Files
Llama-Poro-2-8B-SFT/trainer_state.json
ModelHub XC d0090aca69 初始化项目,由ModelHub XC社区提供模型
Model: LumiOpen/Llama-Poro-2-8B-SFT
Source: Original Platform
2026-08-08 19:38:14 +08:00

1418 lines
34 KiB
JSON

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 500,
"global_step": 9690,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0002063983488132095,
"grad_norm": 21.791844475218177,
"learning_rate": 1.718213058419244e-08,
"loss": 1.2399,
"step": 1
},
{
"epoch": 0.010319917440660475,
"grad_norm": 2.758771436717397,
"learning_rate": 8.59106529209622e-07,
"loss": 1.0835,
"step": 50
},
{
"epoch": 0.02063983488132095,
"grad_norm": 1.6583275647715192,
"learning_rate": 1.718213058419244e-06,
"loss": 0.8281,
"step": 100
},
{
"epoch": 0.030959752321981424,
"grad_norm": 1.052445004470504,
"learning_rate": 2.577319587628866e-06,
"loss": 0.7456,
"step": 150
},
{
"epoch": 0.0412796697626419,
"grad_norm": 0.6644727438734609,
"learning_rate": 3.436426116838488e-06,
"loss": 0.6981,
"step": 200
},
{
"epoch": 0.05159958720330237,
"grad_norm": 0.6511384641685941,
"learning_rate": 4.2955326460481105e-06,
"loss": 0.6756,
"step": 250
},
{
"epoch": 0.06191950464396285,
"grad_norm": 0.6728510847115857,
"learning_rate": 4.995212256623046e-06,
"loss": 0.669,
"step": 300
},
{
"epoch": 0.07223942208462332,
"grad_norm": 0.6002063765996374,
"learning_rate": 4.968613682306629e-06,
"loss": 0.6488,
"step": 350
},
{
"epoch": 0.0825593395252838,
"grad_norm": 0.5807563716846417,
"learning_rate": 4.942015107990212e-06,
"loss": 0.641,
"step": 400
},
{
"epoch": 0.09287925696594428,
"grad_norm": 0.5651403042158113,
"learning_rate": 4.915416533673795e-06,
"loss": 0.6347,
"step": 450
},
{
"epoch": 0.10319917440660474,
"grad_norm": 0.5556237524713133,
"learning_rate": 4.888817959357379e-06,
"loss": 0.6292,
"step": 500
},
{
"epoch": 0.11351909184726522,
"grad_norm": 0.5537920539892396,
"learning_rate": 4.862219385040962e-06,
"loss": 0.6207,
"step": 550
},
{
"epoch": 0.1238390092879257,
"grad_norm": 0.6516923682179434,
"learning_rate": 4.835620810724546e-06,
"loss": 0.6221,
"step": 600
},
{
"epoch": 0.13415892672858618,
"grad_norm": 0.5993983563444428,
"learning_rate": 4.809022236408129e-06,
"loss": 0.6134,
"step": 650
},
{
"epoch": 0.14447884416924664,
"grad_norm": 0.5382978582953953,
"learning_rate": 4.782423662091712e-06,
"loss": 0.6142,
"step": 700
},
{
"epoch": 0.15479876160990713,
"grad_norm": 0.5360001817415995,
"learning_rate": 4.7558250877752955e-06,
"loss": 0.6064,
"step": 750
},
{
"epoch": 0.1651186790505676,
"grad_norm": 0.5196308091067839,
"learning_rate": 4.729226513458879e-06,
"loss": 0.5988,
"step": 800
},
{
"epoch": 0.17543859649122806,
"grad_norm": 0.5427506989599503,
"learning_rate": 4.702627939142462e-06,
"loss": 0.6025,
"step": 850
},
{
"epoch": 0.18575851393188855,
"grad_norm": 0.5521424095703439,
"learning_rate": 4.676029364826046e-06,
"loss": 0.597,
"step": 900
},
{
"epoch": 0.19607843137254902,
"grad_norm": 0.6345336850282564,
"learning_rate": 4.649430790509629e-06,
"loss": 0.5887,
"step": 950
},
{
"epoch": 0.20639834881320948,
"grad_norm": 0.5971258034612797,
"learning_rate": 4.6228322161932124e-06,
"loss": 0.5815,
"step": 1000
},
{
"epoch": 0.21671826625386997,
"grad_norm": 0.5438942448938008,
"learning_rate": 4.596233641876796e-06,
"loss": 0.5805,
"step": 1050
},
{
"epoch": 0.22703818369453044,
"grad_norm": 0.5334432938197203,
"learning_rate": 4.569635067560379e-06,
"loss": 0.5738,
"step": 1100
},
{
"epoch": 0.23735810113519093,
"grad_norm": 0.5347450534004635,
"learning_rate": 4.543036493243962e-06,
"loss": 0.5801,
"step": 1150
},
{
"epoch": 0.2476780185758514,
"grad_norm": 0.5873203374469886,
"learning_rate": 4.516437918927546e-06,
"loss": 0.5834,
"step": 1200
},
{
"epoch": 0.2579979360165119,
"grad_norm": 0.5240586732453324,
"learning_rate": 4.489839344611129e-06,
"loss": 0.5721,
"step": 1250
},
{
"epoch": 0.26831785345717235,
"grad_norm": 0.5575825059292604,
"learning_rate": 4.463240770294713e-06,
"loss": 0.5779,
"step": 1300
},
{
"epoch": 0.2786377708978328,
"grad_norm": 0.5561922023755285,
"learning_rate": 4.436642195978296e-06,
"loss": 0.5739,
"step": 1350
},
{
"epoch": 0.2889576883384933,
"grad_norm": 0.5214056422410267,
"learning_rate": 4.410043621661879e-06,
"loss": 0.5593,
"step": 1400
},
{
"epoch": 0.29927760577915374,
"grad_norm": 0.5509795794922944,
"learning_rate": 4.383445047345462e-06,
"loss": 0.563,
"step": 1450
},
{
"epoch": 0.30959752321981426,
"grad_norm": 0.5698665633020146,
"learning_rate": 4.356846473029046e-06,
"loss": 0.5627,
"step": 1500
},
{
"epoch": 0.31991744066047473,
"grad_norm": 0.5245577515841476,
"learning_rate": 4.3302478987126295e-06,
"loss": 0.5678,
"step": 1550
},
{
"epoch": 0.3302373581011352,
"grad_norm": 0.6001876556912392,
"learning_rate": 4.303649324396213e-06,
"loss": 0.5709,
"step": 1600
},
{
"epoch": 0.34055727554179566,
"grad_norm": 0.5333018384868324,
"learning_rate": 4.277050750079796e-06,
"loss": 0.5581,
"step": 1650
},
{
"epoch": 0.3508771929824561,
"grad_norm": 0.5133912524315866,
"learning_rate": 4.250452175763379e-06,
"loss": 0.5618,
"step": 1700
},
{
"epoch": 0.36119711042311664,
"grad_norm": 0.5077938795371925,
"learning_rate": 4.223853601446962e-06,
"loss": 0.5573,
"step": 1750
},
{
"epoch": 0.3715170278637771,
"grad_norm": 0.5463753688650598,
"learning_rate": 4.1972550271305465e-06,
"loss": 0.5539,
"step": 1800
},
{
"epoch": 0.38183694530443757,
"grad_norm": 0.5787935899358873,
"learning_rate": 4.17065645281413e-06,
"loss": 0.5483,
"step": 1850
},
{
"epoch": 0.39215686274509803,
"grad_norm": 0.5654912899054569,
"learning_rate": 4.144057878497713e-06,
"loss": 0.558,
"step": 1900
},
{
"epoch": 0.4024767801857585,
"grad_norm": 0.4996009494084185,
"learning_rate": 4.117459304181296e-06,
"loss": 0.5456,
"step": 1950
},
{
"epoch": 0.41279669762641896,
"grad_norm": 0.5667395400639976,
"learning_rate": 4.090860729864879e-06,
"loss": 0.5486,
"step": 2000
},
{
"epoch": 0.4231166150670795,
"grad_norm": 0.5188779450810256,
"learning_rate": 4.0642621555484625e-06,
"loss": 0.5464,
"step": 2050
},
{
"epoch": 0.43343653250773995,
"grad_norm": 0.5362183280913333,
"learning_rate": 4.037663581232047e-06,
"loss": 0.5521,
"step": 2100
},
{
"epoch": 0.4437564499484004,
"grad_norm": 0.5473955026723762,
"learning_rate": 4.01106500691563e-06,
"loss": 0.5445,
"step": 2150
},
{
"epoch": 0.4540763673890609,
"grad_norm": 0.5397425583264429,
"learning_rate": 3.984466432599213e-06,
"loss": 0.5585,
"step": 2200
},
{
"epoch": 0.46439628482972134,
"grad_norm": 0.5821347601023747,
"learning_rate": 3.957867858282796e-06,
"loss": 0.5513,
"step": 2250
},
{
"epoch": 0.47471620227038186,
"grad_norm": 0.5167859554911757,
"learning_rate": 3.9312692839663795e-06,
"loss": 0.5401,
"step": 2300
},
{
"epoch": 0.4850361197110423,
"grad_norm": 0.5353436940913985,
"learning_rate": 3.904670709649963e-06,
"loss": 0.5413,
"step": 2350
},
{
"epoch": 0.4953560371517028,
"grad_norm": 0.5450597051287799,
"learning_rate": 3.878072135333547e-06,
"loss": 0.5377,
"step": 2400
},
{
"epoch": 0.5056759545923633,
"grad_norm": 0.5133718526538825,
"learning_rate": 3.85147356101713e-06,
"loss": 0.5411,
"step": 2450
},
{
"epoch": 0.5159958720330238,
"grad_norm": 0.5194969527334257,
"learning_rate": 3.824874986700713e-06,
"loss": 0.5504,
"step": 2500
},
{
"epoch": 0.5263157894736842,
"grad_norm": 0.5244000570689482,
"learning_rate": 3.7982764123842964e-06,
"loss": 0.5373,
"step": 2550
},
{
"epoch": 0.5366357069143447,
"grad_norm": 0.4954661200186007,
"learning_rate": 3.7716778380678796e-06,
"loss": 0.5321,
"step": 2600
},
{
"epoch": 0.5469556243550051,
"grad_norm": 0.507806040946836,
"learning_rate": 3.7450792637514633e-06,
"loss": 0.5363,
"step": 2650
},
{
"epoch": 0.5572755417956656,
"grad_norm": 0.5217141184924292,
"learning_rate": 3.7184806894350465e-06,
"loss": 0.5374,
"step": 2700
},
{
"epoch": 0.5675954592363261,
"grad_norm": 1.0654133163473936,
"learning_rate": 3.69188211511863e-06,
"loss": 0.5387,
"step": 2750
},
{
"epoch": 0.5779153766769866,
"grad_norm": 0.5300385445449849,
"learning_rate": 3.665283540802213e-06,
"loss": 0.5328,
"step": 2800
},
{
"epoch": 0.5882352941176471,
"grad_norm": 0.5195789655014079,
"learning_rate": 3.6386849664857966e-06,
"loss": 0.5448,
"step": 2850
},
{
"epoch": 0.5985552115583075,
"grad_norm": 0.5400701284508418,
"learning_rate": 3.6120863921693798e-06,
"loss": 0.5225,
"step": 2900
},
{
"epoch": 0.608875128998968,
"grad_norm": 0.5192375803111897,
"learning_rate": 3.5854878178529634e-06,
"loss": 0.5354,
"step": 2950
},
{
"epoch": 0.6191950464396285,
"grad_norm": 0.5157981546008934,
"learning_rate": 3.558889243536547e-06,
"loss": 0.5234,
"step": 3000
},
{
"epoch": 0.6295149638802889,
"grad_norm": 0.5189888311937375,
"learning_rate": 3.5322906692201303e-06,
"loss": 0.5261,
"step": 3050
},
{
"epoch": 0.6398348813209495,
"grad_norm": 0.520981935459507,
"learning_rate": 3.505692094903713e-06,
"loss": 0.5277,
"step": 3100
},
{
"epoch": 0.6501547987616099,
"grad_norm": 0.5385057842452714,
"learning_rate": 3.4790935205872967e-06,
"loss": 0.5337,
"step": 3150
},
{
"epoch": 0.6604747162022704,
"grad_norm": 0.5090039339282304,
"learning_rate": 3.45249494627088e-06,
"loss": 0.5246,
"step": 3200
},
{
"epoch": 0.6707946336429309,
"grad_norm": 0.5309810767744108,
"learning_rate": 3.4258963719544636e-06,
"loss": 0.532,
"step": 3250
},
{
"epoch": 0.6811145510835913,
"grad_norm": 0.5117993267505484,
"learning_rate": 3.3992977976380472e-06,
"loss": 0.5329,
"step": 3300
},
{
"epoch": 0.6914344685242518,
"grad_norm": 0.542882178286457,
"learning_rate": 3.3726992233216304e-06,
"loss": 0.5262,
"step": 3350
},
{
"epoch": 0.7017543859649122,
"grad_norm": 0.5485246009385575,
"learning_rate": 3.3461006490052132e-06,
"loss": 0.5288,
"step": 3400
},
{
"epoch": 0.7120743034055728,
"grad_norm": 0.5245153157581836,
"learning_rate": 3.319502074688797e-06,
"loss": 0.5287,
"step": 3450
},
{
"epoch": 0.7223942208462333,
"grad_norm": 0.5408996443293894,
"learning_rate": 3.29290350037238e-06,
"loss": 0.5325,
"step": 3500
},
{
"epoch": 0.7327141382868937,
"grad_norm": 0.539947013464035,
"learning_rate": 3.2663049260559637e-06,
"loss": 0.523,
"step": 3550
},
{
"epoch": 0.7430340557275542,
"grad_norm": 0.5322841773244653,
"learning_rate": 3.2397063517395474e-06,
"loss": 0.5226,
"step": 3600
},
{
"epoch": 0.7533539731682146,
"grad_norm": 0.5150328597209529,
"learning_rate": 3.2131077774231306e-06,
"loss": 0.5182,
"step": 3650
},
{
"epoch": 0.7636738906088751,
"grad_norm": 0.5187564835457917,
"learning_rate": 3.1865092031067134e-06,
"loss": 0.5226,
"step": 3700
},
{
"epoch": 0.7739938080495357,
"grad_norm": 0.5376206106185862,
"learning_rate": 3.159910628790297e-06,
"loss": 0.5221,
"step": 3750
},
{
"epoch": 0.7843137254901961,
"grad_norm": 0.4886509664687733,
"learning_rate": 3.1333120544738802e-06,
"loss": 0.5167,
"step": 3800
},
{
"epoch": 0.7946336429308566,
"grad_norm": 0.5186897778877932,
"learning_rate": 3.106713480157464e-06,
"loss": 0.5215,
"step": 3850
},
{
"epoch": 0.804953560371517,
"grad_norm": 0.5662670913922451,
"learning_rate": 3.0801149058410475e-06,
"loss": 0.5156,
"step": 3900
},
{
"epoch": 0.8152734778121775,
"grad_norm": 0.5068699760497294,
"learning_rate": 3.0535163315246307e-06,
"loss": 0.5117,
"step": 3950
},
{
"epoch": 0.8255933952528379,
"grad_norm": 0.525153451451465,
"learning_rate": 3.0269177572082135e-06,
"loss": 0.5284,
"step": 4000
},
{
"epoch": 0.8359133126934984,
"grad_norm": 0.5415917077347273,
"learning_rate": 3.000319182891797e-06,
"loss": 0.5233,
"step": 4050
},
{
"epoch": 0.846233230134159,
"grad_norm": 0.5122355304442054,
"learning_rate": 2.9737206085753804e-06,
"loss": 0.52,
"step": 4100
},
{
"epoch": 0.8565531475748194,
"grad_norm": 0.5084493187015482,
"learning_rate": 2.947122034258964e-06,
"loss": 0.5122,
"step": 4150
},
{
"epoch": 0.8668730650154799,
"grad_norm": 0.5222318358455478,
"learning_rate": 2.9205234599425477e-06,
"loss": 0.5164,
"step": 4200
},
{
"epoch": 0.8771929824561403,
"grad_norm": 0.4952100113115859,
"learning_rate": 2.893924885626131e-06,
"loss": 0.5184,
"step": 4250
},
{
"epoch": 0.8875128998968008,
"grad_norm": 0.5129926015439157,
"learning_rate": 2.8673263113097137e-06,
"loss": 0.5217,
"step": 4300
},
{
"epoch": 0.8978328173374613,
"grad_norm": 0.5408976129024315,
"learning_rate": 2.8407277369932973e-06,
"loss": 0.5298,
"step": 4350
},
{
"epoch": 0.9081527347781218,
"grad_norm": 0.523128008449258,
"learning_rate": 2.8141291626768805e-06,
"loss": 0.5236,
"step": 4400
},
{
"epoch": 0.9184726522187823,
"grad_norm": 0.545708918135539,
"learning_rate": 2.787530588360464e-06,
"loss": 0.516,
"step": 4450
},
{
"epoch": 0.9287925696594427,
"grad_norm": 0.5643988024472154,
"learning_rate": 2.760932014044048e-06,
"loss": 0.5157,
"step": 4500
},
{
"epoch": 0.9391124871001032,
"grad_norm": 0.49490206521060065,
"learning_rate": 2.7343334397276306e-06,
"loss": 0.5143,
"step": 4550
},
{
"epoch": 0.9494324045407637,
"grad_norm": 0.530477676869888,
"learning_rate": 2.707734865411214e-06,
"loss": 0.5097,
"step": 4600
},
{
"epoch": 0.9597523219814241,
"grad_norm": 0.5419644910097084,
"learning_rate": 2.6811362910947975e-06,
"loss": 0.5047,
"step": 4650
},
{
"epoch": 0.9700722394220846,
"grad_norm": 0.563023481870365,
"learning_rate": 2.654537716778381e-06,
"loss": 0.51,
"step": 4700
},
{
"epoch": 0.9803921568627451,
"grad_norm": 0.5885551415729929,
"learning_rate": 2.6279391424619643e-06,
"loss": 0.5141,
"step": 4750
},
{
"epoch": 0.9907120743034056,
"grad_norm": 0.555238399523142,
"learning_rate": 2.601340568145548e-06,
"loss": 0.5144,
"step": 4800
},
{
"epoch": 1.0,
"eval_loss": 0.5763986110687256,
"eval_runtime": 22.6777,
"eval_samples_per_second": 23.724,
"eval_steps_per_second": 0.397,
"step": 4845
},
{
"epoch": 1.001031991744066,
"grad_norm": 0.5251455186752938,
"learning_rate": 2.5747419938291308e-06,
"loss": 0.5043,
"step": 4850
},
{
"epoch": 1.0113519091847265,
"grad_norm": 0.5102862437645022,
"learning_rate": 2.548143419512714e-06,
"loss": 0.4557,
"step": 4900
},
{
"epoch": 1.021671826625387,
"grad_norm": 0.5329812930765626,
"learning_rate": 2.5215448451962976e-06,
"loss": 0.4593,
"step": 4950
},
{
"epoch": 1.0319917440660475,
"grad_norm": 0.5310044546142733,
"learning_rate": 2.4949462708798813e-06,
"loss": 0.4601,
"step": 5000
},
{
"epoch": 1.0423116615067078,
"grad_norm": 0.5291259662919947,
"learning_rate": 2.4683476965634645e-06,
"loss": 0.4573,
"step": 5050
},
{
"epoch": 1.0526315789473684,
"grad_norm": 0.5758879453674484,
"learning_rate": 2.4417491222470477e-06,
"loss": 0.4545,
"step": 5100
},
{
"epoch": 1.0629514963880289,
"grad_norm": 0.5701692974214282,
"learning_rate": 2.4151505479306313e-06,
"loss": 0.4542,
"step": 5150
},
{
"epoch": 1.0732714138286894,
"grad_norm": 0.584779025941461,
"learning_rate": 2.3885519736142146e-06,
"loss": 0.4494,
"step": 5200
},
{
"epoch": 1.08359133126935,
"grad_norm": 0.5634926571145632,
"learning_rate": 2.3619533992977978e-06,
"loss": 0.4453,
"step": 5250
},
{
"epoch": 1.0939112487100102,
"grad_norm": 0.5287888561997275,
"learning_rate": 2.3353548249813814e-06,
"loss": 0.4494,
"step": 5300
},
{
"epoch": 1.1042311661506707,
"grad_norm": 0.566118676754179,
"learning_rate": 2.3087562506649646e-06,
"loss": 0.4472,
"step": 5350
},
{
"epoch": 1.1145510835913313,
"grad_norm": 0.5591675478155297,
"learning_rate": 2.282157676348548e-06,
"loss": 0.4487,
"step": 5400
},
{
"epoch": 1.1248710010319918,
"grad_norm": 0.5234427712256148,
"learning_rate": 2.2555591020321315e-06,
"loss": 0.4564,
"step": 5450
},
{
"epoch": 1.1351909184726523,
"grad_norm": 0.5341771123435606,
"learning_rate": 2.2289605277157147e-06,
"loss": 0.4546,
"step": 5500
},
{
"epoch": 1.1455108359133126,
"grad_norm": 0.5963942909826432,
"learning_rate": 2.202361953399298e-06,
"loss": 0.4523,
"step": 5550
},
{
"epoch": 1.1558307533539731,
"grad_norm": 0.5668397788463199,
"learning_rate": 2.1757633790828816e-06,
"loss": 0.456,
"step": 5600
},
{
"epoch": 1.1661506707946336,
"grad_norm": 0.5324369168992059,
"learning_rate": 2.1491648047664648e-06,
"loss": 0.458,
"step": 5650
},
{
"epoch": 1.1764705882352942,
"grad_norm": 0.5936240911653264,
"learning_rate": 2.122566230450048e-06,
"loss": 0.4406,
"step": 5700
},
{
"epoch": 1.1867905056759547,
"grad_norm": 0.5813588072096951,
"learning_rate": 2.0959676561336316e-06,
"loss": 0.4542,
"step": 5750
},
{
"epoch": 1.197110423116615,
"grad_norm": 0.585714648308596,
"learning_rate": 2.0693690818172144e-06,
"loss": 0.4435,
"step": 5800
},
{
"epoch": 1.2074303405572755,
"grad_norm": 0.5428768784930765,
"learning_rate": 2.042770507500798e-06,
"loss": 0.4468,
"step": 5850
},
{
"epoch": 1.217750257997936,
"grad_norm": 0.5368254232870538,
"learning_rate": 2.0161719331843817e-06,
"loss": 0.4538,
"step": 5900
},
{
"epoch": 1.2280701754385965,
"grad_norm": 0.548798947651045,
"learning_rate": 1.9895733588679645e-06,
"loss": 0.447,
"step": 5950
},
{
"epoch": 1.238390092879257,
"grad_norm": 0.612099006418541,
"learning_rate": 1.962974784551548e-06,
"loss": 0.4515,
"step": 6000
},
{
"epoch": 1.2487100103199174,
"grad_norm": 0.55639703440266,
"learning_rate": 1.936376210235132e-06,
"loss": 0.446,
"step": 6050
},
{
"epoch": 1.2590299277605779,
"grad_norm": 0.5807813468059144,
"learning_rate": 1.9097776359187146e-06,
"loss": 0.4518,
"step": 6100
},
{
"epoch": 1.2693498452012384,
"grad_norm": 0.5654058519092208,
"learning_rate": 1.8831790616022982e-06,
"loss": 0.4458,
"step": 6150
},
{
"epoch": 1.279669762641899,
"grad_norm": 0.5935622910169027,
"learning_rate": 1.8565804872858817e-06,
"loss": 0.4539,
"step": 6200
},
{
"epoch": 1.2899896800825594,
"grad_norm": 0.5320246374099179,
"learning_rate": 1.8299819129694649e-06,
"loss": 0.4523,
"step": 6250
},
{
"epoch": 1.3003095975232197,
"grad_norm": 0.5993250962476825,
"learning_rate": 1.8033833386530483e-06,
"loss": 0.4525,
"step": 6300
},
{
"epoch": 1.3106295149638802,
"grad_norm": 0.5706555061755686,
"learning_rate": 1.7767847643366317e-06,
"loss": 0.448,
"step": 6350
},
{
"epoch": 1.3209494324045408,
"grad_norm": 0.5604368666415582,
"learning_rate": 1.750186190020215e-06,
"loss": 0.4547,
"step": 6400
},
{
"epoch": 1.3312693498452013,
"grad_norm": 0.5368477750147255,
"learning_rate": 1.7235876157037984e-06,
"loss": 0.4522,
"step": 6450
},
{
"epoch": 1.3415892672858618,
"grad_norm": 0.608531892658329,
"learning_rate": 1.6969890413873818e-06,
"loss": 0.4529,
"step": 6500
},
{
"epoch": 1.351909184726522,
"grad_norm": 0.5524704535036694,
"learning_rate": 1.670390467070965e-06,
"loss": 0.4449,
"step": 6550
},
{
"epoch": 1.3622291021671826,
"grad_norm": 0.608361151275691,
"learning_rate": 1.6437918927545484e-06,
"loss": 0.449,
"step": 6600
},
{
"epoch": 1.3725490196078431,
"grad_norm": 0.5350140506451774,
"learning_rate": 1.6171933184381319e-06,
"loss": 0.4448,
"step": 6650
},
{
"epoch": 1.3828689370485037,
"grad_norm": 0.5657631888716299,
"learning_rate": 1.590594744121715e-06,
"loss": 0.4505,
"step": 6700
},
{
"epoch": 1.3931888544891642,
"grad_norm": 0.5638748961635498,
"learning_rate": 1.5639961698052985e-06,
"loss": 0.4516,
"step": 6750
},
{
"epoch": 1.4035087719298245,
"grad_norm": 0.5838904002390698,
"learning_rate": 1.537397595488882e-06,
"loss": 0.4434,
"step": 6800
},
{
"epoch": 1.413828689370485,
"grad_norm": 0.582688659499904,
"learning_rate": 1.5107990211724652e-06,
"loss": 0.4451,
"step": 6850
},
{
"epoch": 1.4241486068111455,
"grad_norm": 0.5666153072627447,
"learning_rate": 1.4842004468560486e-06,
"loss": 0.4513,
"step": 6900
},
{
"epoch": 1.434468524251806,
"grad_norm": 0.5830919184881627,
"learning_rate": 1.457601872539632e-06,
"loss": 0.4499,
"step": 6950
},
{
"epoch": 1.4447884416924666,
"grad_norm": 0.5392957954275294,
"learning_rate": 1.4310032982232152e-06,
"loss": 0.4526,
"step": 7000
},
{
"epoch": 1.4551083591331269,
"grad_norm": 0.5796751590874428,
"learning_rate": 1.4044047239067987e-06,
"loss": 0.4464,
"step": 7050
},
{
"epoch": 1.4654282765737874,
"grad_norm": 0.649115716424128,
"learning_rate": 1.377806149590382e-06,
"loss": 0.4448,
"step": 7100
},
{
"epoch": 1.475748194014448,
"grad_norm": 0.5437354389754013,
"learning_rate": 1.3512075752739653e-06,
"loss": 0.4488,
"step": 7150
},
{
"epoch": 1.4860681114551084,
"grad_norm": 0.575952575422212,
"learning_rate": 1.3246090009575488e-06,
"loss": 0.4445,
"step": 7200
},
{
"epoch": 1.496388028895769,
"grad_norm": 0.5689643540076003,
"learning_rate": 1.2980104266411322e-06,
"loss": 0.4427,
"step": 7250
},
{
"epoch": 1.5067079463364292,
"grad_norm": 0.5433467470767702,
"learning_rate": 1.2714118523247154e-06,
"loss": 0.4469,
"step": 7300
},
{
"epoch": 1.5170278637770898,
"grad_norm": 0.5785369307750878,
"learning_rate": 1.2448132780082988e-06,
"loss": 0.4418,
"step": 7350
},
{
"epoch": 1.5273477812177503,
"grad_norm": 0.5946065204485477,
"learning_rate": 1.2182147036918823e-06,
"loss": 0.45,
"step": 7400
},
{
"epoch": 1.5376676986584106,
"grad_norm": 0.5276268731765017,
"learning_rate": 1.1916161293754657e-06,
"loss": 0.4425,
"step": 7450
},
{
"epoch": 1.5479876160990713,
"grad_norm": 0.5456969115711289,
"learning_rate": 1.165017555059049e-06,
"loss": 0.4426,
"step": 7500
},
{
"epoch": 1.5583075335397316,
"grad_norm": 0.5626443461851722,
"learning_rate": 1.1384189807426323e-06,
"loss": 0.4457,
"step": 7550
},
{
"epoch": 1.5686274509803921,
"grad_norm": 0.5748093743725187,
"learning_rate": 1.1118204064262155e-06,
"loss": 0.4425,
"step": 7600
},
{
"epoch": 1.5789473684210527,
"grad_norm": 0.6398790495774637,
"learning_rate": 1.085221832109799e-06,
"loss": 0.4461,
"step": 7650
},
{
"epoch": 1.589267285861713,
"grad_norm": 0.5502896358409545,
"learning_rate": 1.0586232577933824e-06,
"loss": 0.4324,
"step": 7700
},
{
"epoch": 1.5995872033023737,
"grad_norm": 0.547970248426798,
"learning_rate": 1.0320246834769656e-06,
"loss": 0.4367,
"step": 7750
},
{
"epoch": 1.609907120743034,
"grad_norm": 0.5553997442333781,
"learning_rate": 1.005426109160549e-06,
"loss": 0.4518,
"step": 7800
},
{
"epoch": 1.6202270381836945,
"grad_norm": 0.5463386099370533,
"learning_rate": 9.788275348441325e-07,
"loss": 0.4494,
"step": 7850
},
{
"epoch": 1.630546955624355,
"grad_norm": 0.5924906419908685,
"learning_rate": 9.522289605277157e-07,
"loss": 0.441,
"step": 7900
},
{
"epoch": 1.6408668730650153,
"grad_norm": 0.5565206258809997,
"learning_rate": 9.256303862112992e-07,
"loss": 0.4417,
"step": 7950
},
{
"epoch": 1.651186790505676,
"grad_norm": 0.587012594186078,
"learning_rate": 8.990318118948826e-07,
"loss": 0.4442,
"step": 8000
},
{
"epoch": 1.6615067079463364,
"grad_norm": 0.5771632055191758,
"learning_rate": 8.724332375784658e-07,
"loss": 0.4456,
"step": 8050
},
{
"epoch": 1.671826625386997,
"grad_norm": 0.56799994088298,
"learning_rate": 8.458346632620493e-07,
"loss": 0.4486,
"step": 8100
},
{
"epoch": 1.6821465428276574,
"grad_norm": 0.569590541861768,
"learning_rate": 8.192360889456326e-07,
"loss": 0.449,
"step": 8150
},
{
"epoch": 1.6924664602683177,
"grad_norm": 0.5795380939294089,
"learning_rate": 7.926375146292158e-07,
"loss": 0.4438,
"step": 8200
},
{
"epoch": 1.7027863777089784,
"grad_norm": 0.5706726979508648,
"learning_rate": 7.660389403127994e-07,
"loss": 0.4443,
"step": 8250
},
{
"epoch": 1.7131062951496387,
"grad_norm": 0.5791014554749523,
"learning_rate": 7.394403659963827e-07,
"loss": 0.4481,
"step": 8300
},
{
"epoch": 1.7234262125902993,
"grad_norm": 0.561090943078198,
"learning_rate": 7.12841791679966e-07,
"loss": 0.4397,
"step": 8350
},
{
"epoch": 1.7337461300309598,
"grad_norm": 0.5863685907601032,
"learning_rate": 6.862432173635495e-07,
"loss": 0.4445,
"step": 8400
},
{
"epoch": 1.74406604747162,
"grad_norm": 0.5730575809697048,
"learning_rate": 6.596446430471328e-07,
"loss": 0.4451,
"step": 8450
},
{
"epoch": 1.7543859649122808,
"grad_norm": 0.5635343819200634,
"learning_rate": 6.330460687307161e-07,
"loss": 0.4382,
"step": 8500
},
{
"epoch": 1.7647058823529411,
"grad_norm": 0.5646651214526662,
"learning_rate": 6.064474944142994e-07,
"loss": 0.4403,
"step": 8550
},
{
"epoch": 1.7750257997936016,
"grad_norm": 0.5486977071368843,
"learning_rate": 5.798489200978829e-07,
"loss": 0.4355,
"step": 8600
},
{
"epoch": 1.7853457172342622,
"grad_norm": 0.5568438386062177,
"learning_rate": 5.532503457814662e-07,
"loss": 0.4477,
"step": 8650
},
{
"epoch": 1.7956656346749225,
"grad_norm": 0.5622396972214505,
"learning_rate": 5.266517714650495e-07,
"loss": 0.4377,
"step": 8700
},
{
"epoch": 1.8059855521155832,
"grad_norm": 0.552378184120207,
"learning_rate": 5.000531971486329e-07,
"loss": 0.4467,
"step": 8750
},
{
"epoch": 1.8163054695562435,
"grad_norm": 0.5708548011902037,
"learning_rate": 4.734546228322162e-07,
"loss": 0.439,
"step": 8800
},
{
"epoch": 1.826625386996904,
"grad_norm": 0.6282302095096289,
"learning_rate": 4.468560485157996e-07,
"loss": 0.4415,
"step": 8850
},
{
"epoch": 1.8369453044375645,
"grad_norm": 0.5526339563083117,
"learning_rate": 4.20257474199383e-07,
"loss": 0.4434,
"step": 8900
},
{
"epoch": 1.8472652218782248,
"grad_norm": 0.5627366388182174,
"learning_rate": 3.936588998829663e-07,
"loss": 0.4406,
"step": 8950
},
{
"epoch": 1.8575851393188856,
"grad_norm": 0.5675273603944594,
"learning_rate": 3.6706032556654965e-07,
"loss": 0.4364,
"step": 9000
},
{
"epoch": 1.8679050567595459,
"grad_norm": 0.6096156008730134,
"learning_rate": 3.40461751250133e-07,
"loss": 0.4421,
"step": 9050
},
{
"epoch": 1.8782249742002064,
"grad_norm": 0.655037790258665,
"learning_rate": 3.1386317693371635e-07,
"loss": 0.4456,
"step": 9100
},
{
"epoch": 1.888544891640867,
"grad_norm": 0.5522478423907845,
"learning_rate": 2.8726460261729973e-07,
"loss": 0.4399,
"step": 9150
},
{
"epoch": 1.8988648090815272,
"grad_norm": 0.5499766592623824,
"learning_rate": 2.606660283008831e-07,
"loss": 0.4447,
"step": 9200
},
{
"epoch": 1.909184726522188,
"grad_norm": 0.6047531326573264,
"learning_rate": 2.3406745398446645e-07,
"loss": 0.4483,
"step": 9250
},
{
"epoch": 1.9195046439628483,
"grad_norm": 0.5307110417886833,
"learning_rate": 2.074688796680498e-07,
"loss": 0.4474,
"step": 9300
},
{
"epoch": 1.9298245614035088,
"grad_norm": 0.5838188935676187,
"learning_rate": 1.8087030535163318e-07,
"loss": 0.4453,
"step": 9350
},
{
"epoch": 1.9401444788441693,
"grad_norm": 0.524912113097606,
"learning_rate": 1.5427173103521653e-07,
"loss": 0.4441,
"step": 9400
},
{
"epoch": 1.9504643962848296,
"grad_norm": 0.5892672773445452,
"learning_rate": 1.2767315671879988e-07,
"loss": 0.4368,
"step": 9450
},
{
"epoch": 1.9607843137254903,
"grad_norm": 0.5480626100352609,
"learning_rate": 1.0107458240238324e-07,
"loss": 0.4408,
"step": 9500
},
{
"epoch": 1.9711042311661506,
"grad_norm": 0.5668784462525065,
"learning_rate": 7.44760080859666e-08,
"loss": 0.4435,
"step": 9550
},
{
"epoch": 1.9814241486068112,
"grad_norm": 0.6110658269716644,
"learning_rate": 4.787743376954996e-08,
"loss": 0.4431,
"step": 9600
},
{
"epoch": 1.9917440660474717,
"grad_norm": 0.5446976742385811,
"learning_rate": 2.127885945313331e-08,
"loss": 0.4367,
"step": 9650
},
{
"epoch": 2.0,
"eval_loss": 0.5610442161560059,
"eval_runtime": 20.0148,
"eval_samples_per_second": 26.88,
"eval_steps_per_second": 0.45,
"step": 9690
},
{
"epoch": 2.0,
"step": 9690,
"total_flos": 4057777727078400.0,
"train_loss": 0.5052211482704485,
"train_runtime": 135745.8608,
"train_samples_per_second": 4.568,
"train_steps_per_second": 0.071
}
],
"logging_steps": 50,
"max_steps": 9690,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 4057777727078400.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}