Hugging Face's logo

Local-Novel-LLM-project
/
Yosegi-2

Text Generation
Transformers
Safetensors
mistral
mergekit
Merge
text-generation-inference
Inference Endpoints
Model card Files Community

Yosegi-2

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the TIES merge method using ./Yosegi-0601 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

base_model: ./Yosegi-0601
dtype: bfloat16
merge_method: ties
parameters:
  int8_mask: 1.0
  normalize: 0.0
slices:
- sources:
  - layer_range: [0, 2]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.9895701336232673
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.5057237984975562
      - filter: mlp
        value: 0.36247235528151495
      - value: 0.0076810835717692014
  - layer_range: [0, 2]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 0.8239779346577963
      weight:
      - filter: self_attn
        value: 0.27499287617186813
      - filter: mlp
        value: 0.10579959634086915
      - value: 0.14502290477239704
  - layer_range: [0, 2]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.9654867628269999
      - value: 0.9584724004158125
      weight:
      - filter: self_attn
        value: 0.059719404899177556
      - filter: mlp
        value: 0.1299695859327612
      - value: 0.18821871354400985
  - layer_range: [0, 2]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 0.9322987005873715
      - filter: mlp
        value: 0.8119693860979944
      - value: 0.7800996941956229
      weight:
      - filter: self_attn
        value: 0.14989333734000856
      - filter: mlp
        value: 0.20525182711733667
      - value: 0.0743540962371737
  - layer_range: [0, 2]
    model: ./Yosegi-0601
- sources:
  - layer_range: [2, 4]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 0.6361163471256639
      - filter: mlp
        value: 0.9983948965135213
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.2433049522842103
      - filter: mlp
        value: 0.11537153133586801
      - value: 0.11236945502439658
  - layer_range: [2, 4]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.13087986863180992
      - filter: mlp
        value: 0.05060452788200992
      - value: 0.029882383396623725
  - layer_range: [2, 4]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.9938109261305853
      - filter: mlp
        value: 0.709432587913349
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.15343343058938377
      - filter: mlp
        value: 0.4105917936868785
      - value: 0.6078632204623161
  - layer_range: [2, 4]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 0.9634269234020544
      weight:
      - filter: self_attn
        value: 0.03750763360681478
      - filter: mlp
        value: 0.29089122858987404
      - value: 0.3408085857388722
  - layer_range: [2, 4]
    model: ./Yosegi-0601
- sources:
  - layer_range: [4, 6]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 0.8057109303418598
      - filter: mlp
        value: 0.9954520808628292
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.02598285706585618
      - filter: mlp
        value: 0.06661629726622949
      - value: 0.1285191000066376
  - layer_range: [4, 6]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 0.9112825916608848
      - filter: mlp
        value: 0.9322557507910056
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.18823564379986454
      - filter: mlp
        value: 0.4552822441636322
      - value: 0.5120525709221785
  - layer_range: [4, 6]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.9869122169774399
      - filter: mlp
        value: 1.0
      - value: 0.9751291459565757
      weight:
      - filter: self_attn
        value: 0.00493134813843582
      - filter: mlp
        value: 0.3008979965262413
      - value: 0.2528466849993097
  - layer_range: [4, 6]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.8956512783019246
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.4197408619693966
      - filter: mlp
        value: 0.1448902874618845
      - value: 0.5196932662212128
  - layer_range: [4, 6]
    model: ./Yosegi-0601
- sources:
  - layer_range: [6, 8]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.05321377226808306
      - filter: mlp
        value: 0.0482589904702303
      - value: 0.433407006546336
  - layer_range: [6, 8]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 0.8300482882633113
      - filter: mlp
        value: 0.8951636861593875
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.35952608658046414
      - filter: mlp
        value: 0.17385333183950857
      - value: 0.6366514725970246
  - layer_range: [6, 8]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.7848308077099464
      - filter: mlp
        value: 0.869549457974157
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.12433943050311849
      - filter: mlp
        value: 0.3065832590226165
      - value: 0.33138948726149514
  - layer_range: [6, 8]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.11885967308786714
      - filter: mlp
        value: 0.29125668567121127
      - value: 0.19251901269486088
  - layer_range: [6, 8]
    model: ./Yosegi-0601
- sources:
  - layer_range: [8, 10]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.9429625513013793
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.4085396076816443
      - filter: mlp
        value: 0.038473657720644636
      - value: 0.35014489493395495
  - layer_range: [8, 10]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.26957216810533163
      - filter: mlp
        value: 0.2393300696241166
      - value: 0.4735322427351712
  - layer_range: [8, 10]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.8594757954447017
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.26101395702355007
      - filter: mlp
        value: 0.3147672140145126
      - value: 0.11658182776184756
  - layer_range: [8, 10]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.6948062341711919
      - value: 0.9312401427737346
      weight:
      - filter: self_attn
        value: 0.1987774487170517
      - filter: mlp
        value: 0.5628384475763534
      - value: 0.2765378221890683
  - layer_range: [8, 10]
    model: ./Yosegi-0601
- sources:
  - layer_range: [10, 12]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.8230035654228713
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.1741591536775035
      - filter: mlp
        value: 0.30563583223301516
      - value: 0.2060419023239155
  - layer_range: [10, 12]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.9991063013557119
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.1470996125766866
      - filter: mlp
        value: 0.06646481892400827
      - value: 0.2645489609472036
  - layer_range: [10, 12]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.6812899560643833
      - filter: mlp
        value: 0.9083104648631823
      - value: 0.9730062683598184
      weight:
      - filter: self_attn
        value: 0.14278507832578724
      - filter: mlp
        value: 0.3475945971407978
      - value: 0.40266546962595284
  - layer_range: [10, 12]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 0.7047231879232164
      - filter: mlp
        value: 0.9148432633716144
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.15341559366405985
      - filter: mlp
        value: 0.20047704006010095
      - value: 0.17364445581398172
  - layer_range: [10, 12]
    model: ./Yosegi-0601
- sources:
  - layer_range: [12, 14]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 0.6974090973508299
      - filter: mlp
        value: 1.0
      - value: 0.9553573565285324
      weight:
      - filter: self_attn
        value: 0.03614401712451334
      - filter: mlp
        value: 0.1287785039219736
      - value: 0.3780545754310749
  - layer_range: [12, 14]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 0.7857328784783159
      - filter: mlp
        value: 1.0
      - value: 0.6631303877423032
      weight:
      - filter: self_attn
        value: 0.21728574423632604
      - filter: mlp
        value: 0.22813107248290188
      - value: 0.1435266378249425
  - layer_range: [12, 14]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.7579910864422339
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.21526786827735228
      - filter: mlp
        value: 0.19769619474642783
      - value: 0.49420458585638627
  - layer_range: [12, 14]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 0.8379590665264793
      - filter: mlp
        value: 1.0
      - value: 0.6778673543559375
      weight:
      - filter: self_attn
        value: 0.060679858649663874
      - filter: mlp
        value: 0.17248738428562518
      - value: 0.05145640258269078
  - layer_range: [12, 14]
    model: ./Yosegi-0601
- sources:
  - layer_range: [14, 16]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.8193296716327286
      - value: 0.709644132681917
      weight:
      - filter: self_attn
        value: 0.09821428505487592
      - filter: mlp
        value: 0.0039875777021436964
      - value: 0.27550746634944184
  - layer_range: [14, 16]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 0.9420135087156387
      - filter: mlp
        value: 1.0
      - value: 0.9478569230341948
      weight:
      - filter: self_attn
        value: 0.32640822225239857
      - filter: mlp
        value: 0.28189746971019747
      - value: 0.09777040841174603
  - layer_range: [14, 16]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.9811539353914964
      - filter: mlp
        value: 1.0
      - value: 0.9947034500579488
      weight:
      - filter: self_attn
        value: 0.015308461456516246
      - filter: mlp
        value: 0.0018966958379955934
      - value: 0.24275389952300747
  - layer_range: [14, 16]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 0.9022355771447704
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.03331841447575224
      - filter: mlp
        value: 0.03561712850019841
      - value: 0.16096143804589919
  - layer_range: [14, 16]
    model: ./Yosegi-0601
- sources:
  - layer_range: [16, 18]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 0.8813466618200871
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.20435001101909528
      - filter: mlp
        value: 0.1516594727144469
      - value: 0.2269819409999868
  - layer_range: [16, 18]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 0.8113796412034742
      weight:
      - filter: self_attn
        value: 0.23760349395229585
      - filter: mlp
        value: 0.1725436279774783
      - value: 0.5818814139457673
  - layer_range: [16, 18]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.9307369835995082
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.0673898519051937
      - filter: mlp
        value: 0.049368399457210624
      - value: 0.2621269048339309
  - layer_range: [16, 18]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 0.8219541044757637
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.21320061393511042
      - filter: mlp
        value: 0.09188781867337345
      - value: 0.27266490524762327
  - layer_range: [16, 18]
    model: ./Yosegi-0601
- sources:
  - layer_range: [18, 20]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.7993530327131696
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.20420262433348008
      - filter: mlp
        value: 0.43400570066910155
      - value: 0.13720822682656159
  - layer_range: [18, 20]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 0.7035563346885239
      weight:
      - filter: self_attn
        value: 0.3313523263002212
      - filter: mlp
        value: 0.356035051194268
      - value: 0.4742357680522683
  - layer_range: [18, 20]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.2475654838180605
      - filter: mlp
        value: 0.35095371882044646
      - value: 0.18536862919946695
  - layer_range: [18, 20]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.02997204931537696
      - filter: mlp
        value: 0.4103581291392323
      - value: 0.19313933251158066
  - layer_range: [18, 20]
    model: ./Yosegi-0601
- sources:
  - layer_range: [20, 22]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 0.5321196166337413
      weight:
      - filter: self_attn
        value: 0.17930537920958298
      - filter: mlp
        value: 0.07662274511683252
      - value: 0.1354315278471591
  - layer_range: [20, 22]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.3768803907042144
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.1592147705254305
      - filter: mlp
        value: 0.18410207999201075
      - value: 0.4928015910047033
  - layer_range: [20, 22]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.37897278298418885
      - filter: mlp
        value: 0.0952591073533606
      - value: 0.03551732810121447
  - layer_range: [20, 22]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.2682334102128691
      - filter: mlp
        value: 0.33485781481395227
      - value: 0.3395139468281392
  - layer_range: [20, 22]
    model: ./Yosegi-0601
- sources:
  - layer_range: [22, 24]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.8002588203446623
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.2549204541625693
      - filter: mlp
        value: 0.3722418477156178
      - value: 0.2410463731352089
  - layer_range: [22, 24]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 0.9220873255898425
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.487455295718532
      - filter: mlp
        value: 0.40022413917173594
      - value: 0.17846009757502157
  - layer_range: [22, 24]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.7696341317318985
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.011267799816515114
      - filter: mlp
        value: 0.5320959832591042
      - value: 0.17095406531325266
  - layer_range: [22, 24]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.556101646343872
      - filter: mlp
        value: 0.5470253909079791
      - value: 0.13241555469863223
  - layer_range: [22, 24]
    model: ./Yosegi-0601
- sources:
  - layer_range: [24, 26]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 0.8667033674916582
      - filter: mlp
        value: 1.0
      - value: 0.9446091486920749
      weight:
      - filter: self_attn
        value: 0.4134110775513897
      - filter: mlp
        value: 0.0181822765943834
      - value: 0.22797659617038232
  - layer_range: [24, 26]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.9839865829690491
      - value: 0.8252981103449059
      weight:
      - filter: self_attn
        value: 0.3310295320944009
      - filter: mlp
        value: 0.05341478458353629
      - value: 0.3588847186159219
  - layer_range: [24, 26]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.8834823812212265
      - value: 0.8195593509048733
      weight:
      - filter: self_attn
        value: 0.3778012590489552
      - filter: mlp
        value: 0.2553204906819882
      - value: 0.23250565137970108
  - layer_range: [24, 26]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.7731602497153744
      - value: 0.8647152680789973
      weight:
      - filter: self_attn
        value: 0.1101209698118704
      - filter: mlp
        value: 0.2399169741437055
      - value: 0.32311925187355206
  - layer_range: [24, 26]
    model: ./Yosegi-0601
- sources:
  - layer_range: [26, 28]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.9508674341172941
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.4312865186270921
      - filter: mlp
        value: 0.28336325917543326
      - value: 0.051826325177477234
  - layer_range: [26, 28]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.8945725432745376
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.03524133636598346
      - filter: mlp
        value: 0.21426126710725438
      - value: 0.31724116335002545
  - layer_range: [26, 28]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.7138130384877139
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.04890129864608137
      - filter: mlp
        value: 0.3324333287494201
      - value: 0.11533647335498036
  - layer_range: [26, 28]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 0.9200281327001997
      weight:
      - filter: self_attn
        value: 0.300842776105564
      - filter: mlp
        value: 0.08363140003203932
      - value: 0.2538677006866867
  - layer_range: [26, 28]
    model: ./Yosegi-0601
- sources:
  - layer_range: [28, 30]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.7116000185808022
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.10977758983122704
      - filter: mlp
        value: 0.1839207861311269
      - value: 0.5426174846632369
  - layer_range: [28, 30]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.8412049419861911
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.3517232690814979
      - filter: mlp
        value: 0.11878679655495025
      - value: 0.432611353923264
  - layer_range: [28, 30]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.7196182744068202
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.29848623969081
      - filter: mlp
        value: 0.034661358236493495
      - value: 0.3438376072572394
  - layer_range: [28, 30]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.051511204430449285
      - filter: mlp
        value: 0.3617968383178797
      - value: 0.2578690795635758
  - layer_range: [28, 30]
    model: ./Yosegi-0601
- sources:
  - layer_range: [30, 32]
    model: mistralai/Mistral-7B-v0.3
    parameters:
      density:
      - filter: self_attn
        value: 0.7971002248466003
      - filter: mlp
        value: 0.8931695149333363
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.07401430804790136
      - filter: mlp
        value: 0.00696466997386886
      - value: 0.08295038526296711
  - layer_range: [30, 32]
    model: meta-math/MetaMath-Mistral-7B
    parameters:
      density:
      - filter: self_attn
        value: 1.0
      - filter: mlp
        value: 0.8158777337631619
      - value: 0.8348784699583887
      weight:
      - filter: self_attn
        value: 0.26799125918248423
      - filter: mlp
        value: 0.08176923813129498
      - value: 0.030317330226146508
  - layer_range: [30, 32]
    model: uukuguy/speechless-zephyr-code-functionary-7b
    parameters:
      density:
      - filter: self_attn
        value: 0.8188850632365792
      - filter: mlp
        value: 0.7463831519693573
      - value: 0.6515317051533988
      weight:
      - filter: self_attn
        value: 0.21122007850953434
      - filter: mlp
        value: 0.1463362342258229
      - value: 0.09176704194956312
  - layer_range: [30, 32]
    model: ./Ninja-v1-RP-expressive-v2-LoRA
    parameters:
      density:
      - filter: self_attn
        value: 0.9313941807354906
      - filter: mlp
        value: 1.0
      - value: 1.0
      weight:
      - filter: self_attn
        value: 0.1443680121177074
      - filter: mlp
        value: 0.08309606396368145
      - value: 0.37059044424517035
  - layer_range: [30, 32]
    model: ./Yosegi-0601
Downloads last month
18
Safetensors
Model size
7.24B params
Tensor type
BF16
·
Inference Examples
Text Generation
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for Local-Novel-LLM-project/Yosegi-2

meta-math/MetaMath-Mistral-7B
mistralai/Mistral-7B-v0.3
uukuguy/speechless-zephyr-code-functionary-7b
Merge model
this model
Merges
2 models