norllm-ai-normistral-7b-sft-qlora / trainer_state.json
hugodk-sch's picture
Model save
3d27451 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.990892531876138,
"eval_steps": 500,
"global_step": 1370,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"grad_norm": 0.1083984375,
"learning_rate": 1.4598540145985402e-06,
"loss": 2.234,
"step": 1
},
{
"epoch": 0.02,
"grad_norm": 0.09521484375,
"learning_rate": 7.2992700729927e-06,
"loss": 2.1882,
"step": 5
},
{
"epoch": 0.04,
"grad_norm": 0.095703125,
"learning_rate": 1.45985401459854e-05,
"loss": 2.2055,
"step": 10
},
{
"epoch": 0.05,
"grad_norm": 0.111328125,
"learning_rate": 2.1897810218978105e-05,
"loss": 2.176,
"step": 15
},
{
"epoch": 0.07,
"grad_norm": 0.109375,
"learning_rate": 2.91970802919708e-05,
"loss": 2.1559,
"step": 20
},
{
"epoch": 0.09,
"grad_norm": 0.11962890625,
"learning_rate": 3.649635036496351e-05,
"loss": 2.16,
"step": 25
},
{
"epoch": 0.11,
"grad_norm": 0.1123046875,
"learning_rate": 4.379562043795621e-05,
"loss": 2.1681,
"step": 30
},
{
"epoch": 0.13,
"grad_norm": 0.11572265625,
"learning_rate": 5.109489051094891e-05,
"loss": 2.0763,
"step": 35
},
{
"epoch": 0.15,
"grad_norm": 0.1171875,
"learning_rate": 5.83941605839416e-05,
"loss": 2.114,
"step": 40
},
{
"epoch": 0.16,
"grad_norm": 0.109375,
"learning_rate": 6.56934306569343e-05,
"loss": 2.0984,
"step": 45
},
{
"epoch": 0.18,
"grad_norm": 0.111328125,
"learning_rate": 7.299270072992701e-05,
"loss": 2.0837,
"step": 50
},
{
"epoch": 0.2,
"grad_norm": 0.1328125,
"learning_rate": 8.029197080291971e-05,
"loss": 2.0903,
"step": 55
},
{
"epoch": 0.22,
"grad_norm": 0.1123046875,
"learning_rate": 8.759124087591242e-05,
"loss": 2.0727,
"step": 60
},
{
"epoch": 0.24,
"grad_norm": 0.1201171875,
"learning_rate": 9.489051094890511e-05,
"loss": 2.0506,
"step": 65
},
{
"epoch": 0.26,
"grad_norm": 0.1181640625,
"learning_rate": 0.00010218978102189782,
"loss": 2.0916,
"step": 70
},
{
"epoch": 0.27,
"grad_norm": 0.1337890625,
"learning_rate": 0.00010948905109489052,
"loss": 2.0582,
"step": 75
},
{
"epoch": 0.29,
"grad_norm": 0.1279296875,
"learning_rate": 0.0001167883211678832,
"loss": 2.0502,
"step": 80
},
{
"epoch": 0.31,
"grad_norm": 0.12451171875,
"learning_rate": 0.0001240875912408759,
"loss": 2.0715,
"step": 85
},
{
"epoch": 0.33,
"grad_norm": 0.1298828125,
"learning_rate": 0.0001313868613138686,
"loss": 1.9955,
"step": 90
},
{
"epoch": 0.35,
"grad_norm": 0.146484375,
"learning_rate": 0.00013868613138686133,
"loss": 2.0099,
"step": 95
},
{
"epoch": 0.36,
"grad_norm": 0.1376953125,
"learning_rate": 0.00014598540145985403,
"loss": 2.0107,
"step": 100
},
{
"epoch": 0.38,
"grad_norm": 0.1533203125,
"learning_rate": 0.00015328467153284672,
"loss": 1.9787,
"step": 105
},
{
"epoch": 0.4,
"grad_norm": 0.1689453125,
"learning_rate": 0.00016058394160583942,
"loss": 1.992,
"step": 110
},
{
"epoch": 0.42,
"grad_norm": 0.1748046875,
"learning_rate": 0.00016788321167883211,
"loss": 2.0616,
"step": 115
},
{
"epoch": 0.44,
"grad_norm": 0.1748046875,
"learning_rate": 0.00017518248175182484,
"loss": 2.0284,
"step": 120
},
{
"epoch": 0.46,
"grad_norm": 0.197265625,
"learning_rate": 0.00018248175182481753,
"loss": 1.9884,
"step": 125
},
{
"epoch": 0.47,
"grad_norm": 0.2578125,
"learning_rate": 0.00018978102189781023,
"loss": 1.995,
"step": 130
},
{
"epoch": 0.49,
"grad_norm": 0.2294921875,
"learning_rate": 0.00019708029197080293,
"loss": 2.0103,
"step": 135
},
{
"epoch": 0.51,
"grad_norm": 0.2021484375,
"learning_rate": 0.00019999707864731247,
"loss": 1.9678,
"step": 140
},
{
"epoch": 0.53,
"grad_norm": 0.255859375,
"learning_rate": 0.0001999792265545627,
"loss": 2.0033,
"step": 145
},
{
"epoch": 0.55,
"grad_norm": 0.25,
"learning_rate": 0.00019994514823658906,
"loss": 1.9277,
"step": 150
},
{
"epoch": 0.56,
"grad_norm": 0.259765625,
"learning_rate": 0.00019989484922416502,
"loss": 2.0091,
"step": 155
},
{
"epoch": 0.58,
"grad_norm": 0.25,
"learning_rate": 0.0001998283376806175,
"loss": 1.9636,
"step": 160
},
{
"epoch": 0.6,
"grad_norm": 0.25390625,
"learning_rate": 0.00019974562440050177,
"loss": 1.9385,
"step": 165
},
{
"epoch": 0.62,
"grad_norm": 0.259765625,
"learning_rate": 0.00019964672280784954,
"loss": 1.9536,
"step": 170
},
{
"epoch": 0.64,
"grad_norm": 0.244140625,
"learning_rate": 0.00019953164895399042,
"loss": 1.9552,
"step": 175
},
{
"epoch": 0.66,
"grad_norm": 0.28125,
"learning_rate": 0.00019940042151494675,
"loss": 1.8901,
"step": 180
},
{
"epoch": 0.67,
"grad_norm": 0.283203125,
"learning_rate": 0.0001992530617884026,
"loss": 1.9541,
"step": 185
},
{
"epoch": 0.69,
"grad_norm": 0.26953125,
"learning_rate": 0.00019908959369024728,
"loss": 1.9515,
"step": 190
},
{
"epoch": 0.71,
"grad_norm": 0.279296875,
"learning_rate": 0.00019891004375069377,
"loss": 1.896,
"step": 195
},
{
"epoch": 0.73,
"grad_norm": 0.279296875,
"learning_rate": 0.0001987144411099731,
"loss": 1.9335,
"step": 200
},
{
"epoch": 0.75,
"grad_norm": 0.380859375,
"learning_rate": 0.00019850281751360497,
"loss": 1.8805,
"step": 205
},
{
"epoch": 0.77,
"grad_norm": 0.296875,
"learning_rate": 0.00019827520730724543,
"loss": 1.8663,
"step": 210
},
{
"epoch": 0.78,
"grad_norm": 0.3125,
"learning_rate": 0.00019803164743111302,
"loss": 1.8169,
"step": 215
},
{
"epoch": 0.8,
"grad_norm": 0.3359375,
"learning_rate": 0.0001977721774139933,
"loss": 1.861,
"step": 220
},
{
"epoch": 0.82,
"grad_norm": 0.296875,
"learning_rate": 0.00019749683936682352,
"loss": 1.8507,
"step": 225
},
{
"epoch": 0.84,
"grad_norm": 0.3046875,
"learning_rate": 0.00019720567797585817,
"loss": 1.8928,
"step": 230
},
{
"epoch": 0.86,
"grad_norm": 0.296875,
"learning_rate": 0.00019689874049541673,
"loss": 1.8767,
"step": 235
},
{
"epoch": 0.87,
"grad_norm": 0.318359375,
"learning_rate": 0.00019657607674021424,
"loss": 1.9053,
"step": 240
},
{
"epoch": 0.89,
"grad_norm": 0.32421875,
"learning_rate": 0.00019623773907727682,
"loss": 1.7781,
"step": 245
},
{
"epoch": 0.91,
"grad_norm": 0.302734375,
"learning_rate": 0.00019588378241744254,
"loss": 1.8099,
"step": 250
},
{
"epoch": 0.93,
"grad_norm": 0.302734375,
"learning_rate": 0.00019551426420644967,
"loss": 1.8546,
"step": 255
},
{
"epoch": 0.95,
"grad_norm": 0.369140625,
"learning_rate": 0.00019512924441561348,
"loss": 1.8158,
"step": 260
},
{
"epoch": 0.97,
"grad_norm": 0.318359375,
"learning_rate": 0.00019472878553209312,
"loss": 1.7982,
"step": 265
},
{
"epoch": 0.98,
"grad_norm": 0.39453125,
"learning_rate": 0.0001943129525487502,
"loss": 1.7274,
"step": 270
},
{
"epoch": 1.0,
"eval_loss": 1.9431506395339966,
"eval_runtime": 74.7123,
"eval_samples_per_second": 3.667,
"eval_steps_per_second": 0.468,
"step": 274
},
{
"epoch": 1.0,
"grad_norm": 0.55078125,
"learning_rate": 0.00019388181295360078,
"loss": 1.7478,
"step": 275
},
{
"epoch": 1.02,
"grad_norm": 0.439453125,
"learning_rate": 0.00019343543671886218,
"loss": 1.61,
"step": 280
},
{
"epoch": 1.04,
"grad_norm": 0.435546875,
"learning_rate": 0.00019297389628959702,
"loss": 1.6174,
"step": 285
},
{
"epoch": 1.06,
"grad_norm": 0.412109375,
"learning_rate": 0.00019249726657195532,
"loss": 1.6191,
"step": 290
},
{
"epoch": 1.07,
"grad_norm": 0.392578125,
"learning_rate": 0.00019200562492101798,
"loss": 1.6129,
"step": 295
},
{
"epoch": 1.09,
"grad_norm": 0.41015625,
"learning_rate": 0.00019149905112824197,
"loss": 1.6052,
"step": 300
},
{
"epoch": 1.11,
"grad_norm": 0.44921875,
"learning_rate": 0.00019097762740851061,
"loss": 1.4974,
"step": 305
},
{
"epoch": 1.13,
"grad_norm": 0.4375,
"learning_rate": 0.0001904414383867907,
"loss": 1.598,
"step": 310
},
{
"epoch": 1.15,
"grad_norm": 0.4609375,
"learning_rate": 0.00018989057108439777,
"loss": 1.5788,
"step": 315
},
{
"epoch": 1.17,
"grad_norm": 0.373046875,
"learning_rate": 0.0001893251149048732,
"loss": 1.6078,
"step": 320
},
{
"epoch": 1.18,
"grad_norm": 0.55078125,
"learning_rate": 0.0001887451616194743,
"loss": 1.527,
"step": 325
},
{
"epoch": 1.2,
"grad_norm": 0.482421875,
"learning_rate": 0.0001881508053522801,
"loss": 1.4852,
"step": 330
},
{
"epoch": 1.22,
"grad_norm": 0.458984375,
"learning_rate": 0.00018754214256491562,
"loss": 1.5538,
"step": 335
},
{
"epoch": 1.24,
"grad_norm": 0.439453125,
"learning_rate": 0.0001869192720408963,
"loss": 1.5196,
"step": 340
},
{
"epoch": 1.26,
"grad_norm": 0.49609375,
"learning_rate": 0.00018628229486959596,
"loss": 1.495,
"step": 345
},
{
"epoch": 1.28,
"grad_norm": 0.443359375,
"learning_rate": 0.00018563131442984044,
"loss": 1.465,
"step": 350
},
{
"epoch": 1.29,
"grad_norm": 0.447265625,
"learning_rate": 0.00018496643637312942,
"loss": 1.3923,
"step": 355
},
{
"epoch": 1.31,
"grad_norm": 0.455078125,
"learning_rate": 0.00018428776860648994,
"loss": 1.5173,
"step": 360
},
{
"epoch": 1.33,
"grad_norm": 0.494140625,
"learning_rate": 0.0001835954212749632,
"loss": 1.4434,
"step": 365
},
{
"epoch": 1.35,
"grad_norm": 0.4765625,
"learning_rate": 0.0001828895067437287,
"loss": 1.4619,
"step": 370
},
{
"epoch": 1.37,
"grad_norm": 0.55078125,
"learning_rate": 0.0001821701395798677,
"loss": 1.3675,
"step": 375
},
{
"epoch": 1.38,
"grad_norm": 0.58984375,
"learning_rate": 0.00018143743653376942,
"loss": 1.4109,
"step": 380
},
{
"epoch": 1.4,
"grad_norm": 0.55078125,
"learning_rate": 0.0001806915165201831,
"loss": 1.3592,
"step": 385
},
{
"epoch": 1.42,
"grad_norm": 0.466796875,
"learning_rate": 0.00017993250059891838,
"loss": 1.3634,
"step": 390
},
{
"epoch": 1.44,
"grad_norm": 0.55859375,
"learning_rate": 0.00017916051195519797,
"loss": 1.2936,
"step": 395
},
{
"epoch": 1.46,
"grad_norm": 0.48828125,
"learning_rate": 0.00017837567587966507,
"loss": 1.4582,
"step": 400
},
{
"epoch": 1.48,
"grad_norm": 0.5078125,
"learning_rate": 0.00017757811974804927,
"loss": 1.3851,
"step": 405
},
{
"epoch": 1.49,
"grad_norm": 0.5234375,
"learning_rate": 0.00017676797300049393,
"loss": 1.4325,
"step": 410
},
{
"epoch": 1.51,
"grad_norm": 0.55859375,
"learning_rate": 0.0001759453671205487,
"loss": 1.3731,
"step": 415
},
{
"epoch": 1.53,
"grad_norm": 0.44921875,
"learning_rate": 0.00017511043561382997,
"loss": 1.5819,
"step": 420
},
{
"epoch": 1.55,
"grad_norm": 0.4921875,
"learning_rate": 0.0001742633139863538,
"loss": 1.4186,
"step": 425
},
{
"epoch": 1.57,
"grad_norm": 0.5859375,
"learning_rate": 0.00017340413972254367,
"loss": 1.2765,
"step": 430
},
{
"epoch": 1.58,
"grad_norm": 0.498046875,
"learning_rate": 0.00017253305226291724,
"loss": 1.2749,
"step": 435
},
{
"epoch": 1.6,
"grad_norm": 0.60546875,
"learning_rate": 0.00017165019298145585,
"loss": 1.3995,
"step": 440
},
{
"epoch": 1.62,
"grad_norm": 0.55078125,
"learning_rate": 0.00017075570516266012,
"loss": 1.3697,
"step": 445
},
{
"epoch": 1.64,
"grad_norm": 0.58984375,
"learning_rate": 0.00016984973397829524,
"loss": 1.2694,
"step": 450
},
{
"epoch": 1.66,
"grad_norm": 0.5390625,
"learning_rate": 0.0001689324264638304,
"loss": 1.3461,
"step": 455
},
{
"epoch": 1.68,
"grad_norm": 0.5859375,
"learning_rate": 0.00016800393149457539,
"loss": 1.2316,
"step": 460
},
{
"epoch": 1.69,
"grad_norm": 0.6015625,
"learning_rate": 0.0001670643997615189,
"loss": 1.2794,
"step": 465
},
{
"epoch": 1.71,
"grad_norm": 0.53125,
"learning_rate": 0.0001661139837468717,
"loss": 1.185,
"step": 470
},
{
"epoch": 1.73,
"grad_norm": 0.58203125,
"learning_rate": 0.00016515283769931972,
"loss": 1.2398,
"step": 475
},
{
"epoch": 1.75,
"grad_norm": 0.55859375,
"learning_rate": 0.0001641811176089899,
"loss": 1.2424,
"step": 480
},
{
"epoch": 1.77,
"grad_norm": 0.60546875,
"learning_rate": 0.00016319898118213365,
"loss": 1.24,
"step": 485
},
{
"epoch": 1.79,
"grad_norm": 0.58984375,
"learning_rate": 0.00016220658781553176,
"loss": 1.2045,
"step": 490
},
{
"epoch": 1.8,
"grad_norm": 0.60546875,
"learning_rate": 0.00016120409857062507,
"loss": 1.2237,
"step": 495
},
{
"epoch": 1.82,
"grad_norm": 0.61328125,
"learning_rate": 0.0001601916761473747,
"loss": 1.329,
"step": 500
},
{
"epoch": 1.84,
"grad_norm": 0.7265625,
"learning_rate": 0.00015916948485785647,
"loss": 1.0863,
"step": 505
},
{
"epoch": 1.86,
"grad_norm": 0.55078125,
"learning_rate": 0.0001581376905995939,
"loss": 1.2565,
"step": 510
},
{
"epoch": 1.88,
"grad_norm": 0.59765625,
"learning_rate": 0.0001570964608286336,
"loss": 1.127,
"step": 515
},
{
"epoch": 1.89,
"grad_norm": 0.625,
"learning_rate": 0.0001560459645323678,
"loss": 1.147,
"step": 520
},
{
"epoch": 1.91,
"grad_norm": 0.62890625,
"learning_rate": 0.0001549863722021084,
"loss": 1.1599,
"step": 525
},
{
"epoch": 1.93,
"grad_norm": 0.65234375,
"learning_rate": 0.00015391785580541698,
"loss": 1.1667,
"step": 530
},
{
"epoch": 1.95,
"grad_norm": 0.6015625,
"learning_rate": 0.00015284058875819504,
"loss": 1.2302,
"step": 535
},
{
"epoch": 1.97,
"grad_norm": 0.66796875,
"learning_rate": 0.00015175474589653942,
"loss": 1.1781,
"step": 540
},
{
"epoch": 1.99,
"grad_norm": 0.734375,
"learning_rate": 0.00015066050344836706,
"loss": 1.1514,
"step": 545
},
{
"epoch": 2.0,
"eval_loss": 1.7110929489135742,
"eval_runtime": 74.7203,
"eval_samples_per_second": 3.667,
"eval_steps_per_second": 0.468,
"step": 549
},
{
"epoch": 2.0,
"grad_norm": 0.53515625,
"learning_rate": 0.00014955803900481389,
"loss": 1.0657,
"step": 550
},
{
"epoch": 2.02,
"grad_norm": 0.8125,
"learning_rate": 0.0001484475314914125,
"loss": 0.9519,
"step": 555
},
{
"epoch": 2.04,
"grad_norm": 0.66796875,
"learning_rate": 0.00014732916113905335,
"loss": 0.8776,
"step": 560
},
{
"epoch": 2.06,
"grad_norm": 0.76171875,
"learning_rate": 0.0001462031094547339,
"loss": 0.9081,
"step": 565
},
{
"epoch": 2.08,
"grad_norm": 0.703125,
"learning_rate": 0.0001450695591921008,
"loss": 0.9279,
"step": 570
},
{
"epoch": 2.09,
"grad_norm": 0.73828125,
"learning_rate": 0.00014392869432178971,
"loss": 0.7395,
"step": 575
},
{
"epoch": 2.11,
"grad_norm": 0.671875,
"learning_rate": 0.00014278070000156768,
"loss": 0.9795,
"step": 580
},
{
"epoch": 2.13,
"grad_norm": 0.72265625,
"learning_rate": 0.00014162576254628267,
"loss": 0.8563,
"step": 585
},
{
"epoch": 2.15,
"grad_norm": 0.73046875,
"learning_rate": 0.00014046406939762545,
"loss": 0.8419,
"step": 590
},
{
"epoch": 2.17,
"grad_norm": 0.70703125,
"learning_rate": 0.00013929580909370862,
"loss": 0.82,
"step": 595
},
{
"epoch": 2.19,
"grad_norm": 0.67578125,
"learning_rate": 0.00013812117123846747,
"loss": 0.9267,
"step": 600
},
{
"epoch": 2.2,
"grad_norm": 0.69921875,
"learning_rate": 0.0001369403464708884,
"loss": 0.8248,
"step": 605
},
{
"epoch": 2.22,
"grad_norm": 0.73046875,
"learning_rate": 0.00013575352643406844,
"loss": 0.8331,
"step": 610
},
{
"epoch": 2.24,
"grad_norm": 0.69140625,
"learning_rate": 0.00013456090374411284,
"loss": 0.8908,
"step": 615
},
{
"epoch": 2.26,
"grad_norm": 0.734375,
"learning_rate": 0.00013336267195887398,
"loss": 0.8693,
"step": 620
},
{
"epoch": 2.28,
"grad_norm": 0.953125,
"learning_rate": 0.00013215902554653772,
"loss": 0.8886,
"step": 625
},
{
"epoch": 2.3,
"grad_norm": 0.8046875,
"learning_rate": 0.00013095015985406215,
"loss": 0.7254,
"step": 630
},
{
"epoch": 2.31,
"grad_norm": 0.7421875,
"learning_rate": 0.00012973627107547346,
"loss": 0.8929,
"step": 635
},
{
"epoch": 2.33,
"grad_norm": 0.8671875,
"learning_rate": 0.00012851755622002444,
"loss": 0.7416,
"step": 640
},
{
"epoch": 2.35,
"grad_norm": 0.6796875,
"learning_rate": 0.0001272942130802207,
"loss": 0.7426,
"step": 645
},
{
"epoch": 2.37,
"grad_norm": 0.73046875,
"learning_rate": 0.00012606644019971968,
"loss": 0.7973,
"step": 650
},
{
"epoch": 2.39,
"grad_norm": 0.7265625,
"learning_rate": 0.00012483443684110793,
"loss": 0.7846,
"step": 655
},
{
"epoch": 2.4,
"grad_norm": 0.7265625,
"learning_rate": 0.00012359840295356151,
"loss": 0.9146,
"step": 660
},
{
"epoch": 2.42,
"grad_norm": 0.734375,
"learning_rate": 0.00012235853914039515,
"loss": 0.8241,
"step": 665
},
{
"epoch": 2.44,
"grad_norm": 0.6640625,
"learning_rate": 0.0001211150466265049,
"loss": 0.811,
"step": 670
},
{
"epoch": 2.46,
"grad_norm": 0.66015625,
"learning_rate": 0.00011986812722571039,
"loss": 0.7336,
"step": 675
},
{
"epoch": 2.48,
"grad_norm": 0.6875,
"learning_rate": 0.00011861798330800125,
"loss": 0.727,
"step": 680
},
{
"epoch": 2.5,
"grad_norm": 0.7578125,
"learning_rate": 0.00011736481776669306,
"loss": 0.6652,
"step": 685
},
{
"epoch": 2.51,
"grad_norm": 0.625,
"learning_rate": 0.00011610883398549879,
"loss": 0.6868,
"step": 690
},
{
"epoch": 2.53,
"grad_norm": 0.734375,
"learning_rate": 0.00011485023580552039,
"loss": 0.7602,
"step": 695
},
{
"epoch": 2.55,
"grad_norm": 0.73046875,
"learning_rate": 0.00011358922749216614,
"loss": 0.7176,
"step": 700
},
{
"epoch": 2.57,
"grad_norm": 0.73828125,
"learning_rate": 0.00011232601370199936,
"loss": 0.7167,
"step": 705
},
{
"epoch": 2.59,
"grad_norm": 0.76953125,
"learning_rate": 0.00011106079944952317,
"loss": 0.7278,
"step": 710
},
{
"epoch": 2.6,
"grad_norm": 0.75390625,
"learning_rate": 0.00010979379007390782,
"loss": 0.778,
"step": 715
},
{
"epoch": 2.62,
"grad_norm": 0.73046875,
"learning_rate": 0.00010852519120566475,
"loss": 0.7099,
"step": 720
},
{
"epoch": 2.64,
"grad_norm": 0.671875,
"learning_rate": 0.00010725520873327361,
"loss": 0.781,
"step": 725
},
{
"epoch": 2.66,
"grad_norm": 0.71875,
"learning_rate": 0.00010598404876976749,
"loss": 0.6719,
"step": 730
},
{
"epoch": 2.68,
"grad_norm": 0.7890625,
"learning_rate": 0.00010471191761928146,
"loss": 0.8674,
"step": 735
},
{
"epoch": 2.7,
"grad_norm": 0.73828125,
"learning_rate": 0.00010343902174357039,
"loss": 0.6813,
"step": 740
},
{
"epoch": 2.71,
"grad_norm": 0.703125,
"learning_rate": 0.00010216556772850094,
"loss": 0.795,
"step": 745
},
{
"epoch": 2.73,
"grad_norm": 0.7109375,
"learning_rate": 0.00010089176225052354,
"loss": 0.6741,
"step": 750
},
{
"epoch": 2.75,
"grad_norm": 0.71484375,
"learning_rate": 9.96178120431296e-05,
"loss": 0.6977,
"step": 755
},
{
"epoch": 2.77,
"grad_norm": 0.734375,
"learning_rate": 9.83439238632996e-05,
"loss": 0.6683,
"step": 760
},
{
"epoch": 2.79,
"grad_norm": 0.65625,
"learning_rate": 9.707030445794715e-05,
"loss": 0.6262,
"step": 765
},
{
"epoch": 2.81,
"grad_norm": 0.78125,
"learning_rate": 9.579716053036479e-05,
"loss": 0.6627,
"step": 770
},
{
"epoch": 2.82,
"grad_norm": 0.765625,
"learning_rate": 9.452469870667713e-05,
"loss": 0.7535,
"step": 775
},
{
"epoch": 2.84,
"grad_norm": 0.6640625,
"learning_rate": 9.325312550230581e-05,
"loss": 0.7468,
"step": 780
},
{
"epoch": 2.86,
"grad_norm": 0.59765625,
"learning_rate": 9.198264728845332e-05,
"loss": 0.6141,
"step": 785
},
{
"epoch": 2.88,
"grad_norm": 0.8125,
"learning_rate": 9.071347025860958e-05,
"loss": 0.6955,
"step": 790
},
{
"epoch": 2.9,
"grad_norm": 0.65234375,
"learning_rate": 8.944580039508748e-05,
"loss": 0.6839,
"step": 795
},
{
"epoch": 2.91,
"grad_norm": 0.71875,
"learning_rate": 8.81798434355931e-05,
"loss": 0.6004,
"step": 800
},
{
"epoch": 2.93,
"grad_norm": 0.69140625,
"learning_rate": 8.69158048398349e-05,
"loss": 0.7424,
"step": 805
},
{
"epoch": 2.95,
"grad_norm": 0.703125,
"learning_rate": 8.565388975617873e-05,
"loss": 0.6301,
"step": 810
},
{
"epoch": 2.97,
"grad_norm": 0.6875,
"learning_rate": 8.439430298835304e-05,
"loss": 0.74,
"step": 815
},
{
"epoch": 2.99,
"grad_norm": 0.7578125,
"learning_rate": 8.313724896220994e-05,
"loss": 0.645,
"step": 820
},
{
"epoch": 3.0,
"eval_loss": 1.5109344720840454,
"eval_runtime": 74.6963,
"eval_samples_per_second": 3.668,
"eval_steps_per_second": 0.469,
"step": 823
},
{
"epoch": 3.01,
"grad_norm": 0.59765625,
"learning_rate": 8.188293169254775e-05,
"loss": 0.5792,
"step": 825
},
{
"epoch": 3.02,
"grad_norm": 0.8359375,
"learning_rate": 8.063155475000037e-05,
"loss": 0.6678,
"step": 830
},
{
"epoch": 3.04,
"grad_norm": 0.71875,
"learning_rate": 7.938332122799834e-05,
"loss": 0.3506,
"step": 835
},
{
"epoch": 3.06,
"grad_norm": 0.828125,
"learning_rate": 7.813843370980787e-05,
"loss": 0.4464,
"step": 840
},
{
"epoch": 3.08,
"grad_norm": 0.6640625,
"learning_rate": 7.68970942356521e-05,
"loss": 0.4639,
"step": 845
},
{
"epoch": 3.1,
"grad_norm": 0.66015625,
"learning_rate": 7.565950426992096e-05,
"loss": 0.4815,
"step": 850
},
{
"epoch": 3.11,
"grad_norm": 0.6953125,
"learning_rate": 7.442586466847441e-05,
"loss": 0.4089,
"step": 855
},
{
"epoch": 3.13,
"grad_norm": 0.67578125,
"learning_rate": 7.319637564604412e-05,
"loss": 0.4647,
"step": 860
},
{
"epoch": 3.15,
"grad_norm": 0.734375,
"learning_rate": 7.197123674373942e-05,
"loss": 0.4837,
"step": 865
},
{
"epoch": 3.17,
"grad_norm": 0.71875,
"learning_rate": 7.075064679666294e-05,
"loss": 0.5208,
"step": 870
},
{
"epoch": 3.19,
"grad_norm": 0.703125,
"learning_rate": 6.953480390164e-05,
"loss": 0.4568,
"step": 875
},
{
"epoch": 3.21,
"grad_norm": 0.7890625,
"learning_rate": 6.832390538506868e-05,
"loss": 0.5692,
"step": 880
},
{
"epoch": 3.22,
"grad_norm": 0.72265625,
"learning_rate": 6.711814777089427e-05,
"loss": 0.4652,
"step": 885
},
{
"epoch": 3.24,
"grad_norm": 0.74609375,
"learning_rate": 6.591772674871434e-05,
"loss": 0.5665,
"step": 890
},
{
"epoch": 3.26,
"grad_norm": 0.69921875,
"learning_rate": 6.472283714201919e-05,
"loss": 0.5147,
"step": 895
},
{
"epoch": 3.28,
"grad_norm": 0.70703125,
"learning_rate": 6.353367287657262e-05,
"loss": 0.5033,
"step": 900
},
{
"epoch": 3.3,
"grad_norm": 0.75,
"learning_rate": 6.235042694893862e-05,
"loss": 0.5159,
"step": 905
},
{
"epoch": 3.32,
"grad_norm": 0.59765625,
"learning_rate": 6.117329139515894e-05,
"loss": 0.4865,
"step": 910
},
{
"epoch": 3.33,
"grad_norm": 0.6953125,
"learning_rate": 6.000245725958618e-05,
"loss": 0.4963,
"step": 915
},
{
"epoch": 3.35,
"grad_norm": 0.69140625,
"learning_rate": 5.883811456387821e-05,
"loss": 0.4766,
"step": 920
},
{
"epoch": 3.37,
"grad_norm": 0.671875,
"learning_rate": 5.7680452276158457e-05,
"loss": 0.4681,
"step": 925
},
{
"epoch": 3.39,
"grad_norm": 0.73046875,
"learning_rate": 5.6529658280346906e-05,
"loss": 0.4764,
"step": 930
},
{
"epoch": 3.41,
"grad_norm": 0.62109375,
"learning_rate": 5.5385919345667715e-05,
"loss": 0.4547,
"step": 935
},
{
"epoch": 3.42,
"grad_norm": 0.6953125,
"learning_rate": 5.424942109633692e-05,
"loss": 0.5382,
"step": 940
},
{
"epoch": 3.44,
"grad_norm": 0.85546875,
"learning_rate": 5.3120347981436444e-05,
"loss": 0.4404,
"step": 945
},
{
"epoch": 3.46,
"grad_norm": 0.703125,
"learning_rate": 5.199888324497907e-05,
"loss": 0.5514,
"step": 950
},
{
"epoch": 3.48,
"grad_norm": 0.60546875,
"learning_rate": 5.088520889616828e-05,
"loss": 0.5287,
"step": 955
},
{
"epoch": 3.5,
"grad_norm": 0.72265625,
"learning_rate": 4.9779505679859096e-05,
"loss": 0.4759,
"step": 960
},
{
"epoch": 3.52,
"grad_norm": 0.73828125,
"learning_rate": 4.8681953047223914e-05,
"loss": 0.5279,
"step": 965
},
{
"epoch": 3.53,
"grad_norm": 0.71484375,
"learning_rate": 4.759272912662822e-05,
"loss": 0.4184,
"step": 970
},
{
"epoch": 3.55,
"grad_norm": 0.57421875,
"learning_rate": 4.651201069472125e-05,
"loss": 0.4301,
"step": 975
},
{
"epoch": 3.57,
"grad_norm": 0.734375,
"learning_rate": 4.543997314774553e-05,
"loss": 0.4781,
"step": 980
},
{
"epoch": 3.59,
"grad_norm": 0.64453125,
"learning_rate": 4.437679047307116e-05,
"loss": 0.5563,
"step": 985
},
{
"epoch": 3.61,
"grad_norm": 0.6640625,
"learning_rate": 4.332263522095815e-05,
"loss": 0.4216,
"step": 990
},
{
"epoch": 3.62,
"grad_norm": 0.72265625,
"learning_rate": 4.227767847655205e-05,
"loss": 0.5113,
"step": 995
},
{
"epoch": 3.64,
"grad_norm": 0.6015625,
"learning_rate": 4.124208983211799e-05,
"loss": 0.5323,
"step": 1000
},
{
"epoch": 3.66,
"grad_norm": 0.71875,
"learning_rate": 4.021603735951608e-05,
"loss": 0.5079,
"step": 1005
},
{
"epoch": 3.68,
"grad_norm": 0.703125,
"learning_rate": 3.919968758292425e-05,
"loss": 0.4655,
"step": 1010
},
{
"epoch": 3.7,
"grad_norm": 0.71875,
"learning_rate": 3.8193205451811995e-05,
"loss": 0.5073,
"step": 1015
},
{
"epoch": 3.72,
"grad_norm": 0.609375,
"learning_rate": 3.719675431416972e-05,
"loss": 0.4882,
"step": 1020
},
{
"epoch": 3.73,
"grad_norm": 0.640625,
"learning_rate": 3.62104958899982e-05,
"loss": 0.4177,
"step": 1025
},
{
"epoch": 3.75,
"grad_norm": 0.671875,
"learning_rate": 3.5234590245061684e-05,
"loss": 0.5276,
"step": 1030
},
{
"epoch": 3.77,
"grad_norm": 0.69921875,
"learning_rate": 3.426919576491024e-05,
"loss": 0.357,
"step": 1035
},
{
"epoch": 3.79,
"grad_norm": 0.671875,
"learning_rate": 3.3314469129174364e-05,
"loss": 0.4146,
"step": 1040
},
{
"epoch": 3.81,
"grad_norm": 0.64453125,
"learning_rate": 3.237056528613624e-05,
"loss": 0.3781,
"step": 1045
},
{
"epoch": 3.83,
"grad_norm": 0.6328125,
"learning_rate": 3.1437637427582564e-05,
"loss": 0.4652,
"step": 1050
},
{
"epoch": 3.84,
"grad_norm": 0.67578125,
"learning_rate": 3.0515836963942056e-05,
"loss": 0.5685,
"step": 1055
},
{
"epoch": 3.86,
"grad_norm": 0.703125,
"learning_rate": 2.960531349971204e-05,
"loss": 0.5038,
"step": 1060
},
{
"epoch": 3.88,
"grad_norm": 0.62109375,
"learning_rate": 2.8706214809178355e-05,
"loss": 0.4237,
"step": 1065
},
{
"epoch": 3.9,
"grad_norm": 0.765625,
"learning_rate": 2.7818686812432136e-05,
"loss": 0.6172,
"step": 1070
},
{
"epoch": 3.92,
"grad_norm": 0.72265625,
"learning_rate": 2.6942873551687587e-05,
"loss": 0.5244,
"step": 1075
},
{
"epoch": 3.93,
"grad_norm": 0.62890625,
"learning_rate": 2.6078917167904527e-05,
"loss": 0.428,
"step": 1080
},
{
"epoch": 3.95,
"grad_norm": 0.640625,
"learning_rate": 2.5226957877719436e-05,
"loss": 0.3934,
"step": 1085
},
{
"epoch": 3.97,
"grad_norm": 0.64453125,
"learning_rate": 2.4387133950688912e-05,
"loss": 0.5376,
"step": 1090
},
{
"epoch": 3.99,
"grad_norm": 0.62109375,
"learning_rate": 2.3559581686849065e-05,
"loss": 0.4291,
"step": 1095
},
{
"epoch": 4.0,
"eval_loss": 1.4415178298950195,
"eval_runtime": 74.7485,
"eval_samples_per_second": 3.666,
"eval_steps_per_second": 0.468,
"step": 1098
},
{
"epoch": 4.01,
"grad_norm": 0.60546875,
"learning_rate": 2.2744435394594497e-05,
"loss": 0.4588,
"step": 1100
},
{
"epoch": 4.03,
"grad_norm": 0.484375,
"learning_rate": 2.1941827368880796e-05,
"loss": 0.3858,
"step": 1105
},
{
"epoch": 4.04,
"grad_norm": 0.6171875,
"learning_rate": 2.115188786975336e-05,
"loss": 0.4656,
"step": 1110
},
{
"epoch": 4.06,
"grad_norm": 0.50390625,
"learning_rate": 2.037474510120676e-05,
"loss": 0.3973,
"step": 1115
},
{
"epoch": 4.08,
"grad_norm": 0.6015625,
"learning_rate": 1.9610525190377983e-05,
"loss": 0.4184,
"step": 1120
},
{
"epoch": 4.1,
"grad_norm": 0.61328125,
"learning_rate": 1.885935216707617e-05,
"loss": 0.3977,
"step": 1125
},
{
"epoch": 4.12,
"grad_norm": 0.70703125,
"learning_rate": 1.8121347943653332e-05,
"loss": 0.4274,
"step": 1130
},
{
"epoch": 4.13,
"grad_norm": 0.5859375,
"learning_rate": 1.7396632295218317e-05,
"loss": 0.326,
"step": 1135
},
{
"epoch": 4.15,
"grad_norm": 0.625,
"learning_rate": 1.668532284019789e-05,
"loss": 0.4174,
"step": 1140
},
{
"epoch": 4.17,
"grad_norm": 0.67578125,
"learning_rate": 1.5987535021247667e-05,
"loss": 0.3984,
"step": 1145
},
{
"epoch": 4.19,
"grad_norm": 0.60546875,
"learning_rate": 1.530338208651627e-05,
"loss": 0.5323,
"step": 1150
},
{
"epoch": 4.21,
"grad_norm": 0.67578125,
"learning_rate": 1.463297507126562e-05,
"loss": 0.398,
"step": 1155
},
{
"epoch": 4.23,
"grad_norm": 0.703125,
"learning_rate": 1.3976422779850384e-05,
"loss": 0.3888,
"step": 1160
},
{
"epoch": 4.24,
"grad_norm": 0.56640625,
"learning_rate": 1.3333831768059301e-05,
"loss": 0.3988,
"step": 1165
},
{
"epoch": 4.26,
"grad_norm": 0.625,
"learning_rate": 1.270530632582191e-05,
"loss": 0.3928,
"step": 1170
},
{
"epoch": 4.28,
"grad_norm": 0.61328125,
"learning_rate": 1.2090948460282414e-05,
"loss": 0.4208,
"step": 1175
},
{
"epoch": 4.3,
"grad_norm": 0.58984375,
"learning_rate": 1.149085787924451e-05,
"loss": 0.4415,
"step": 1180
},
{
"epoch": 4.32,
"grad_norm": 0.7109375,
"learning_rate": 1.0905131974989158e-05,
"loss": 0.3831,
"step": 1185
},
{
"epoch": 4.34,
"grad_norm": 0.6640625,
"learning_rate": 1.0333865808468202e-05,
"loss": 0.4559,
"step": 1190
},
{
"epoch": 4.35,
"grad_norm": 0.5625,
"learning_rate": 9.777152093876396e-06,
"loss": 0.3927,
"step": 1195
},
{
"epoch": 4.37,
"grad_norm": 0.6171875,
"learning_rate": 9.235081183604233e-06,
"loss": 0.4036,
"step": 1200
},
{
"epoch": 4.39,
"grad_norm": 0.5234375,
"learning_rate": 8.70774105357407e-06,
"loss": 0.3153,
"step": 1205
},
{
"epoch": 4.41,
"grad_norm": 0.51953125,
"learning_rate": 8.195217288962175e-06,
"loss": 0.2749,
"step": 1210
},
{
"epoch": 4.43,
"grad_norm": 0.71484375,
"learning_rate": 7.697593070308317e-06,
"loss": 0.4395,
"step": 1215
},
{
"epoch": 4.44,
"grad_norm": 0.6796875,
"learning_rate": 7.214949160016115e-06,
"loss": 0.4126,
"step": 1220
},
{
"epoch": 4.46,
"grad_norm": 0.67578125,
"learning_rate": 6.747363889245606e-06,
"loss": 0.4111,
"step": 1225
},
{
"epoch": 4.48,
"grad_norm": 0.6328125,
"learning_rate": 6.294913145200321e-06,
"loss": 0.3986,
"step": 1230
},
{
"epoch": 4.5,
"grad_norm": 0.625,
"learning_rate": 5.857670358811096e-06,
"loss": 0.4057,
"step": 1235
},
{
"epoch": 4.52,
"grad_norm": 0.71875,
"learning_rate": 5.435706492818693e-06,
"loss": 0.478,
"step": 1240
},
{
"epoch": 4.54,
"grad_norm": 0.640625,
"learning_rate": 5.0290900302566335e-06,
"loss": 0.449,
"step": 1245
},
{
"epoch": 4.55,
"grad_norm": 0.70703125,
"learning_rate": 4.63788696333678e-06,
"loss": 0.4115,
"step": 1250
},
{
"epoch": 4.57,
"grad_norm": 0.52734375,
"learning_rate": 4.262160782739067e-06,
"loss": 0.3495,
"step": 1255
},
{
"epoch": 4.59,
"grad_norm": 0.640625,
"learning_rate": 3.901972467307191e-06,
"loss": 0.3864,
"step": 1260
},
{
"epoch": 4.61,
"grad_norm": 0.69921875,
"learning_rate": 3.5573804741519833e-06,
"loss": 0.4566,
"step": 1265
},
{
"epoch": 4.63,
"grad_norm": 0.51171875,
"learning_rate": 3.2284407291641174e-06,
"loss": 0.4479,
"step": 1270
},
{
"epoch": 4.64,
"grad_norm": 0.66015625,
"learning_rate": 2.9152066179374603e-06,
"loss": 0.4232,
"step": 1275
},
{
"epoch": 4.66,
"grad_norm": 0.6328125,
"learning_rate": 2.6177289771049274e-06,
"loss": 0.3441,
"step": 1280
},
{
"epoch": 4.68,
"grad_norm": 0.68359375,
"learning_rate": 2.33605608608779e-06,
"loss": 0.3774,
"step": 1285
},
{
"epoch": 4.7,
"grad_norm": 0.70703125,
"learning_rate": 2.0702336592602122e-06,
"loss": 0.4859,
"step": 1290
},
{
"epoch": 4.72,
"grad_norm": 0.625,
"learning_rate": 1.8203048385299181e-06,
"loss": 0.4231,
"step": 1295
},
{
"epoch": 4.74,
"grad_norm": 0.671875,
"learning_rate": 1.5863101863364483e-06,
"loss": 0.441,
"step": 1300
},
{
"epoch": 4.75,
"grad_norm": 0.734375,
"learning_rate": 1.3682876790681254e-06,
"loss": 0.3982,
"step": 1305
},
{
"epoch": 4.77,
"grad_norm": 0.71875,
"learning_rate": 1.1662727008984964e-06,
"loss": 0.4582,
"step": 1310
},
{
"epoch": 4.79,
"grad_norm": 0.5625,
"learning_rate": 9.802980380437588e-07,
"loss": 0.3836,
"step": 1315
},
{
"epoch": 4.81,
"grad_norm": 0.72265625,
"learning_rate": 8.103938734415839e-07,
"loss": 0.4008,
"step": 1320
},
{
"epoch": 4.83,
"grad_norm": 0.5859375,
"learning_rate": 6.565877818526245e-07,
"loss": 0.353,
"step": 1325
},
{
"epoch": 4.85,
"grad_norm": 0.5859375,
"learning_rate": 5.189047253851942e-07,
"loss": 0.4758,
"step": 1330
},
{
"epoch": 4.86,
"grad_norm": 0.609375,
"learning_rate": 3.973670494440307e-07,
"loss": 0.4848,
"step": 1335
},
{
"epoch": 4.88,
"grad_norm": 0.5703125,
"learning_rate": 2.919944791037632e-07,
"loss": 0.3638,
"step": 1340
},
{
"epoch": 4.9,
"grad_norm": 0.6796875,
"learning_rate": 2.0280411590754e-07,
"loss": 0.4863,
"step": 1345
},
{
"epoch": 4.92,
"grad_norm": 0.6484375,
"learning_rate": 1.2981043509159297e-07,
"loss": 0.4502,
"step": 1350
},
{
"epoch": 4.94,
"grad_norm": 0.7578125,
"learning_rate": 7.302528323589464e-08,
"loss": 0.4023,
"step": 1355
},
{
"epoch": 4.95,
"grad_norm": 0.65234375,
"learning_rate": 3.2457876341585033e-08,
"loss": 0.4602,
"step": 1360
},
{
"epoch": 4.97,
"grad_norm": 0.61328125,
"learning_rate": 8.114798335157048e-09,
"loss": 0.4061,
"step": 1365
},
{
"epoch": 4.99,
"grad_norm": 0.421875,
"learning_rate": 0.0,
"loss": 0.3392,
"step": 1370
},
{
"epoch": 4.99,
"eval_loss": 1.4403423070907593,
"eval_runtime": 74.7328,
"eval_samples_per_second": 3.666,
"eval_steps_per_second": 0.468,
"step": 1370
},
{
"epoch": 4.99,
"step": 1370,
"total_flos": 9.82065973089534e+17,
"train_loss": 1.004000898782354,
"train_runtime": 10982.2688,
"train_samples_per_second": 0.998,
"train_steps_per_second": 0.125
}
],
"logging_steps": 5,
"max_steps": 1370,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 100,
"total_flos": 9.82065973089534e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}