zephyr-tinyllama-sft-qlora / trainer_state.json
Ritvik19's picture
Upload 13 files
002c335 verified
raw
history blame
20.3 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.999068850303993,
"eval_steps": 500,
"global_step": 570,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0017527523689543736,
"grad_norm": 0.435546875,
"learning_rate": 3.5087719298245615e-06,
"loss": 1.6619,
"step": 1
},
{
"epoch": 0.008763761844771869,
"grad_norm": 0.455078125,
"learning_rate": 1.7543859649122806e-05,
"loss": 1.6809,
"step": 5
},
{
"epoch": 0.017527523689543738,
"grad_norm": 0.455078125,
"learning_rate": 3.508771929824561e-05,
"loss": 1.6799,
"step": 10
},
{
"epoch": 0.026291285534315607,
"grad_norm": 0.4375,
"learning_rate": 5.2631578947368424e-05,
"loss": 1.6325,
"step": 15
},
{
"epoch": 0.035055047379087476,
"grad_norm": 0.40234375,
"learning_rate": 7.017543859649122e-05,
"loss": 1.5568,
"step": 20
},
{
"epoch": 0.04381880922385934,
"grad_norm": 0.2236328125,
"learning_rate": 8.771929824561403e-05,
"loss": 1.4713,
"step": 25
},
{
"epoch": 0.05258257106863121,
"grad_norm": 0.1572265625,
"learning_rate": 0.00010526315789473685,
"loss": 1.397,
"step": 30
},
{
"epoch": 0.06134633291340308,
"grad_norm": 0.11474609375,
"learning_rate": 0.00012280701754385965,
"loss": 1.3622,
"step": 35
},
{
"epoch": 0.07011009475817495,
"grad_norm": 0.0966796875,
"learning_rate": 0.00014035087719298245,
"loss": 1.338,
"step": 40
},
{
"epoch": 0.07887385660294681,
"grad_norm": 0.072265625,
"learning_rate": 0.00015789473684210527,
"loss": 1.3272,
"step": 45
},
{
"epoch": 0.08763761844771868,
"grad_norm": 0.04833984375,
"learning_rate": 0.00017543859649122806,
"loss": 1.3016,
"step": 50
},
{
"epoch": 0.09640138029249055,
"grad_norm": 0.043701171875,
"learning_rate": 0.00019298245614035088,
"loss": 1.2995,
"step": 55
},
{
"epoch": 0.10516514213726243,
"grad_norm": 0.046142578125,
"learning_rate": 0.00019998312416333227,
"loss": 1.2801,
"step": 60
},
{
"epoch": 0.11392890398203429,
"grad_norm": 0.036376953125,
"learning_rate": 0.0001998800146766861,
"loss": 1.2739,
"step": 65
},
{
"epoch": 0.12269266582680616,
"grad_norm": 0.032470703125,
"learning_rate": 0.00019968326771610797,
"loss": 1.2633,
"step": 70
},
{
"epoch": 0.13145642767157803,
"grad_norm": 0.029541015625,
"learning_rate": 0.00019939306773179497,
"loss": 1.2596,
"step": 75
},
{
"epoch": 0.1402201895163499,
"grad_norm": 0.030029296875,
"learning_rate": 0.00019900968678611666,
"loss": 1.2441,
"step": 80
},
{
"epoch": 0.14898395136112177,
"grad_norm": 0.0308837890625,
"learning_rate": 0.00019853348429855672,
"loss": 1.2414,
"step": 85
},
{
"epoch": 0.15774771320589362,
"grad_norm": 0.02978515625,
"learning_rate": 0.0001979649067087574,
"loss": 1.2428,
"step": 90
},
{
"epoch": 0.1665114750506655,
"grad_norm": 0.035888671875,
"learning_rate": 0.00019730448705798239,
"loss": 1.2387,
"step": 95
},
{
"epoch": 0.17527523689543736,
"grad_norm": 0.03125,
"learning_rate": 0.00019655284448939094,
"loss": 1.2568,
"step": 100
},
{
"epoch": 0.18403899874020924,
"grad_norm": 0.03173828125,
"learning_rate": 0.00019571068366759143,
"loss": 1.2212,
"step": 105
},
{
"epoch": 0.1928027605849811,
"grad_norm": 0.0306396484375,
"learning_rate": 0.00019477879411801844,
"loss": 1.24,
"step": 110
},
{
"epoch": 0.20156652242975298,
"grad_norm": 0.0341796875,
"learning_rate": 0.00019375804948675306,
"loss": 1.2301,
"step": 115
},
{
"epoch": 0.21033028427452485,
"grad_norm": 0.03564453125,
"learning_rate": 0.00019264940672148018,
"loss": 1.2178,
"step": 120
},
{
"epoch": 0.2190940461192967,
"grad_norm": 0.03955078125,
"learning_rate": 0.00019145390517435012,
"loss": 1.218,
"step": 125
},
{
"epoch": 0.22785780796406857,
"grad_norm": 0.036865234375,
"learning_rate": 0.00019017266562758659,
"loss": 1.2244,
"step": 130
},
{
"epoch": 0.23662156980884044,
"grad_norm": 0.040771484375,
"learning_rate": 0.00018880688924275378,
"loss": 1.2208,
"step": 135
},
{
"epoch": 0.24538533165361232,
"grad_norm": 0.0380859375,
"learning_rate": 0.00018735785643466784,
"loss": 1.221,
"step": 140
},
{
"epoch": 0.25414909349838416,
"grad_norm": 0.036865234375,
"learning_rate": 0.00018582692567100867,
"loss": 1.2084,
"step": 145
},
{
"epoch": 0.26291285534315606,
"grad_norm": 0.03759765625,
"learning_rate": 0.00018421553219875658,
"loss": 1.2209,
"step": 150
},
{
"epoch": 0.2716766171879279,
"grad_norm": 0.038330078125,
"learning_rate": 0.00018252518669864936,
"loss": 1.2136,
"step": 155
},
{
"epoch": 0.2804403790326998,
"grad_norm": 0.0361328125,
"learning_rate": 0.0001807574738689193,
"loss": 1.2214,
"step": 160
},
{
"epoch": 0.28920414087747165,
"grad_norm": 0.035888671875,
"learning_rate": 0.00017891405093963938,
"loss": 1.2213,
"step": 165
},
{
"epoch": 0.29796790272224355,
"grad_norm": 0.034912109375,
"learning_rate": 0.00017699664611907072,
"loss": 1.2084,
"step": 170
},
{
"epoch": 0.3067316645670154,
"grad_norm": 0.040771484375,
"learning_rate": 0.0001750070569734681,
"loss": 1.2028,
"step": 175
},
{
"epoch": 0.31549542641178724,
"grad_norm": 0.036865234375,
"learning_rate": 0.0001729471487418621,
"loss": 1.2065,
"step": 180
},
{
"epoch": 0.32425918825655914,
"grad_norm": 0.035888671875,
"learning_rate": 0.00017081885258739846,
"loss": 1.2002,
"step": 185
},
{
"epoch": 0.333022950101331,
"grad_norm": 0.037841796875,
"learning_rate": 0.0001686241637868734,
"loss": 1.2058,
"step": 190
},
{
"epoch": 0.3417867119461029,
"grad_norm": 0.0361328125,
"learning_rate": 0.00016636513986016213,
"loss": 1.2084,
"step": 195
},
{
"epoch": 0.35055047379087473,
"grad_norm": 0.03759765625,
"learning_rate": 0.00016404389864129533,
"loss": 1.199,
"step": 200
},
{
"epoch": 0.35931423563564663,
"grad_norm": 0.042724609375,
"learning_rate": 0.00016166261629298995,
"loss": 1.2023,
"step": 205
},
{
"epoch": 0.3680779974804185,
"grad_norm": 0.044189453125,
"learning_rate": 0.00015922352526649803,
"loss": 1.1999,
"step": 210
},
{
"epoch": 0.3768417593251903,
"grad_norm": 0.036376953125,
"learning_rate": 0.00015672891220868432,
"loss": 1.1882,
"step": 215
},
{
"epoch": 0.3856055211699622,
"grad_norm": 0.041748046875,
"learning_rate": 0.00015418111581829574,
"loss": 1.2016,
"step": 220
},
{
"epoch": 0.39436928301473406,
"grad_norm": 0.041748046875,
"learning_rate": 0.00015158252465343242,
"loss": 1.1994,
"step": 225
},
{
"epoch": 0.40313304485950596,
"grad_norm": 0.048828125,
"learning_rate": 0.00014893557489227517,
"loss": 1.2054,
"step": 230
},
{
"epoch": 0.4118968067042778,
"grad_norm": 0.0419921875,
"learning_rate": 0.00014624274804916958,
"loss": 1.1868,
"step": 235
},
{
"epoch": 0.4206605685490497,
"grad_norm": 0.041259765625,
"learning_rate": 0.00014350656864820733,
"loss": 1.2056,
"step": 240
},
{
"epoch": 0.42942433039382155,
"grad_norm": 0.039306640625,
"learning_rate": 0.00014072960185648577,
"loss": 1.2042,
"step": 245
},
{
"epoch": 0.4381880922385934,
"grad_norm": 0.037353515625,
"learning_rate": 0.00013791445107926478,
"loss": 1.1992,
"step": 250
},
{
"epoch": 0.4469518540833653,
"grad_norm": 0.0380859375,
"learning_rate": 0.00013506375551927547,
"loss": 1.2025,
"step": 255
},
{
"epoch": 0.45571561592813714,
"grad_norm": 0.036865234375,
"learning_rate": 0.00013218018770246858,
"loss": 1.2102,
"step": 260
},
{
"epoch": 0.46447937777290904,
"grad_norm": 0.036865234375,
"learning_rate": 0.0001292664509725226,
"loss": 1.2105,
"step": 265
},
{
"epoch": 0.4732431396176809,
"grad_norm": 0.044189453125,
"learning_rate": 0.00012632527695645993,
"loss": 1.1963,
"step": 270
},
{
"epoch": 0.48200690146245273,
"grad_norm": 0.0380859375,
"learning_rate": 0.00012335942300374788,
"loss": 1.1911,
"step": 275
},
{
"epoch": 0.49077066330722463,
"grad_norm": 0.037109375,
"learning_rate": 0.00012037166960128443,
"loss": 1.1901,
"step": 280
},
{
"epoch": 0.4995344251519965,
"grad_norm": 0.035888671875,
"learning_rate": 0.00011736481776669306,
"loss": 1.1952,
"step": 285
},
{
"epoch": 0.5082981869967683,
"grad_norm": 0.0361328125,
"learning_rate": 0.00011434168642236964,
"loss": 1.1864,
"step": 290
},
{
"epoch": 0.5170619488415402,
"grad_norm": 0.0390625,
"learning_rate": 0.00011130510975274409,
"loss": 1.1909,
"step": 295
},
{
"epoch": 0.5258257106863121,
"grad_norm": 0.039794921875,
"learning_rate": 0.00010825793454723325,
"loss": 1.1998,
"step": 300
},
{
"epoch": 0.534589472531084,
"grad_norm": 0.043212890625,
"learning_rate": 0.00010520301753137724,
"loss": 1.1918,
"step": 305
},
{
"epoch": 0.5433532343758558,
"grad_norm": 0.037109375,
"learning_rate": 0.00010214322268866032,
"loss": 1.1821,
"step": 310
},
{
"epoch": 0.5521169962206277,
"grad_norm": 0.044189453125,
"learning_rate": 9.908141857552737e-05,
"loss": 1.1964,
"step": 315
},
{
"epoch": 0.5608807580653996,
"grad_norm": 0.037109375,
"learning_rate": 9.602047563211359e-05,
"loss": 1.1802,
"step": 320
},
{
"epoch": 0.5696445199101714,
"grad_norm": 0.038818359375,
"learning_rate": 9.296326349120785e-05,
"loss": 1.1869,
"step": 325
},
{
"epoch": 0.5784082817549433,
"grad_norm": 0.039306640625,
"learning_rate": 8.991264828797319e-05,
"loss": 1.2014,
"step": 330
},
{
"epoch": 0.5871720435997152,
"grad_norm": 0.03515625,
"learning_rate": 8.687148997294621e-05,
"loss": 1.1903,
"step": 335
},
{
"epoch": 0.5959358054444871,
"grad_norm": 0.03857421875,
"learning_rate": 8.384263963083453e-05,
"loss": 1.1862,
"step": 340
},
{
"epoch": 0.6046995672892589,
"grad_norm": 0.037353515625,
"learning_rate": 8.082893680762619e-05,
"loss": 1.1919,
"step": 345
},
{
"epoch": 0.6134633291340308,
"grad_norm": 0.03662109375,
"learning_rate": 7.783320684851614e-05,
"loss": 1.1874,
"step": 350
},
{
"epoch": 0.6222270909788027,
"grad_norm": 0.0380859375,
"learning_rate": 7.485825824914659e-05,
"loss": 1.2009,
"step": 355
},
{
"epoch": 0.6309908528235745,
"grad_norm": 0.03955078125,
"learning_rate": 7.190688002264308e-05,
"loss": 1.1853,
"step": 360
},
{
"epoch": 0.6397546146683464,
"grad_norm": 0.037841796875,
"learning_rate": 6.898183908491617e-05,
"loss": 1.1868,
"step": 365
},
{
"epoch": 0.6485183765131183,
"grad_norm": 0.039306640625,
"learning_rate": 6.608587766067852e-05,
"loss": 1.1845,
"step": 370
},
{
"epoch": 0.6572821383578902,
"grad_norm": 0.038818359375,
"learning_rate": 6.322171071261071e-05,
"loss": 1.1923,
"step": 375
},
{
"epoch": 0.666045900202662,
"grad_norm": 0.0380859375,
"learning_rate": 6.039202339608432e-05,
"loss": 1.1929,
"step": 380
},
{
"epoch": 0.6748096620474339,
"grad_norm": 0.038330078125,
"learning_rate": 5.7599468541830356e-05,
"loss": 1.1872,
"step": 385
},
{
"epoch": 0.6835734238922058,
"grad_norm": 0.03759765625,
"learning_rate": 5.484666416891109e-05,
"loss": 1.1766,
"step": 390
},
{
"epoch": 0.6923371857369776,
"grad_norm": 0.036865234375,
"learning_rate": 5.2136191030328455e-05,
"loss": 1.1938,
"step": 395
},
{
"epoch": 0.7011009475817495,
"grad_norm": 0.038330078125,
"learning_rate": 4.9470590193569044e-05,
"loss": 1.1953,
"step": 400
},
{
"epoch": 0.7098647094265214,
"grad_norm": 0.037353515625,
"learning_rate": 4.685236065835443e-05,
"loss": 1.1926,
"step": 405
},
{
"epoch": 0.7186284712712933,
"grad_norm": 0.03466796875,
"learning_rate": 4.4283957013829846e-05,
"loss": 1.1892,
"step": 410
},
{
"epoch": 0.727392233116065,
"grad_norm": 0.037841796875,
"learning_rate": 4.176778713738787e-05,
"loss": 1.1871,
"step": 415
},
{
"epoch": 0.736155994960837,
"grad_norm": 0.035888671875,
"learning_rate": 3.9306209937284346e-05,
"loss": 1.1879,
"step": 420
},
{
"epoch": 0.7449197568056088,
"grad_norm": 0.034423828125,
"learning_rate": 3.69015331411628e-05,
"loss": 1.2001,
"step": 425
},
{
"epoch": 0.7536835186503806,
"grad_norm": 0.03662109375,
"learning_rate": 3.455601113256073e-05,
"loss": 1.1817,
"step": 430
},
{
"epoch": 0.7624472804951525,
"grad_norm": 0.036865234375,
"learning_rate": 3.227184283742591e-05,
"loss": 1.19,
"step": 435
},
{
"epoch": 0.7712110423399244,
"grad_norm": 0.036376953125,
"learning_rate": 3.0051169662624225e-05,
"loss": 1.201,
"step": 440
},
{
"epoch": 0.7799748041846963,
"grad_norm": 0.035888671875,
"learning_rate": 2.789607348837153e-05,
"loss": 1.1962,
"step": 445
},
{
"epoch": 0.7887385660294681,
"grad_norm": 0.035400390625,
"learning_rate": 2.5808574716471856e-05,
"loss": 1.1805,
"step": 450
},
{
"epoch": 0.79750232787424,
"grad_norm": 0.036865234375,
"learning_rate": 2.379063037619146e-05,
"loss": 1.1879,
"step": 455
},
{
"epoch": 0.8062660897190119,
"grad_norm": 0.0361328125,
"learning_rate": 2.184413228954468e-05,
"loss": 1.1902,
"step": 460
},
{
"epoch": 0.8150298515637837,
"grad_norm": 0.035888671875,
"learning_rate": 1.9970905297711606e-05,
"loss": 1.1754,
"step": 465
},
{
"epoch": 0.8237936134085556,
"grad_norm": 0.033935546875,
"learning_rate": 1.8172705550250092e-05,
"loss": 1.1926,
"step": 470
},
{
"epoch": 0.8325573752533275,
"grad_norm": 0.034912109375,
"learning_rate": 1.6451218858706374e-05,
"loss": 1.1832,
"step": 475
},
{
"epoch": 0.8413211370980994,
"grad_norm": 0.033935546875,
"learning_rate": 1.4808059116167305e-05,
"loss": 1.1767,
"step": 480
},
{
"epoch": 0.8500848989428712,
"grad_norm": 0.03515625,
"learning_rate": 1.3244766784236307e-05,
"loss": 1.1978,
"step": 485
},
{
"epoch": 0.8588486607876431,
"grad_norm": 0.035888671875,
"learning_rate": 1.176280744885121e-05,
"loss": 1.1975,
"step": 490
},
{
"epoch": 0.867612422632415,
"grad_norm": 0.03466796875,
"learning_rate": 1.0363570446297999e-05,
"loss": 1.1886,
"step": 495
},
{
"epoch": 0.8763761844771868,
"grad_norm": 0.03466796875,
"learning_rate": 9.048367560708604e-06,
"loss": 1.1954,
"step": 500
},
{
"epoch": 0.8851399463219587,
"grad_norm": 0.036376953125,
"learning_rate": 7.818431794263836e-06,
"loss": 1.1965,
"step": 505
},
{
"epoch": 0.8939037081667306,
"grad_norm": 0.03515625,
"learning_rate": 6.674916211254289e-06,
"loss": 1.2116,
"step": 510
},
{
"epoch": 0.9026674700115024,
"grad_norm": 0.033935546875,
"learning_rate": 5.618892857083069e-06,
"loss": 1.1851,
"step": 515
},
{
"epoch": 0.9114312318562743,
"grad_norm": 0.035400390625,
"learning_rate": 4.65135175322361e-06,
"loss": 1.1957,
"step": 520
},
{
"epoch": 0.9201949937010462,
"grad_norm": 0.0361328125,
"learning_rate": 3.7731999690749585e-06,
"loss": 1.1806,
"step": 525
},
{
"epoch": 0.9289587555458181,
"grad_norm": 0.034423828125,
"learning_rate": 2.9852607715846193e-06,
"loss": 1.1924,
"step": 530
},
{
"epoch": 0.9377225173905899,
"grad_norm": 0.03564453125,
"learning_rate": 2.288272853436013e-06,
"loss": 1.1911,
"step": 535
},
{
"epoch": 0.9464862792353618,
"grad_norm": 0.033935546875,
"learning_rate": 1.6828896405244988e-06,
"loss": 1.1815,
"step": 540
},
{
"epoch": 0.9552500410801337,
"grad_norm": 0.034423828125,
"learning_rate": 1.1696786793707781e-06,
"loss": 1.2068,
"step": 545
},
{
"epoch": 0.9640138029249055,
"grad_norm": 0.03515625,
"learning_rate": 7.491211050462798e-07,
"loss": 1.1929,
"step": 550
},
{
"epoch": 0.9727775647696774,
"grad_norm": 0.035400390625,
"learning_rate": 4.216111901092501e-07,
"loss": 1.1962,
"step": 555
},
{
"epoch": 0.9815413266144493,
"grad_norm": 0.036865234375,
"learning_rate": 1.8745597497433765e-07,
"loss": 1.1793,
"step": 560
},
{
"epoch": 0.9903050884592212,
"grad_norm": 0.03515625,
"learning_rate": 4.687498006236135e-08,
"loss": 1.2006,
"step": 565
},
{
"epoch": 0.999068850303993,
"grad_norm": 0.034912109375,
"learning_rate": 0.0,
"loss": 1.1908,
"step": 570
},
{
"epoch": 0.999068850303993,
"eval_loss": 1.1943339109420776,
"eval_runtime": 1451.0124,
"eval_samples_per_second": 11.141,
"eval_steps_per_second": 2.786,
"step": 570
},
{
"epoch": 0.999068850303993,
"step": 570,
"total_flos": 1.8775686401910374e+18,
"train_loss": 0.1465015503398159,
"train_runtime": 6527.562,
"train_samples_per_second": 22.375,
"train_steps_per_second": 0.087
}
],
"logging_steps": 5,
"max_steps": 570,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"total_flos": 1.8775686401910374e+18,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}