© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 22519
    Prev
    Next
    INDEX
    Explanations

    So

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    é£İ顺
    -0.06
    æļĤè¡ĮåĬŀæ³ķ
    -0.06
    è§Ħ模以ä¸Ĭ
    -0.06
    inati
    -0.06
     humanoid
    -0.06
     جÙĪÙĦ
    -0.05
    åĺĽ
    -0.05
    goed
    -0.05
     beh
    -0.05
    inery
    -0.05
    POSITIVE LOGITS
    å¾Ĺ失
    0.06
    æĿĥè¡¡
    0.06
    åĪ©å®³åħ³ç³»
    0.06
     edge
    0.06
    agine
    0.06
     threat
    0.06
    Ñī
    0.05
    лик
    0.05
    失
    0.05
    Risk
    0.05
    Activations Density 0.034%

    No Known Activations