0.7664393782615662,
"learning_rate": 0.0001315108218023621,
"loss": 0.6754,
"num_input_tokens_seen": 8928608,
"step": 235
},
{
"epoch": 2.4427480916030535,
"grad_norm": 0.7355414032936096,
"learning_rate": 0.0001289645514619963,
"loss": 0.6518,
"num_input_tokens_seen": 9120032,
"step": 240
},
{
"epoch": 2.4936386768447836,
"grad_norm": 0.7242420315742493,
"learning_rate": 0.00012639761180675098,
"loss": 0.6331,
"num_input_tokens_seen": 9307840,
"step": 245
},
{
"epoch": 2.544529262086514,
"grad_norm": 0.7869409322738647,
"learning_rate": 0.00012381183462345982,
"loss": 0.6588,
"num_input_tokens_seen": 9499104,
"step": 250
},
{
"epoch": 2.595419847328244,
"grad_norm": 0.7066891193389893,
"learning_rate": 0.0001212090651415537,
"loss": 0.6657,
"num_input_tokens_seen": 9694752,
"step": 255
},
{
"epoch": 2.6463104325699747,
"grad_norm": 0.8193861842155457,
"learning_rate": 0.00011859116071629149,
"loss": 0.6707,
"num_input_tokens_seen": 9889824,
"step": 260
},
{
"epoch": 2.6972010178117047,
"grad_norm": 0.7292617559432983,
"learning_rate": 0.00011595998950333793,
"loss": 0.6624,
"num_input_tokens_seen": 10080896,
"step": 265
},
{
"epoch": 2.7480916030534353,
"grad_norm": 0.7750299572944641,
"learning_rate": 0.00011331742912563413,
"loss": 0.6671,
"num_input_tokens_seen": 10278144,
"step": 270
},
{
"epoch": 2.7989821882951653,
"grad_norm": 0.7738965153694153,
"learning_rate": 0.00011066536533351202,
"loss": 0.6654,
"num_input_tokens_seen": 10466528,
"step": 275
},
{
"epoch": 2.849872773536896,
"grad_norm": 0.8403007984161377,
"learning_rate": 0.00010800569065900933,
"loss": 0.6681,
"num_input_tokens_seen": 10649024,
"step": 280
},
{
"epoch": 2.900763358778626,
"grad_norm": 0.8651427626609802,
"learning_rate": 0.0001053403030653449,
"loss": 0.6718,
"num_input_tokens_seen": 10834816,
"step": 285
},
{
"epoch": 2.9516539440203564,
"grad_norm": 0.8150460720062256,
"learning_rate": 0.00010267110459251823,
"loss": 0.637,
"num_input_tokens_seen": 11027616,
"step": 290
},
{
"epoch": 3.0025445292620865,
"grad_norm": 0.7902624011039734,
"learning_rate": 0.0001,
"loss": 0.6567,
"num_input_tokens_seen": 11217664,
"step": 295
},
{
"epoch": 3.053435114503817,
"grad_norm": 0.7837531566619873,
"learning_rate": 9.73288954074818e-05,
"loss": 0.5179,
"num_input_tokens_seen": 11406624,
"step": 300
},
{
"epoch": 3.104325699745547,
"grad_norm": 0.893497109413147,
"learning_rate": 9.46596969346551e-05,
"loss": 0.5329,
"num_input_tokens_seen": 11596096,
"step": 305
},
{
"epoch": 3.1552162849872776,
"grad_norm": 0.8641018867492676,
"learning_rate": 9.199430934099068e-05,
"loss": 0.5039,
"num_input_tokens_seen": 11788992,
"step": 310
},
{
"epoch": 3.2061068702290076,
"grad_norm": 0.9465248584747314,
"learning_rate": 8.933463466648798e-05,
"loss": 0.5085,
"num_input_tokens_seen": 11986976,
"step": 315
},
{
"epoch": 3.2569974554707377,
"grad_norm": 0.8461422920227051,
"learning_rate": 8.66825708743659e-05,
"loss": 0.5065,
"num_input_tokens_seen": 12179136,
"step": 320
},
{
"epoch": 3.3078880407124682,
"grad_norm": 0.9939844608306885,
"learning_rate": 8.404001049666211e-05,
"loss": 0.5059,
"num_input_tokens_seen": 12368608,
"step": 325
},
{
"epoch": 3.3587786259541983,
"grad_norm": 0.8322144150733948,
"learning_rate": 8.140883928370855e-05,
"loss": 0.5089,
"num_input_tokens_seen": 12556256,
"step": 330
},
{
"epoch": 3.409669211195929,
"grad_norm": 0.8849288821220398,
"learning_rate": 7.879093485844635e-05,
"loss": 0.4853,
"num_input_tokens_seen": 12746784,
"step": 335
},
{
"epoch": 3.460559796437659,
"grad_norm": 0.898762047290802,
"learning_rate": 7.618816537654018e-05,
"loss": 0.5119,
"num_input_tokens_seen": 12944096,
"step": 340
},
{
"epoch": 3.5114503816793894,
"grad_norm": 0.9055672883987427,
"learning_rate": 7.360238819324903e-05,
"loss": 0.5024,
"num_input_tokens_seen": 13139904,
"step": 345
},
{
"epoch": 3.5623409669211195,
"grad_norm": 0.8258459568023682,
"learning_rate": 7.10354485380037e-05,
"loss": 0.5074,
"num_input_tokens_seen": 13328480,
"step": 350
},
{
"epoch": 3.61323155216285,
"grad_norm": 0.8779503703117371,
"learning_rate": 6.848917819763793e-05,
"loss": 0.5078,
"num_input_tokens_seen": 13518112,
"step": 355
},
{
"epoch": 3.66412213740458,
"grad_norm": 0.8894705176353455,
"learning_rate": 6.596539420921171e-05,
"loss": 0.5211,
"num_input_tokens_seen": 13705632,
"step": 360
},
{
"epoch": 3.7150127226463106,
"grad_norm": 0.8531593680381775,
"learning_rate": 6.34658975633605e-05,
"loss": 0.5042,
"num_input_tokens_seen": 13896896,
"step": 365
},
{
"epoch": 3.7659033078880406,
"grad_norm": 0.859188973903656,
"learning_rate": 6.0992471919095315e-05,
"loss": 0.5038,
"num_input_tokens_seen": 14081888,
"step": 370
},
{
"epoch": 3.816793893129771,
"grad_norm": 0.9424790740013123,
"learning_rate": 5.854688233097045e-05,
"loss": 0.5046,
"num_input_tokens_seen": 14270560,
"step": 375
},
{
"epoch": 3.867684478371501,
"grad_norm": 0.90933758020401,
"learning_rate": 5.613087398952792e-05,
"loss": 0.5196,
"num_input_tokens_seen": 14461120,
"step": 380
},
{
"epoch": 3.9185750636132317,
"grad_norm": 0.9108320474624634,
"learning_rate": 5.37461709759165e-05,
"loss": 0.4898,
"num_input_tokens_seen": 14648096,
"step": 385
},
{
"epoch": 3.969465648854962,
"grad_norm": 0.8353007435798645,
"learning_rate": 5.139447503157513e-05,
"loss": 0.5283,
"num_input_tokens_seen": 14839840,
"step": 390
},
{
"epoch": 4.020356234096692,
"grad_norm": 0.8553094863891602,
"learning_rate": 4.9077464343857694e-05,
"loss": 0.4485,
"num_input_tokens_seen": 15020448,
"step": 395
},
{
"epoch": 4.071246819338422,
"grad_norm": 0.9495893120765686,
"learning_rate": 4.6796792348466356e-05,
"loss": 0.4041,
"num_input_tokens_seen": 15210368,
"step": 400
},
{
"epoch": 4.122137404580153,
"grad_norm": 0.9808993339538574,
"learning_rate": 4.4554086549547715e-05,
"loss": 0.4028,
"num_input_tokens_seen": 15399296,
"step": 405
},
{
"epoch": 4.1730279898218825,
"grad_norm": 1.0066555738449097,
"learning_rate": 4.23509473582941e-05,
"loss": 0.4001,
"num_input_tokens_seen": 15580416,
"step": 410
},
{
"epoch": 4.223918575063613,
"grad_norm": 0.987274706363678,
"learning_rate": 4.0188946950878404e-05,
"loss": 0.3725,
"num_input_tokens_seen": 15768480,
"step": 415
},
{
"epoch": 4.2748091603053435,
"grad_norm": 0.8921579718589783,
"learning_rate": 3.806962814653779e-05,
"loss": 0.3941,
"num_input_tokens_seen": 15951072,
"step": 420
},
{
"epoch": 4.325699745547074,
"grad_norm": 1.0258628129959106,
"learning_rate": 3.5994503306606497e-05,
"loss": 0.4274,
"num_input_tokens_seen": 16148512,
"step": 425
},
{
"epoch": 4.376590330788804,
"grad_norm": 0.9532390236854553,
"learning_rate": 3.3965053255284084e-05,
"loss": 0.394,
"num_input_tokens_seen": 16341600,
"step": 430
},
{
"epoch": 4.427480916030534,
"grad_norm": 0.9402347803115845,
"learning_rate": 3.198272622290804e-05,
"loss": 0.4072,
"num_input_tokens_seen": 16533792,
"step": 435
},
{
"epoch": 4.478371501272265,
"grad_norm": 0.9012470841407776,
"learning_rate": 3.0048936812486615e-05,
"loss": 0.4025,
"num_input_tokens_seen": 16728128,
"step": 440
},
{
"epoch": 4.529262086513995,
"grad_norm": 1.046890377998352,
"learning_rate": 2.8165064990227252e-05,
"loss": 0.3966,
"num_input_tokens_seen": 16922560,
"step": 445
},
{
"epoch": 4.580152671755725,
"grad_norm": 0.9989413619041443,
"learning_rate": 2.6332455100783083e-05,
"loss": 0.394,
"num_input_tokens_seen": 17106272,
"step": 450
},
{
"epoch": 4.631043256997455,
"grad_norm": 0.9461767077445984,
"learning_rate": 2.4552414907918564e-05,
"loss": 0.4152,
"num_input_tokens_seen": 17298240,
"step": 455
},
{
"epoch": 4.681933842239186,
"grad_norm": 1.0528404712677002,
"learning_rate": 2.282621466127982e-05,
"loss": 0.408,
"num_input_tokens_seen": 17490624,
"step": 460
},
{
"epoch": 4.732824427480916,
"grad_norm": 1.0558110475540161,
"learning_rate": 2.1155086189935224e-05,
"loss": 0.3924,
"num_input_tokens_seen": 17677632,
"step": 465
},
{
"epoch": 4.783715012722646,
"grad_norm": 0.8918471932411194,
"learning_rate": 1.9540222023333166e-05,
"loss": 0.4161,
"num_input_tokens_seen": 17867872,
"step": 470
},
{
"epoch": 4.8346055979643765,
"grad_norm": 0.9665652513504028,
"learning_rate": 1.7982774540304403e-05,
"loss": 0.3904,
"num_input_tokens_seen": 18055552,
"step": 475
},
{
"epoch": 4.885496183206107,
"grad_norm": 0.9551387429237366,
"learning_rate": 1.6483855146716152e-05,
"loss": 0.4057,
"num_input_tokens_seen": 18247104,
"step": 480
},
{
"epoch": 4.9363867684478375,
"grad_norm": 0.9778920412063599,
"learning_rate": 1.504453348236461e-05,
"loss": 0.3934,
"num_input_tokens_seen": 18434752,
"step": 485
},
{
"epoch": 4.987277353689567,
"grad_norm": 0.9678117632865906,
"learning_rate": 1.3665836657672493e-05,
"loss": 0.4161,
"num_input_tokens_seen": 18633824,
"step": 490
},
{
"epoch": 5.038167938931298,
"grad_norm": 0.8953499794006348,
"learning_rate": 1.2348748520735221e-05,
"loss": 0.3599,
"num_input_tokens_seen": 18825024,
"step": 495
},
{
"epoch": 5.089058524173028,
"grad_norm": 0.9666320085525513,
"learning_rate": 1.1094208955240081e-05,
"loss": 0.356,
"num_input_tokens_seen": 19013664,
"step": 500
}
],
"logging_steps": 5,
"max_steps": 588,
"num_input_tokens_seen": 19013664,
"num_train_epochs": 6,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 8.585691706551173e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}