grg commited on
Commit
74922c5
·
1 Parent(s): 63e90ba

Adding reasoning models and open generation for non-reasoning models

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. static/leaderboard.csv +27 -49
  2. static/leaderboard_old.csv +51 -0
  3. static/models_data/Cydonia-22B-v1.2/cfa_metrics.csv +9 -9
  4. static/models_data/Cydonia-22B-v1.2/matrix.svg +282 -286
  5. static/models_data/Cydonia-22B-v1.2/ranks.svg +0 -0
  6. static/models_data/Cydonia-22B-v1.2/structure.svg +0 -0
  7. static/models_data/Falcon3-10B-Instruct/matrix.svg +24 -24
  8. static/models_data/Falcon3-10B-Instruct/ranks.svg +0 -0
  9. static/models_data/Falcon3-10B-Instruct/structure.svg +0 -0
  10. static/models_data/Ministral-8B-Instruct-2410/matrix.svg +24 -24
  11. static/models_data/Ministral-8B-Instruct-2410/ranks.svg +0 -0
  12. static/models_data/Ministral-8B-Instruct-2410/structure.svg +0 -0
  13. static/models_data/Ministrations-8B-v1/cfa_metrics.csv +9 -9
  14. static/models_data/Ministrations-8B-v1/matrix.svg +342 -338
  15. static/models_data/Ministrations-8B-v1/ranks.svg +0 -0
  16. static/models_data/Ministrations-8B-v1/structure.svg +0 -0
  17. static/models_data/Mistral-7B-Instruct-v0.1/matrix.svg +24 -24
  18. static/models_data/Mistral-7B-Instruct-v0.1/ranks.svg +0 -0
  19. static/models_data/Mistral-7B-Instruct-v0.1/structure.svg +0 -0
  20. static/models_data/Mistral-7B-Instruct-v0.2/matrix.svg +24 -24
  21. static/models_data/Mistral-7B-Instruct-v0.2/ranks.svg +0 -0
  22. static/models_data/Mistral-7B-Instruct-v0.2/structure.svg +0 -0
  23. static/models_data/Mistral-7B-Instruct-v0.3/matrix.svg +24 -24
  24. static/models_data/Mistral-7B-Instruct-v0.3/ranks.svg +0 -0
  25. static/models_data/Mistral-7B-Instruct-v0.3/structure.svg +0 -0
  26. static/models_data/Mistral-Large-Instruct-2407/matrix.svg +24 -24
  27. static/models_data/Mistral-Large-Instruct-2407/ranks.svg +0 -0
  28. static/models_data/Mistral-Large-Instruct-2407/structure.svg +0 -0
  29. static/models_data/Mistral-Nemo-Instruct-2407/cfa_metrics.csv +9 -9
  30. static/models_data/Mistral-Nemo-Instruct-2407/matrix.svg +351 -337
  31. static/models_data/Mistral-Nemo-Instruct-2407/ranks.svg +0 -0
  32. static/models_data/Mistral-Nemo-Instruct-2407/structure.svg +0 -0
  33. static/models_data/Mistral-Small-Instruct-2409/matrix.svg +24 -24
  34. static/models_data/Mistral-Small-Instruct-2409/ranks.svg +0 -0
  35. static/models_data/Mistral-Small-Instruct-2409/structure.svg +0 -0
  36. static/models_data/Mixtral-8x22B-Instruct-v0.1/matrix.svg +24 -24
  37. static/models_data/Mixtral-8x22B-Instruct-v0.1/ranks.svg +0 -0
  38. static/models_data/Mixtral-8x22B-Instruct-v0.1/structure.svg +0 -0
  39. static/models_data/Mixtral-8x7B-Instruct-v0.1/matrix.svg +24 -24
  40. static/models_data/Mixtral-8x7B-Instruct-v0.1/ranks.svg +0 -0
  41. static/models_data/Mixtral-8x7B-Instruct-v0.1/structure.svg +0 -0
  42. static/models_data/Nautilus-70B-v0.1/matrix.svg +24 -24
  43. static/models_data/Nautilus-70B-v0.1/ranks.svg +0 -0
  44. static/models_data/Nautilus-70B-v0.1/structure.svg +0 -0
  45. static/models_data/Qwen2-72B-Instruct/matrix.svg +24 -24
  46. static/models_data/Qwen2-72B-Instruct/ranks.svg +0 -0
  47. static/models_data/Qwen2-72B-Instruct/structure.svg +0 -0
  48. static/models_data/Qwen2-7B-Instruct/matrix.svg +24 -24
  49. static/models_data/Qwen2-7B-Instruct/ranks.svg +0 -0
  50. static/models_data/Qwen2-7B-Instruct/structure.svg +0 -0
static/leaderboard.csv CHANGED
@@ -1,50 +1,28 @@
1
  Model,Ordinal (Win rate),Cardinal (Score),RO Stability,Stress,CFI,SRMR,RMSEA
2
- Dolphin3.0-Mistral-24B,0.43518518518518523,0.4109145498091707,0.30325113823532956,0.2448575989123979,0.42509722222222224,0.5472416666666666,0.5586861111111112
3
- dolphin-2.9.2-qwen2-72b,0.4288194444444444,0.419293239999323,0.3563924678118336,0.23967117347497235,0.37208888888888886,0.6048944444444444,0.5987472222222222
4
- Mistral-Large-Instruct-2411,0.7309027777777778,0.6622695500174903,0.6757331288014612,0.1797177817325889,0.5662249999999999,0.37859999999999994,0.4126833333333333
5
- Dracarys-72B-Instruct,0.5734953703703703,0.5493254808101572,0.6288139430434564,0.20048803680368033,0.34225833333333333,0.6266499999999999,0.6357722222222223
6
- Qwen2.5-14B-Instruct-1M,0.5807291666666666,0.54563004132866,0.4384933916905111,0.23996656946609862,0.606225,0.38121388888888896,0.37397777777777785
7
- phi-4,0.369212962962963,0.3704371621418185,0.1281161101838446,0.26194492137860814,0.5092972222222223,0.4079166666666667,0.38840277777777776
8
- Llama-3.1-Tulu-3-70B-DPO,0.36863425925925924,0.3596139752709628,0.29702121128617454,0.24487582075477365,0.2888666666666666,0.6782194444444445,0.6769444444444443
9
- Falcon3-10B-Instruct,0.2760416666666667,0.3089574335019922,0.11488806927790406,0.2675372535401236,0.40780833333333333,0.57435,0.5538138888888888
10
- Ministrations-8B-v1,0.5975115740740741,0.5629212654972728,0.5060368869776407,0.22452742393237857,0.5590166666666667,0.42490000000000006,0.4303666666666667
11
- Cydonia-22B-v1.2,0.7118055555555556,0.6554547368791501,0.6193557112491432,0.19415994996337282,0.6361166666666667,0.3338083333333334,0.3419333333333332
12
- Nautilus-70B-v0.1,0.7847222222222222,0.7074335111695115,0.6326443266767067,0.1814758840174009,0.7506972222222221,0.20927777777777778,0.23105277777777777
13
- Ministral-8B-Instruct-2410,0.5465856481481483,0.5196862499633303,0.4120903823316509,0.24045486295329566,0.5791333333333334,0.4199666666666668,0.4095833333333334
14
- llama-3.1-nemotron-70B-instruct,0.8726851851851852,0.751782963334874,0.7174031652092134,0.16209339860230643,0.7561694444444446,0.21189444444444439,0.23753055555555547
15
- hermes_3_llama_3.1_70b,0.5257523148148149,0.47953495168653426,0.2589918470169178,0.22941277457539844,0.6493000000000001,0.3096972222222222,0.3098777777777778
16
- hermes_3_llama_3.1_8b,0.41290509259259256,0.4117221943281449,0.16520527634373441,0.25311021741644446,0.5822944444444444,0.3526722222222223,0.3435555555555555
17
- gemma-2-2b-it,0.31886574074074076,0.3309858600428668,0.14746606707946294,0.263080165752695,0.40932500000000005,0.550475,0.5377472222222222
18
- gemma-2-9b-it,0.7051504629629629,0.6020857503693501,0.43782539244147833,0.20116278903333318,0.7543666666666667,0.23989444444444452,0.24792499999999995
19
- gemma-2-27b-it,0.5969328703703705,0.5270946699366518,0.3917304045417486,0.2058170364515589,0.5997861111111111,0.37121111111111116,0.37292222222222227
20
- phi-3-mini-128k-instruct,0.2934027777777778,0.32984992817164005,0.039299993295009855,0.281800547806919,0.5861361111111111,0.42524166666666674,0.3974944444444444
21
- phi-3-medium-128k-instruct,0.28964120370370366,0.30802986933853177,0.09692037989916814,0.2651981204439735,0.43025555555555556,0.5503277777777777,0.5381722222222222
22
- phi-3.5-mini-instruct,0.2152777777777778,0.2680653144619754,0.0361229186530762,0.28422749224983457,0.40715555555555555,0.5721138888888888,0.5507833333333333
23
- phi-3.5-MoE-instruct,0.35185185185185186,0.36128192067041315,0.10985291697837646,0.2739229692168671,0.5530944444444444,0.4248777777777778,0.40345
24
- Mistral-7B-Instruct-v0.1,0.19618055555555555,0.26609566354811315,0.027216280472015988,0.2829498135031582,0.38917777777777773,0.5561138888888888,0.530213888888889
25
- Mistral-7B-Instruct-v0.2,0.30497685185185186,0.32133832899241477,0.14417876497818388,0.265188983528973,0.3802722222222222,0.5727305555555555,0.5483611111111111
26
- Mistral-7B-Instruct-v0.3,0.22164351851851852,0.26572479479146804,0.07960539866974455,0.2742399030139009,0.31385,0.6241,0.6081333333333333
27
- Mixtral-8x7B-Instruct-v0.1,0.38512731481481477,0.3819009850972602,0.21473356319081474,0.2624402608740656,0.45275,0.5034666666666667,0.4905694444444444
28
- Mixtral-8x22B-Instruct-v0.1,0.3012152777777778,0.31529864972153404,0.1414001940345544,0.2548838005881672,0.3772361111111111,0.5810888888888889,0.5844750000000001
29
- command_r_plus,0.5413773148148148,0.4995356672762356,0.3429686514651868,0.23811982320641845,0.6033000000000001,0.3740166666666668,0.3667527777777777
30
- llama_3_8b_instruct,0.44675925925925924,0.4295836112681494,0.24527785038654715,0.245806400289881,0.5498222222222222,0.42656388888888896,0.42189444444444446
31
- llama_3_70b_instruct,0.7621527777777778,0.6839540364836003,0.607020698814379,0.18525883672204868,0.7210055555555557,0.2346083333333333,0.25758888888888887
32
- llama_3.1_8b_instruct,0.5324074074074073,0.4786874422110324,0.4295080949846363,0.22060228669473025,0.4305722222222223,0.5455027777777777,0.553
33
- llama_3.1_70b_instruct,0.8125,0.7172545013390067,0.691365862744007,0.1709718847084183,0.6979472222222223,0.2636777777777777,0.2907250000000001
34
- llama_3.3_70b_instruct,0.8394097222222222,0.7312232033647639,0.7140528865685338,0.16560091935602395,0.7000888888888889,0.2521833333333334,0.2887305555555556
35
- llama_3.1_405b_instruct_4bit,0.7364004629629629,0.6490864350383405,0.7232098126552619,0.1702199925365422,0.4875722222222223,0.4963444444444445,0.5211555555555556
36
- llama_3.2_1b_instruct,0.19502314814814817,0.2522036562381785,0.027192115495770382,0.29255310096654275,0.37450000000000006,0.5990222222222223,0.5740638888888888
37
- llama_3.2_3b_instruct,0.35127314814814814,0.3615804465210719,0.13450325180647235,0.27485276839064654,0.5017,0.44956666666666667,0.4226500000000001
38
- Qwen2-7B-Instruct,0.3619791666666667,0.36370005127542027,0.25108519506513916,0.25776537005719313,0.3560861111111111,0.6009722222222222,0.5920888888888889
39
- Qwen2-72B-Instruct,0.5732060185185185,0.5461212335522644,0.6465993243020925,0.20297742879025626,0.3045,0.6543138888888889,0.6646361111111111
40
- Qwen2.5-0.5B-Instruct,0.267650462962963,0.3005554090516966,0.002970456550606876,0.2928913315666324,0.5371250000000001,0.44709722222222226,0.404575
41
- Qwen2.5-7B-Instruct,0.5570023148148148,0.5163098181421168,0.333554494486959,0.2505866550331236,0.6473694444444444,0.30400277777777773,0.29651944444444434
42
- Qwen2.5-32B-Instruct,0.7335069444444444,0.656917654644944,0.6724190751477237,0.1806656189868978,0.5603222222222223,0.40237500000000004,0.41161666666666663
43
- Qwen2.5-72B-Instruct,0.8171296296296295,0.7104489147495714,0.6974116787371809,0.16176650806326276,0.6734583333333333,0.2993,0.3184472222222223
44
- gpt-3.5-turbo-0125,0.21267361111111108,0.28218378886707396,0.08240359836763214,0.28728574920060357,0.3873055555555555,0.599925,0.572238888888889
45
- gpt-4o-0513,0.6643518518518517,0.5989532974661671,0.5122163952167618,0.19201420113771173,0.6235416666666667,0.34458611111111115,0.3441805555555555
46
- gpt-4o-mini-2024-07-18,0.32754629629629634,0.3418785071827972,0.13575309046266867,0.2707065266105181,0.44214722222222214,0.5004583333333332,0.47896666666666665
47
- Mistral-Large-Instruct-2407,0.8388310185185185,0.7374229691535793,0.7644582301049158,0.16944638941325085,0.6510750000000001,0.31028611111111104,0.3297916666666667
48
- Mistral-Nemo-Instruct-2407,0.5494791666666666,0.5262426956484347,0.4414072595011627,0.21142636170606344,0.5161,0.42923055555555545,0.43113055555555546
49
- Mistral-Small-Instruct-2409,0.7644675925925926,0.6890378862258165,0.6416815833333804,0.1894343546381,0.6840472222222221,0.2601583333333335,0.2888777777777778
50
- dummy,0.17100694444444445,0.2291015386716794,-0.009004148398032956,0.2928877637010999,0.3755222222222222,0.622275,0.5915305555555557
 
1
  Model,Ordinal (Win rate),Cardinal (Score),RO Stability,Stress,CFI,SRMR,RMSEA
2
+ reka-flash-3,0.45940170940170943,0.4899723723251134,0.2885614355228928,0.21908343015733012,0.6356444444444445,0.32448055555555566,0.3265472222222222
3
+ DeepSeek-V3-0324,0.30982905982905984,0.3893755207780696,0.3004930917799752,0.23355431200645516,0.3651611111111111,0.5979083333333333,0.6206666666666667
4
+ DeepSeek-V3-0324_user,0.44070512820512825,0.49753828417140256,0.3776747268598861,0.22438707851276918,0.5427472222222222,0.4287916666666667,0.4199611111111111
5
+ gemma-3-27b-it,0.3894230769230769,0.5040524142149732,0.4494497801681616,0.22345480695286235,0.4511777777777778,0.4855138888888889,0.5075888888888889
6
+ Llama-3.3-70B-Instruct,0.7083333333333334,0.6737308210529425,0.689130950089758,0.17019112082437948,0.5869277777777778,0.38021666666666665,0.4031972222222221
7
+ Llama-3.1-70B-Instruct,0.7494658119658121,0.7109651955039333,0.6723010699062582,0.1770688267046745,0.7082666666666667,0.2531305555555555,0.27954999999999997
8
+ Llama-3.1-Nemotron-70B-Instruct,0.7291666666666665,0.7021372470167542,0.6749074485556226,0.16936515142178976,0.6728861111111111,0.2835055555555557,0.3025472222222223
9
+ Llama-3.1-8B-Instruct,0.5080128205128206,0.5681123119595507,0.46930842769163483,0.21202410397902305,0.5946111111111112,0.35378055555555554,0.3611416666666667
10
+ Llama-3.2-3B-Instruct,0.219017094017094,0.30725051735750397,0.06779772942677353,0.2865006677359514,0.48239166666666666,0.5190555555555556,0.49002222222222214
11
+ Llama-3.2-1B-Instruct,0.13354700854700854,0.23848817421801657,0.006872157585997037,0.29254990326652297,0.33736111111111117,0.5983222222222222,0.5660722222222222
12
+ Mistral-Large-Instruct-2411,0.5657051282051282,0.6032900501455456,0.5261735061045867,0.2009208454762036,0.5878305555555555,0.32947777777777776,0.33580555555555547
13
+ Mistral-Large-Instruct-2407,0.7964743589743589,0.7374229691535793,0.7644582301049158,0.16944638941325085,0.6510750000000001,0.31028611111111104,0.3297916666666667
14
+ Mistral-Nemo-Instruct-2407,0.31303418803418803,0.4060740463440762,0.21398622636798964,0.25768864583046025,0.5191694444444444,0.4492111111111111,0.41962222222222223
15
+ Mistral-Small-3.1-24B-Instruct-2503,0.6436965811965812,0.6471972798701351,0.5693346095644334,0.18619908818554043,0.6869249999999999,0.29532500000000006,0.305161111111111
16
+ QwQ-32B,0.8263888888888888,0.7719313363889678,0.8091295835194909,0.1769481985217758,0.6934333333333333,0.27074999999999994,0.30680277777777776
17
+ Qwen2.5-VL-72B-Instruct,0.8482905982905984,0.7728617287376658,0.7921085092426651,0.15380409595822364,0.7021583333333334,0.24572499999999997,0.2881694444444445
18
+ Qwen2.5-VL-7B-Instruct,0.3365384615384615,0.4265332391154229,0.22155167860085653,0.25685191259115403,0.5591361111111112,0.39983055555555547,0.37639444444444436
19
+ Qwen2.5-VL-3B-Instruct,0.12286324786324787,0.22915813420909414,0.021149465964301183,0.2913022346288958,0.3104166666666667,0.6484194444444444,0.6220277777777778
20
+ Qwen2.5-72B-Instruct,0.7110042735042735,0.6892855084702477,0.6634524803081552,0.17931196458174936,0.6639638888888889,0.30247222222222225,0.3217055555555556
21
+ Qwen2.5-32B-Instruct,0.5229700854700855,0.5797225465553821,0.6034765993068393,0.20283713589541164,0.46560833333333324,0.5226250000000001,0.5162722222222222
22
+ Qwen2.5-14B-Instruct-1M,0.3782051282051282,0.46648036370033313,0.36401726539161106,0.2400178186304459,0.4806611111111111,0.4841472222222223,0.48072222222222216
23
+ phi-4,0.20192307692307696,0.29450852038492736,0.06465994553954729,0.27174661907877007,0.41245277777777783,0.5301916666666666,0.5130861111111111
24
+ Dracarys2-72B-Instruct,0.7216880341880343,0.6991833993340636,0.6630452722509824,0.18543611655364134,0.6917972222222222,0.27243055555555573,0.2926444444444445
25
+ Nautilus-70B-v0.1,0.716880341880342,0.7074335111695115,0.6326443266767067,0.1814758840174009,0.7506972222222221,0.20927777777777778,0.23105277777777777
26
+ Cydonia-22B-v1.2,0.6052350427350427,0.6469610565596798,0.5951243715693844,0.20282847824454076,0.6382611111111111,0.3130722222222222,0.3271694444444445
27
+ Ministrations-8B-v1,0.3862179487179487,0.4584643004722592,0.32151643372620586,0.2588041089045269,0.5249611111111112,0.4474500000000001,0.4370583333333333
28
+ dummy,0.13835470085470086,0.2291015386716794,-0.009004148398032956,0.2928877637010999,0.3755222222222222,0.622275,0.5915305555555557
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
static/leaderboard_old.csv ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Model,Ordinal (Win rate),Cardinal (Score),RO Stability,Stress,CFI,SRMR,RMSEA
2
+ DeepSeek-V3-0324,0.3812358276643991,0.37334436936722804,0.057675582487030516,0.271102930565853,0.6624527777777778,0.34640277777777784,0.28889444444444445
3
+ Dolphin3.0-Mistral-24B,0.4407596371882086,0.4109145498091707,0.30325113823532956,0.2448575989123979,0.42509722222222224,0.5472416666666666,0.5586861111111112
4
+ dolphin-2.9.2-qwen2-72b,0.43339002267573695,0.419293239999323,0.3563924678118336,0.23967117347497235,0.37208888888888886,0.6048944444444444,0.5987472222222222
5
+ Mistral-Large-Instruct-2411,0.7315759637188209,0.6622695500174903,0.6757331288014612,0.1797177817325889,0.5662249999999999,0.37859999999999994,0.4126833333333333
6
+ Dracarys-72B-Instruct,0.5751133786848073,0.5493254808101572,0.6288139430434564,0.20048803680368033,0.34225833333333333,0.6266499999999999,0.6357722222222223
7
+ Qwen2.5-14B-Instruct-1M,0.5844671201814058,0.54563004132866,0.4384933916905111,0.23996656946609862,0.606225,0.38121388888888896,0.37397777777777785
8
+ phi-4,0.3724489795918367,0.3704371621418185,0.1281161101838446,0.26194492137860814,0.5092972222222223,0.4079166666666667,0.38840277777777776
9
+ Llama-3.1-Tulu-3-70B-DPO,0.37329931972789115,0.3596139752709628,0.29702121128617454,0.24487582075477365,0.2888666666666666,0.6782194444444445,0.6769444444444443
10
+ Falcon3-10B-Instruct,0.27919501133786845,0.3089574335019922,0.11488806927790406,0.2675372535401236,0.40780833333333333,0.57435,0.5538138888888888
11
+ Ministrations-8B-v1,0.6000566893424035,0.5629212654972728,0.5060368869776407,0.22452742393237857,0.5590166666666667,0.42490000000000006,0.4303666666666667
12
+ Cydonia-22B-v1.2,0.7137188208616779,0.6554547368791501,0.6193557112491432,0.19415994996337282,0.6361166666666667,0.3338083333333334,0.3419333333333332
13
+ Nautilus-70B-v0.1,0.7854308390022675,0.7074335111695115,0.6326443266767067,0.1814758840174009,0.7506972222222221,0.20927777777777778,0.23105277777777777
14
+ Ministral-8B-Instruct-2410,0.5507369614512471,0.5196862499633303,0.4120903823316509,0.24045486295329566,0.5791333333333334,0.4199666666666668,0.4095833333333334
15
+ llama-3.1-nemotron-70B-instruct,0.8735827664399093,0.751782963334874,0.7174031652092134,0.16209339860230643,0.7561694444444446,0.21189444444444439,0.23753055555555547
16
+ hermes_3_llama_3.1_70b,0.5291950113378685,0.47953495168653426,0.2589918470169178,0.22941277457539844,0.6493000000000001,0.3096972222222222,0.3098777777777778
17
+ hermes_3_llama_3.1_8b,0.4172335600907029,0.4117221943281449,0.16520527634373441,0.25311021741644446,0.5822944444444444,0.3526722222222223,0.3435555555555555
18
+ gemma-2-2b-it,0.3222789115646259,0.3309858600428668,0.14746606707946294,0.263080165752695,0.40932500000000005,0.550475,0.5377472222222222
19
+ gemma-2-9b-it,0.7083333333333334,0.6020857503693501,0.43782539244147833,0.20116278903333318,0.7543666666666667,0.23989444444444452,0.24792499999999995
20
+ gemma-2-27b-it,0.6006235827664399,0.5270946699366518,0.3917304045417486,0.2058170364515589,0.5997861111111111,0.37121111111111116,0.37292222222222227
21
+ phi-3-mini-128k-instruct,0.29563492063492064,0.32984992817164005,0.039299993295009855,0.281800547806919,0.5861361111111111,0.42524166666666674,0.3974944444444444
22
+ phi-3-medium-128k-instruct,0.2905328798185941,0.30802986933853177,0.09692037989916814,0.2651981204439735,0.43025555555555556,0.5503277777777777,0.5381722222222222
23
+ phi-3.5-mini-instruct,0.21683673469387757,0.2680653144619754,0.0361229186530762,0.28422749224983457,0.40715555555555555,0.5721138888888888,0.5507833333333333
24
+ phi-3.5-MoE-instruct,0.35374149659863946,0.36128192067041315,0.10985291697837646,0.2739229692168671,0.5530944444444444,0.4248777777777778,0.40345
25
+ Mistral-7B-Instruct-v0.1,0.19784580498866214,0.26609566354811315,0.027216280472015988,0.2829498135031582,0.38917777777777773,0.5561138888888888,0.530213888888889
26
+ Mistral-7B-Instruct-v0.2,0.30895691609977327,0.32133832899241477,0.14417876497818388,0.265188983528973,0.3802722222222222,0.5727305555555555,0.5483611111111111
27
+ Mistral-7B-Instruct-v0.3,0.22392290249433108,0.26572479479146804,0.07960539866974455,0.2742399030139009,0.31385,0.6241,0.6081333333333333
28
+ Mixtral-8x7B-Instruct-v0.1,0.38945578231292516,0.3819009850972602,0.21473356319081474,0.2624402608740656,0.45275,0.5034666666666667,0.4905694444444444
29
+ Mixtral-8x22B-Instruct-v0.1,0.3055555555555556,0.31529864972153404,0.1414001940345544,0.2548838005881672,0.3772361111111111,0.5810888888888889,0.5844750000000001
30
+ command_r_plus,0.5456349206349207,0.4995356672762356,0.3429686514651868,0.23811982320641845,0.6033000000000001,0.3740166666666668,0.3667527777777777
31
+ llama_3_8b_instruct,0.4503968253968254,0.4295836112681494,0.24527785038654715,0.245806400289881,0.5498222222222222,0.42656388888888896,0.42189444444444446
32
+ llama_3_70b_instruct,0.7638888888888888,0.6839540364836003,0.607020698814379,0.18525883672204868,0.7210055555555557,0.2346083333333333,0.25758888888888887
33
+ llama_3.1_8b_instruct,0.5362811791383221,0.4786874422110324,0.4295080949846363,0.22060228669473025,0.4305722222222223,0.5455027777777777,0.553
34
+ llama_3.1_70b_instruct,0.8120748299319728,0.7172545013390067,0.691365862744007,0.1709718847084183,0.6979472222222223,0.2636777777777777,0.2907250000000001
35
+ llama_3.3_70b_instruct,0.8395691609977324,0.7312232033647639,0.7140528865685338,0.16560091935602395,0.7000888888888889,0.2521833333333334,0.2887305555555556
36
+ llama_3.1_405b_instruct_4bit,0.7358276643990931,0.6490864350383405,0.7232098126552619,0.1702199925365422,0.4875722222222223,0.4963444444444445,0.5211555555555556
37
+ llama_3.2_1b_instruct,0.1961451247165533,0.2522036562381785,0.027192115495770382,0.29255310096654275,0.37450000000000006,0.5990222222222223,0.5740638888888888
38
+ llama_3.2_3b_instruct,0.3543083900226757,0.3615804465210719,0.13450325180647235,0.27485276839064654,0.5017,0.44956666666666667,0.4226500000000001
39
+ Qwen2-7B-Instruct,0.36621315192743764,0.36370005127542027,0.25108519506513916,0.25776537005719313,0.3560861111111111,0.6009722222222222,0.5920888888888889
40
+ Qwen2-72B-Instruct,0.5748299319727891,0.5461212335522644,0.6465993243020925,0.20297742879025626,0.3045,0.6543138888888889,0.6646361111111111
41
+ Qwen2.5-0.5B-Instruct,0.26842403628117917,0.3005554090516966,0.002970456550606876,0.2928913315666324,0.5371250000000001,0.44709722222222226,0.404575
42
+ Qwen2.5-7B-Instruct,0.5620748299319729,0.5163098181421168,0.333554494486959,0.2505866550331236,0.6473694444444444,0.30400277777777773,0.29651944444444434
43
+ Qwen2.5-32B-Instruct,0.7349773242630386,0.656917654644944,0.6724190751477237,0.1806656189868978,0.5603222222222223,0.40237500000000004,0.41161666666666663
44
+ Qwen2.5-72B-Instruct,0.8166099773242631,0.7104489147495714,0.6974116787371809,0.16176650806326276,0.6734583333333333,0.2993,0.3184472222222223
45
+ gpt-3.5-turbo-0125,0.21456916099773238,0.28218378886707396,0.08240359836763214,0.28728574920060357,0.3873055555555555,0.599925,0.572238888888889
46
+ gpt-4o-0513,0.6669501133786847,0.5989532974661671,0.5122163952167618,0.19201420113771173,0.6235416666666667,0.34458611111111115,0.3441805555555555
47
+ gpt-4o-mini-2024-07-18,0.3310657596371882,0.3418785071827972,0.13575309046266867,0.2707065266105181,0.44214722222222214,0.5004583333333332,0.47896666666666665
48
+ Mistral-Large-Instruct-2407,0.8373015873015872,0.7374229691535793,0.7644582301049158,0.16944638941325085,0.6510750000000001,0.31028611111111104,0.3297916666666667
49
+ Mistral-Nemo-Instruct-2407,0.5521541950113379,0.5262426956484347,0.4414072595011627,0.21142636170606344,0.5161,0.42923055555555545,0.43113055555555546
50
+ Mistral-Small-Instruct-2409,0.7655895691609977,0.6890378862258165,0.6416815833333804,0.1894343546381,0.6840472222222221,0.2601583333333335,0.2888777777777778
51
+ dummy,0.17006802721088435,0.2291015386716794,-0.009004148398032956,0.2928877637010999,0.3755222222222222,0.622275,0.5915305555555557
static/models_data/Cydonia-22B-v1.2/cfa_metrics.csv CHANGED
@@ -1,10 +1,10 @@
1
  Context chunk,CFI,TLI,SRMR,RMSEA
2
- chunk_0,0.8068,0.739625,0.1121,0.14629999999999999
3
- chunk_1,0.6067499999999999,0.563925,0.345375,0.35409999999999997
4
- chunk_2,0.403675,0.376425,0.56435,0.55855
5
- chunk_3,0.6270249999999999,0.5892,0.32565,0.34327500000000005
6
- chunk_4,0.8815999999999999,0.840025,0.10122500000000001,0.117225
7
- chunk_chess_0,0.6381,0.604975,0.34,0.35275
8
- chunk_grammar_1,0.865825,0.8127499999999999,0.104025,0.11407499999999998
9
- chunk_no_conv,0.697075,0.6856500000000001,0.321125,0.30562500000000004
10
- chunk_svs_no_conv,0.1982,0.180525,0.7904249999999999,0.7855
 
1
  Context chunk,CFI,TLI,SRMR,RMSEA
2
+ chunk_0,0.38787499999999997,0.349075,0.55785,0.5835250000000001
3
+ chunk_1,0.895525,0.8709,0.098275,0.08115
4
+ chunk_2,0.8167249999999999,0.7593500000000001,0.109725,0.13575
5
+ chunk_3,0.820225,0.7622500000000001,0.12269999999999999,0.13224999999999998
6
+ chunk_4,0.6108,0.551775,0.34125,0.35359999999999997
7
+ chunk_chess_0,0.6475500000000001,0.6178,0.329075,0.345675
8
+ chunk_grammar_1,0.61065,0.572025,0.35305,0.35135
9
+ chunk_no_conv,0.834975,0.77905,0.11562499999999999,0.1638
10
+ chunk_svs_no_conv,0.120025,0.09015,0.7901,0.797425
static/models_data/Cydonia-22B-v1.2/matrix.svg CHANGED
static/models_data/Cydonia-22B-v1.2/ranks.svg CHANGED
static/models_data/Cydonia-22B-v1.2/structure.svg CHANGED
static/models_data/Falcon3-10B-Instruct/matrix.svg CHANGED
static/models_data/Falcon3-10B-Instruct/ranks.svg CHANGED
static/models_data/Falcon3-10B-Instruct/structure.svg CHANGED
static/models_data/Ministral-8B-Instruct-2410/matrix.svg CHANGED
static/models_data/Ministral-8B-Instruct-2410/ranks.svg CHANGED
static/models_data/Ministral-8B-Instruct-2410/structure.svg CHANGED
static/models_data/Ministrations-8B-v1/cfa_metrics.csv CHANGED
@@ -1,10 +1,10 @@
1
  Context chunk,CFI,TLI,SRMR,RMSEA
2
- chunk_0,0.6514249999999999,0.6182000000000001,0.339675,0.3367
3
- chunk_1,0.48062499999999997,0.5191,0.5442750000000001,0.5115000000000001
4
- chunk_2,0.418025,0.412975,0.5533750000000001,0.5383249999999999
5
- chunk_3,0.44335,0.41459999999999997,0.5395,0.56965
6
- chunk_4,0.8795,0.8342,0.10127499999999999,0.12219999999999999
7
- chunk_chess_0,0.690575,0.667175,0.3052,0.33315
8
- chunk_grammar_1,0.419425,0.394525,0.556825,0.550925
9
- chunk_no_conv,0.4447,0.42155,0.552075,0.559525
10
- chunk_svs_no_conv,0.603525,0.567325,0.3319,0.35132500000000005
 
1
  Context chunk,CFI,TLI,SRMR,RMSEA
2
+ chunk_0,0.46715,0.49015,0.5363,0.529325
3
+ chunk_1,0.202875,0.1907,0.774875,0.7669250000000001
4
+ chunk_2,0.481425,0.6037250000000001,0.5422750000000001,0.51175
5
+ chunk_3,0.6098250000000001,0.5543,0.32732500000000003,0.314075
6
+ chunk_4,0.428925,0.401625,0.542175,0.547875
7
+ chunk_chess_0,0.650825,0.6739999999999999,0.32,0.29962500000000003
8
+ chunk_grammar_1,0.583025,0.525525,0.326875,0.321575
9
+ chunk_no_conv,0.43115,0.405575,0.55255,0.549325
10
+ chunk_svs_no_conv,0.8694500000000001,0.83165,0.104675,0.09305
static/models_data/Ministrations-8B-v1/matrix.svg CHANGED
static/models_data/Ministrations-8B-v1/ranks.svg CHANGED
static/models_data/Ministrations-8B-v1/structure.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.1/matrix.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.1/ranks.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.1/structure.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.2/matrix.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.2/ranks.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.2/structure.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.3/matrix.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.3/ranks.svg CHANGED
static/models_data/Mistral-7B-Instruct-v0.3/structure.svg CHANGED
static/models_data/Mistral-Large-Instruct-2407/matrix.svg CHANGED
static/models_data/Mistral-Large-Instruct-2407/ranks.svg CHANGED
static/models_data/Mistral-Large-Instruct-2407/structure.svg CHANGED
static/models_data/Mistral-Nemo-Instruct-2407/cfa_metrics.csv CHANGED
@@ -1,10 +1,10 @@
1
  Context chunk,CFI,TLI,SRMR,RMSEA
2
- chunk_0,0.415225,0.38944999999999996,0.55465,0.5547249999999999
3
- chunk_1,0.36997500000000005,0.33294999999999997,0.5631999999999999,0.55495
4
- chunk_2,0.907125,0.8771,0.08945,0.079925
5
- chunk_3,0.6317,0.606625,0.3249,0.32384999999999997
6
- chunk_4,0.383275,0.34015,0.5604500000000001,0.5612
7
- chunk_chess_0,0.4017,0.3686,0.54955,0.568475
8
- chunk_grammar_1,0.364375,0.357275,0.550575,0.524525
9
- chunk_no_conv,0.440975,0.409975,0.5550499999999999,0.58055
10
- chunk_svs_no_conv,0.73055,0.6734749999999999,0.11524999999999999,0.131975
 
1
  Context chunk,CFI,TLI,SRMR,RMSEA
2
+ chunk_0,0.25,1.392025,0.772925,0.75
3
+ chunk_1,0.9480999999999999,1.2264249999999999,0.08925,0.017525
4
+ chunk_2,0.2768,0.18375,0.5579999999999999,0.5459750000000001
5
+ chunk_3,0.2818,0.21995,0.558425,0.5501
6
+ chunk_4,0.923175,1.5393999999999999,0.09607500000000001,0.03105
7
+ chunk_chess_0,0.356975,0.2954,0.548775,0.539325
8
+ chunk_grammar_1,0.485425,1.534625,0.545725,0.508975
9
+ chunk_no_conv,0.477075,0.471075,0.54905,0.51215
10
+ chunk_svs_no_conv,0.673175,0.64615,0.324675,0.3215
static/models_data/Mistral-Nemo-Instruct-2407/matrix.svg CHANGED
static/models_data/Mistral-Nemo-Instruct-2407/ranks.svg CHANGED
static/models_data/Mistral-Nemo-Instruct-2407/structure.svg CHANGED
static/models_data/Mistral-Small-Instruct-2409/matrix.svg CHANGED
static/models_data/Mistral-Small-Instruct-2409/ranks.svg CHANGED
static/models_data/Mistral-Small-Instruct-2409/structure.svg CHANGED
static/models_data/Mixtral-8x22B-Instruct-v0.1/matrix.svg CHANGED
static/models_data/Mixtral-8x22B-Instruct-v0.1/ranks.svg CHANGED
static/models_data/Mixtral-8x22B-Instruct-v0.1/structure.svg CHANGED
static/models_data/Mixtral-8x7B-Instruct-v0.1/matrix.svg CHANGED
static/models_data/Mixtral-8x7B-Instruct-v0.1/ranks.svg CHANGED
static/models_data/Mixtral-8x7B-Instruct-v0.1/structure.svg CHANGED
static/models_data/Nautilus-70B-v0.1/matrix.svg CHANGED
static/models_data/Nautilus-70B-v0.1/ranks.svg CHANGED
static/models_data/Nautilus-70B-v0.1/structure.svg CHANGED
static/models_data/Qwen2-72B-Instruct/matrix.svg CHANGED
static/models_data/Qwen2-72B-Instruct/ranks.svg CHANGED
static/models_data/Qwen2-72B-Instruct/structure.svg CHANGED
static/models_data/Qwen2-7B-Instruct/matrix.svg CHANGED
static/models_data/Qwen2-7B-Instruct/ranks.svg CHANGED
static/models_data/Qwen2-7B-Instruct/structure.svg CHANGED