coding_llamaduo_result3 / trainer_state.json
chansung's picture
Model save
2b79151 verified
raw
history blame
28.7 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"eval_steps": 500,
"global_step": 820,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.006097560975609756,
"grad_norm": 470.0,
"learning_rate": 2.4390243902439027e-06,
"loss": 50.3127,
"step": 1
},
{
"epoch": 0.03048780487804878,
"grad_norm": 362.0,
"learning_rate": 1.2195121951219513e-05,
"loss": 41.9975,
"step": 5
},
{
"epoch": 0.06097560975609756,
"grad_norm": 199.0,
"learning_rate": 2.4390243902439026e-05,
"loss": 35.2694,
"step": 10
},
{
"epoch": 0.09146341463414634,
"grad_norm": 27.875,
"learning_rate": 3.6585365853658535e-05,
"loss": 22.3723,
"step": 15
},
{
"epoch": 0.12195121951219512,
"grad_norm": 21.375,
"learning_rate": 4.878048780487805e-05,
"loss": 18.6519,
"step": 20
},
{
"epoch": 0.1524390243902439,
"grad_norm": 5.8125,
"learning_rate": 6.097560975609756e-05,
"loss": 16.0212,
"step": 25
},
{
"epoch": 0.18292682926829268,
"grad_norm": 4.5625,
"learning_rate": 7.317073170731707e-05,
"loss": 15.6762,
"step": 30
},
{
"epoch": 0.21341463414634146,
"grad_norm": 5.9375,
"learning_rate": 8.53658536585366e-05,
"loss": 13.4942,
"step": 35
},
{
"epoch": 0.24390243902439024,
"grad_norm": 13.875,
"learning_rate": 9.75609756097561e-05,
"loss": 13.2206,
"step": 40
},
{
"epoch": 0.27439024390243905,
"grad_norm": 40.0,
"learning_rate": 0.00010975609756097563,
"loss": 9.5294,
"step": 45
},
{
"epoch": 0.3048780487804878,
"grad_norm": 8.0625,
"learning_rate": 0.00012195121951219512,
"loss": 2.8066,
"step": 50
},
{
"epoch": 0.3353658536585366,
"grad_norm": 2.421875,
"learning_rate": 0.00013414634146341464,
"loss": 1.5733,
"step": 55
},
{
"epoch": 0.36585365853658536,
"grad_norm": 2.0625,
"learning_rate": 0.00014634146341463414,
"loss": 1.3089,
"step": 60
},
{
"epoch": 0.39634146341463417,
"grad_norm": 3.15625,
"learning_rate": 0.00015853658536585366,
"loss": 1.1767,
"step": 65
},
{
"epoch": 0.4268292682926829,
"grad_norm": 7.0625,
"learning_rate": 0.0001707317073170732,
"loss": 1.1124,
"step": 70
},
{
"epoch": 0.4573170731707317,
"grad_norm": 3.265625,
"learning_rate": 0.0001829268292682927,
"loss": 1.0577,
"step": 75
},
{
"epoch": 0.4878048780487805,
"grad_norm": 6.59375,
"learning_rate": 0.0001951219512195122,
"loss": 0.9769,
"step": 80
},
{
"epoch": 0.5182926829268293,
"grad_norm": 10.6875,
"learning_rate": 0.00019999184556954776,
"loss": 0.9411,
"step": 85
},
{
"epoch": 0.5487804878048781,
"grad_norm": 1.171875,
"learning_rate": 0.0001999420177550043,
"loss": 0.939,
"step": 90
},
{
"epoch": 0.5792682926829268,
"grad_norm": 2.25,
"learning_rate": 0.00019984691491033906,
"loss": 0.9085,
"step": 95
},
{
"epoch": 0.6097560975609756,
"grad_norm": 0.97265625,
"learning_rate": 0.00019970658011837404,
"loss": 0.8808,
"step": 100
},
{
"epoch": 0.6402439024390244,
"grad_norm": 2.046875,
"learning_rate": 0.00019952107695258992,
"loss": 0.8201,
"step": 105
},
{
"epoch": 0.6707317073170732,
"grad_norm": 0.828125,
"learning_rate": 0.00019929048944832638,
"loss": 0.7852,
"step": 110
},
{
"epoch": 0.7012195121951219,
"grad_norm": 2.3125,
"learning_rate": 0.00019901492206471325,
"loss": 0.7695,
"step": 115
},
{
"epoch": 0.7317073170731707,
"grad_norm": 14.3125,
"learning_rate": 0.00019869449963734893,
"loss": 0.7996,
"step": 120
},
{
"epoch": 0.7621951219512195,
"grad_norm": 1.0859375,
"learning_rate": 0.00019832936732174834,
"loss": 0.7611,
"step": 125
},
{
"epoch": 0.7926829268292683,
"grad_norm": 29.0,
"learning_rate": 0.00019791969052758562,
"loss": 0.7835,
"step": 130
},
{
"epoch": 0.823170731707317,
"grad_norm": 0.6015625,
"learning_rate": 0.00019746565484376132,
"loss": 0.759,
"step": 135
},
{
"epoch": 0.8536585365853658,
"grad_norm": 0.57421875,
"learning_rate": 0.00019696746595432828,
"loss": 0.7598,
"step": 140
},
{
"epoch": 0.8841463414634146,
"grad_norm": 0.9140625,
"learning_rate": 0.0001964253495453141,
"loss": 0.7279,
"step": 145
},
{
"epoch": 0.9146341463414634,
"grad_norm": 1.484375,
"learning_rate": 0.00019583955120248237,
"loss": 0.7304,
"step": 150
},
{
"epoch": 0.9451219512195121,
"grad_norm": 0.7890625,
"learning_rate": 0.00019521033630007928,
"loss": 0.7659,
"step": 155
},
{
"epoch": 0.975609756097561,
"grad_norm": 3.03125,
"learning_rate": 0.00019453798988061535,
"loss": 0.7425,
"step": 160
},
{
"epoch": 1.0,
"eval_loss": 1.182375431060791,
"eval_runtime": 1.0783,
"eval_samples_per_second": 1.855,
"eval_steps_per_second": 0.927,
"step": 164
},
{
"epoch": 1.0060975609756098,
"grad_norm": 0.8515625,
"learning_rate": 0.00019382281652573785,
"loss": 0.6928,
"step": 165
},
{
"epoch": 1.0365853658536586,
"grad_norm": 0.9375,
"learning_rate": 0.00019306514021825118,
"loss": 0.6592,
"step": 170
},
{
"epoch": 1.0670731707317074,
"grad_norm": 0.8125,
"learning_rate": 0.00019226530419534833,
"loss": 0.6442,
"step": 175
},
{
"epoch": 1.0975609756097562,
"grad_norm": 0.72265625,
"learning_rate": 0.00019142367079312021,
"loss": 0.659,
"step": 180
},
{
"epoch": 1.1280487804878048,
"grad_norm": 0.462890625,
"learning_rate": 0.00019054062128241264,
"loss": 0.6355,
"step": 185
},
{
"epoch": 1.1585365853658536,
"grad_norm": 1.1015625,
"learning_rate": 0.00018961655569610557,
"loss": 0.6304,
"step": 190
},
{
"epoch": 1.1890243902439024,
"grad_norm": 0.5625,
"learning_rate": 0.0001886518926478932,
"loss": 0.6414,
"step": 195
},
{
"epoch": 1.2195121951219512,
"grad_norm": 0.5,
"learning_rate": 0.00018764706914264635,
"loss": 0.6429,
"step": 200
},
{
"epoch": 1.25,
"grad_norm": 1.046875,
"learning_rate": 0.00018660254037844388,
"loss": 0.6442,
"step": 205
},
{
"epoch": 1.2804878048780488,
"grad_norm": 0.73046875,
"learning_rate": 0.00018551877954036162,
"loss": 0.6394,
"step": 210
},
{
"epoch": 1.3109756097560976,
"grad_norm": 0.87109375,
"learning_rate": 0.00018439627758611385,
"loss": 0.6256,
"step": 215
},
{
"epoch": 1.3414634146341464,
"grad_norm": 0.890625,
"learning_rate": 0.00018323554302364272,
"loss": 0.6397,
"step": 220
},
{
"epoch": 1.3719512195121952,
"grad_norm": 0.96875,
"learning_rate": 0.00018203710168075788,
"loss": 0.6138,
"step": 225
},
{
"epoch": 1.4024390243902438,
"grad_norm": 1.0625,
"learning_rate": 0.0001808014964669293,
"loss": 0.6305,
"step": 230
},
{
"epoch": 1.4329268292682926,
"grad_norm": 0.51953125,
"learning_rate": 0.00017952928712734268,
"loss": 0.6213,
"step": 235
},
{
"epoch": 1.4634146341463414,
"grad_norm": 1.78125,
"learning_rate": 0.00017822104998932713,
"loss": 0.6233,
"step": 240
},
{
"epoch": 1.4939024390243902,
"grad_norm": 0.7265625,
"learning_rate": 0.00017687737770127185,
"loss": 0.6489,
"step": 245
},
{
"epoch": 1.524390243902439,
"grad_norm": 0.8984375,
"learning_rate": 0.00017549887896414851,
"loss": 0.6058,
"step": 250
},
{
"epoch": 1.5548780487804879,
"grad_norm": 1.015625,
"learning_rate": 0.0001740861782557618,
"loss": 0.6164,
"step": 255
},
{
"epoch": 1.5853658536585367,
"grad_norm": 0.73046875,
"learning_rate": 0.0001726399155478529,
"loss": 0.606,
"step": 260
},
{
"epoch": 1.6158536585365852,
"grad_norm": 0.5703125,
"learning_rate": 0.00017116074601618417,
"loss": 0.6051,
"step": 265
},
{
"epoch": 1.6463414634146343,
"grad_norm": 0.7109375,
"learning_rate": 0.0001696493397437357,
"loss": 0.6798,
"step": 270
},
{
"epoch": 1.6768292682926829,
"grad_norm": 1.6953125,
"learning_rate": 0.00016810638141714934,
"loss": 0.6205,
"step": 275
},
{
"epoch": 1.7073170731707317,
"grad_norm": 1.015625,
"learning_rate": 0.00016653257001655652,
"loss": 0.6627,
"step": 280
},
{
"epoch": 1.7378048780487805,
"grad_norm": 0.84765625,
"learning_rate": 0.0001649286184989315,
"loss": 0.6152,
"step": 285
},
{
"epoch": 1.7682926829268293,
"grad_norm": 1.21875,
"learning_rate": 0.0001632952534751122,
"loss": 0.5972,
"step": 290
},
{
"epoch": 1.798780487804878,
"grad_norm": 0.4609375,
"learning_rate": 0.00016163321488063637,
"loss": 0.6053,
"step": 295
},
{
"epoch": 1.8292682926829267,
"grad_norm": 0.423828125,
"learning_rate": 0.00015994325564054122,
"loss": 0.5965,
"step": 300
},
{
"epoch": 1.8597560975609757,
"grad_norm": 0.9296875,
"learning_rate": 0.00015822614132827837,
"loss": 0.6023,
"step": 305
},
{
"epoch": 1.8902439024390243,
"grad_norm": 0.671875,
"learning_rate": 0.00015648264981889934,
"loss": 0.5916,
"step": 310
},
{
"epoch": 1.9207317073170733,
"grad_norm": 0.68359375,
"learning_rate": 0.00015471357093666804,
"loss": 0.6082,
"step": 315
},
{
"epoch": 1.951219512195122,
"grad_norm": 0.578125,
"learning_rate": 0.00015291970609726007,
"loss": 0.6176,
"step": 320
},
{
"epoch": 1.9817073170731707,
"grad_norm": 0.55859375,
"learning_rate": 0.00015110186794471103,
"loss": 0.6196,
"step": 325
},
{
"epoch": 2.0,
"eval_loss": 1.1435465812683105,
"eval_runtime": 1.0793,
"eval_samples_per_second": 1.853,
"eval_steps_per_second": 0.927,
"step": 328
},
{
"epoch": 2.0121951219512195,
"grad_norm": 0.5703125,
"learning_rate": 0.00014926087998327837,
"loss": 0.5623,
"step": 330
},
{
"epoch": 2.042682926829268,
"grad_norm": 0.82421875,
"learning_rate": 0.00014739757620438307,
"loss": 0.5231,
"step": 335
},
{
"epoch": 2.073170731707317,
"grad_norm": 1.25,
"learning_rate": 0.0001455128007088009,
"loss": 0.5611,
"step": 340
},
{
"epoch": 2.1036585365853657,
"grad_norm": 0.5,
"learning_rate": 0.00014360740732427367,
"loss": 0.5497,
"step": 345
},
{
"epoch": 2.1341463414634148,
"grad_norm": 0.443359375,
"learning_rate": 0.00014168225921871433,
"loss": 0.5154,
"step": 350
},
{
"epoch": 2.1646341463414633,
"grad_norm": 0.498046875,
"learning_rate": 0.00013973822850918055,
"loss": 0.5307,
"step": 355
},
{
"epoch": 2.1951219512195124,
"grad_norm": 0.79296875,
"learning_rate": 0.0001377761958667946,
"loss": 0.5085,
"step": 360
},
{
"epoch": 2.225609756097561,
"grad_norm": 0.8046875,
"learning_rate": 0.00013579705011778766,
"loss": 0.5339,
"step": 365
},
{
"epoch": 2.2560975609756095,
"grad_norm": 0.7890625,
"learning_rate": 0.00013380168784085027,
"loss": 0.5263,
"step": 370
},
{
"epoch": 2.2865853658536586,
"grad_norm": 0.5234375,
"learning_rate": 0.00013179101296097035,
"loss": 0.5106,
"step": 375
},
{
"epoch": 2.317073170731707,
"grad_norm": 0.62890625,
"learning_rate": 0.00012976593633994346,
"loss": 0.5371,
"step": 380
},
{
"epoch": 2.347560975609756,
"grad_norm": 0.4140625,
"learning_rate": 0.0001277273753637408,
"loss": 0.5188,
"step": 385
},
{
"epoch": 2.3780487804878048,
"grad_norm": 0.66015625,
"learning_rate": 0.00012567625352692127,
"loss": 0.5309,
"step": 390
},
{
"epoch": 2.408536585365854,
"grad_norm": 1.21875,
"learning_rate": 0.0001236135000142765,
"loss": 0.5024,
"step": 395
},
{
"epoch": 2.4390243902439024,
"grad_norm": 0.58203125,
"learning_rate": 0.00012154004927989815,
"loss": 0.5291,
"step": 400
},
{
"epoch": 2.4695121951219514,
"grad_norm": 0.478515625,
"learning_rate": 0.00011945684062385803,
"loss": 0.5261,
"step": 405
},
{
"epoch": 2.5,
"grad_norm": 0.431640625,
"learning_rate": 0.00011736481776669306,
"loss": 0.5218,
"step": 410
},
{
"epoch": 2.5304878048780486,
"grad_norm": 0.447265625,
"learning_rate": 0.00011526492842188745,
"loss": 0.5259,
"step": 415
},
{
"epoch": 2.5609756097560976,
"grad_norm": 0.41796875,
"learning_rate": 0.0001131581238665465,
"loss": 0.5077,
"step": 420
},
{
"epoch": 2.591463414634146,
"grad_norm": 0.42578125,
"learning_rate": 0.00011104535851045539,
"loss": 0.5356,
"step": 425
},
{
"epoch": 2.6219512195121952,
"grad_norm": 0.62890625,
"learning_rate": 0.00010892758946371944,
"loss": 0.5158,
"step": 430
},
{
"epoch": 2.652439024390244,
"grad_norm": 0.46484375,
"learning_rate": 0.00010680577610318072,
"loss": 0.5297,
"step": 435
},
{
"epoch": 2.682926829268293,
"grad_norm": 0.4609375,
"learning_rate": 0.00010468087963780789,
"loss": 0.5015,
"step": 440
},
{
"epoch": 2.7134146341463414,
"grad_norm": 0.447265625,
"learning_rate": 0.00010255386267325602,
"loss": 0.5169,
"step": 445
},
{
"epoch": 2.7439024390243905,
"grad_norm": 0.6171875,
"learning_rate": 0.00010042568877579388,
"loss": 0.5034,
"step": 450
},
{
"epoch": 2.774390243902439,
"grad_norm": 0.419921875,
"learning_rate": 9.829732203579584e-05,
"loss": 0.5113,
"step": 455
},
{
"epoch": 2.8048780487804876,
"grad_norm": 0.396484375,
"learning_rate": 9.616972663099647e-05,
"loss": 0.5106,
"step": 460
},
{
"epoch": 2.8353658536585367,
"grad_norm": 0.4765625,
"learning_rate": 9.404386638970542e-05,
"loss": 0.5135,
"step": 465
},
{
"epoch": 2.8658536585365852,
"grad_norm": 0.390625,
"learning_rate": 9.192070435418079e-05,
"loss": 0.5222,
"step": 470
},
{
"epoch": 2.8963414634146343,
"grad_norm": 0.55859375,
"learning_rate": 8.980120234435849e-05,
"loss": 0.5214,
"step": 475
},
{
"epoch": 2.926829268292683,
"grad_norm": 0.6328125,
"learning_rate": 8.768632052213531e-05,
"loss": 0.5408,
"step": 480
},
{
"epoch": 2.9573170731707314,
"grad_norm": 0.490234375,
"learning_rate": 8.557701695640321e-05,
"loss": 0.5012,
"step": 485
},
{
"epoch": 2.9878048780487805,
"grad_norm": 0.41796875,
"learning_rate": 8.347424718903151e-05,
"loss": 0.5124,
"step": 490
},
{
"epoch": 3.0,
"eval_loss": 1.1460928916931152,
"eval_runtime": 1.0779,
"eval_samples_per_second": 1.855,
"eval_steps_per_second": 0.928,
"step": 492
},
{
"epoch": 3.018292682926829,
"grad_norm": 0.71484375,
"learning_rate": 8.13789638019942e-05,
"loss": 0.4767,
"step": 495
},
{
"epoch": 3.048780487804878,
"grad_norm": 0.53515625,
"learning_rate": 7.929211598583794e-05,
"loss": 0.4404,
"step": 500
},
{
"epoch": 3.0792682926829267,
"grad_norm": 0.546875,
"learning_rate": 7.721464910968627e-05,
"loss": 0.447,
"step": 505
},
{
"epoch": 3.1097560975609757,
"grad_norm": 0.40234375,
"learning_rate": 7.514750429297528e-05,
"loss": 0.4379,
"step": 510
},
{
"epoch": 3.1402439024390243,
"grad_norm": 0.59765625,
"learning_rate": 7.309161797911441e-05,
"loss": 0.4287,
"step": 515
},
{
"epoch": 3.1707317073170733,
"grad_norm": 0.4375,
"learning_rate": 7.104792151126515e-05,
"loss": 0.428,
"step": 520
},
{
"epoch": 3.201219512195122,
"grad_norm": 0.4296875,
"learning_rate": 6.901734071043071e-05,
"loss": 0.4448,
"step": 525
},
{
"epoch": 3.231707317073171,
"grad_norm": 0.4453125,
"learning_rate": 6.700079545604708e-05,
"loss": 0.4297,
"step": 530
},
{
"epoch": 3.2621951219512195,
"grad_norm": 0.435546875,
"learning_rate": 6.499919926926566e-05,
"loss": 0.423,
"step": 535
},
{
"epoch": 3.292682926829268,
"grad_norm": 0.44921875,
"learning_rate": 6.301345889911637e-05,
"loss": 0.4218,
"step": 540
},
{
"epoch": 3.323170731707317,
"grad_norm": 0.56640625,
"learning_rate": 6.104447391173858e-05,
"loss": 0.4354,
"step": 545
},
{
"epoch": 3.3536585365853657,
"grad_norm": 0.47265625,
"learning_rate": 5.909313628286601e-05,
"loss": 0.4233,
"step": 550
},
{
"epoch": 3.3841463414634148,
"grad_norm": 0.42578125,
"learning_rate": 5.716032999375006e-05,
"loss": 0.4634,
"step": 555
},
{
"epoch": 3.4146341463414633,
"grad_norm": 0.419921875,
"learning_rate": 5.524693063070492e-05,
"loss": 0.4268,
"step": 560
},
{
"epoch": 3.4451219512195124,
"grad_norm": 0.494140625,
"learning_rate": 5.335380498845559e-05,
"loss": 0.4333,
"step": 565
},
{
"epoch": 3.475609756097561,
"grad_norm": 0.46875,
"learning_rate": 5.148181067746862e-05,
"loss": 0.471,
"step": 570
},
{
"epoch": 3.5060975609756095,
"grad_norm": 0.44921875,
"learning_rate": 4.963179573544357e-05,
"loss": 0.4116,
"step": 575
},
{
"epoch": 3.5365853658536586,
"grad_norm": 0.458984375,
"learning_rate": 4.7804598243140666e-05,
"loss": 0.4301,
"step": 580
},
{
"epoch": 3.567073170731707,
"grad_norm": 0.431640625,
"learning_rate": 4.60010459447196e-05,
"loss": 0.4359,
"step": 585
},
{
"epoch": 3.597560975609756,
"grad_norm": 0.44921875,
"learning_rate": 4.422195587276058e-05,
"loss": 0.4322,
"step": 590
},
{
"epoch": 3.6280487804878048,
"grad_norm": 0.52734375,
"learning_rate": 4.2468133978137945e-05,
"loss": 0.4283,
"step": 595
},
{
"epoch": 3.658536585365854,
"grad_norm": 0.51171875,
"learning_rate": 4.0740374764914136e-05,
"loss": 0.4297,
"step": 600
},
{
"epoch": 3.6890243902439024,
"grad_norm": 0.52734375,
"learning_rate": 3.903946093041877e-05,
"loss": 0.4386,
"step": 605
},
{
"epoch": 3.7195121951219514,
"grad_norm": 0.4453125,
"learning_rate": 3.736616301067694e-05,
"loss": 0.4279,
"step": 610
},
{
"epoch": 3.75,
"grad_norm": 0.53125,
"learning_rate": 3.5721239031346066e-05,
"loss": 0.454,
"step": 615
},
{
"epoch": 3.7804878048780486,
"grad_norm": 0.447265625,
"learning_rate": 3.410543416432069e-05,
"loss": 0.4497,
"step": 620
},
{
"epoch": 3.8109756097560976,
"grad_norm": 0.48828125,
"learning_rate": 3.2519480390159806e-05,
"loss": 0.4405,
"step": 625
},
{
"epoch": 3.841463414634146,
"grad_norm": 0.55078125,
"learning_rate": 3.096409616649023e-05,
"loss": 0.4354,
"step": 630
},
{
"epoch": 3.8719512195121952,
"grad_norm": 0.443359375,
"learning_rate": 2.9439986102536043e-05,
"loss": 0.4382,
"step": 635
},
{
"epoch": 3.902439024390244,
"grad_norm": 0.416015625,
"learning_rate": 2.794784063992131e-05,
"loss": 0.4204,
"step": 640
},
{
"epoch": 3.932926829268293,
"grad_norm": 0.515625,
"learning_rate": 2.6488335739891178e-05,
"loss": 0.4199,
"step": 645
},
{
"epoch": 3.9634146341463414,
"grad_norm": 0.5234375,
"learning_rate": 2.50621325770927e-05,
"loss": 0.4275,
"step": 650
},
{
"epoch": 3.9939024390243905,
"grad_norm": 0.4765625,
"learning_rate": 2.366987724005404e-05,
"loss": 0.4403,
"step": 655
},
{
"epoch": 4.0,
"eval_loss": 1.206821322441101,
"eval_runtime": 1.0795,
"eval_samples_per_second": 1.853,
"eval_steps_per_second": 0.926,
"step": 656
},
{
"epoch": 4.024390243902439,
"grad_norm": 0.408203125,
"learning_rate": 2.2312200438498043e-05,
"loss": 0.3802,
"step": 660
},
{
"epoch": 4.054878048780488,
"grad_norm": 0.455078125,
"learning_rate": 2.0989717217622652e-05,
"loss": 0.3864,
"step": 665
},
{
"epoch": 4.085365853658536,
"grad_norm": 0.443359375,
"learning_rate": 1.9703026679477256e-05,
"loss": 0.3842,
"step": 670
},
{
"epoch": 4.115853658536586,
"grad_norm": 0.5546875,
"learning_rate": 1.8452711711561842e-05,
"loss": 0.3832,
"step": 675
},
{
"epoch": 4.146341463414634,
"grad_norm": 0.447265625,
"learning_rate": 1.7239338722771327e-05,
"loss": 0.3686,
"step": 680
},
{
"epoch": 4.176829268292683,
"grad_norm": 0.466796875,
"learning_rate": 1.6063457386805004e-05,
"loss": 0.3955,
"step": 685
},
{
"epoch": 4.2073170731707314,
"grad_norm": 0.51953125,
"learning_rate": 1.4925600393157324e-05,
"loss": 0.3847,
"step": 690
},
{
"epoch": 4.237804878048781,
"grad_norm": 0.435546875,
"learning_rate": 1.3826283205802427e-05,
"loss": 0.3829,
"step": 695
},
{
"epoch": 4.2682926829268295,
"grad_norm": 0.47265625,
"learning_rate": 1.2766003829682505e-05,
"loss": 0.3862,
"step": 700
},
{
"epoch": 4.298780487804878,
"grad_norm": 0.4609375,
"learning_rate": 1.1745242585104955e-05,
"loss": 0.3965,
"step": 705
},
{
"epoch": 4.329268292682927,
"grad_norm": 0.494140625,
"learning_rate": 1.0764461890151112e-05,
"loss": 0.3807,
"step": 710
},
{
"epoch": 4.359756097560975,
"grad_norm": 0.466796875,
"learning_rate": 9.824106051194859e-06,
"loss": 0.3722,
"step": 715
},
{
"epoch": 4.390243902439025,
"grad_norm": 0.451171875,
"learning_rate": 8.924601061626048e-06,
"loss": 0.3865,
"step": 720
},
{
"epoch": 4.420731707317073,
"grad_norm": 0.455078125,
"learning_rate": 8.066354408870048e-06,
"loss": 0.3827,
"step": 725
},
{
"epoch": 4.451219512195122,
"grad_norm": 0.470703125,
"learning_rate": 7.249754889790539e-06,
"loss": 0.3932,
"step": 730
},
{
"epoch": 4.4817073170731705,
"grad_norm": 0.44140625,
"learning_rate": 6.475172434559573e-06,
"loss": 0.3762,
"step": 735
},
{
"epoch": 4.512195121951219,
"grad_norm": 0.482421875,
"learning_rate": 5.742957939074412e-06,
"loss": 0.3825,
"step": 740
},
{
"epoch": 4.5426829268292686,
"grad_norm": 0.435546875,
"learning_rate": 5.0534431059970685e-06,
"loss": 0.3846,
"step": 745
},
{
"epoch": 4.573170731707317,
"grad_norm": 0.48828125,
"learning_rate": 4.40694029448877e-06,
"loss": 0.3748,
"step": 750
},
{
"epoch": 4.603658536585366,
"grad_norm": 0.453125,
"learning_rate": 3.803742378707198e-06,
"loss": 0.4205,
"step": 755
},
{
"epoch": 4.634146341463414,
"grad_norm": 0.47265625,
"learning_rate": 3.2441226151306404e-06,
"loss": 0.3816,
"step": 760
},
{
"epoch": 4.664634146341464,
"grad_norm": 0.443359375,
"learning_rate": 2.7283345187693264e-06,
"loss": 0.3782,
"step": 765
},
{
"epoch": 4.695121951219512,
"grad_norm": 0.52734375,
"learning_rate": 2.256611748319792e-06,
"loss": 0.3842,
"step": 770
},
{
"epoch": 4.725609756097561,
"grad_norm": 0.453125,
"learning_rate": 1.8291680003145073e-06,
"loss": 0.4031,
"step": 775
},
{
"epoch": 4.7560975609756095,
"grad_norm": 0.4453125,
"learning_rate": 1.4461969123145457e-06,
"loss": 0.3891,
"step": 780
},
{
"epoch": 4.786585365853659,
"grad_norm": 0.443359375,
"learning_rate": 1.107871975189234e-06,
"loss": 0.365,
"step": 785
},
{
"epoch": 4.817073170731708,
"grad_norm": 0.54296875,
"learning_rate": 8.143464545226298e-07,
"loss": 0.3849,
"step": 790
},
{
"epoch": 4.847560975609756,
"grad_norm": 0.578125,
"learning_rate": 5.657533211820942e-07,
"loss": 0.3848,
"step": 795
},
{
"epoch": 4.878048780487805,
"grad_norm": 0.50390625,
"learning_rate": 3.622051910808666e-07,
"loss": 0.3739,
"step": 800
},
{
"epoch": 4.908536585365853,
"grad_norm": 0.47265625,
"learning_rate": 2.037942741615617e-07,
"loss": 0.392,
"step": 805
},
{
"epoch": 4.939024390243903,
"grad_norm": 0.46484375,
"learning_rate": 9.059233262386225e-08,
"loss": 0.391,
"step": 810
},
{
"epoch": 4.969512195121951,
"grad_norm": 0.419921875,
"learning_rate": 2.2650648415334376e-08,
"loss": 0.3862,
"step": 815
},
{
"epoch": 5.0,
"grad_norm": 0.486328125,
"learning_rate": 0.0,
"loss": 0.4072,
"step": 820
},
{
"epoch": 5.0,
"eval_loss": 1.248147964477539,
"eval_runtime": 1.0787,
"eval_samples_per_second": 1.854,
"eval_steps_per_second": 0.927,
"step": 820
},
{
"epoch": 5.0,
"step": 820,
"total_flos": 6.28853392705323e+17,
"train_loss": 1.6822980254161648,
"train_runtime": 5646.8844,
"train_samples_per_second": 1.159,
"train_steps_per_second": 0.145
}
],
"logging_steps": 5,
"max_steps": 820,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 100,
"total_flos": 6.28853392705323e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}