© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 21770
    Prev
    Next
    INDEX
    Explanations

    way to create

    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
    Not in Any Lists

    No Comments

    Negative Logits
     kaut
    -0.06
    kau
    -0.06
     Ø¥ÙĨÙĩ
    -0.06
    ลà¹Į
    -0.05
    ganti
    -0.05
    pädagog
    -0.05
    odle
    -0.05
     ôt
    -0.05
    enius
    -0.05
    roe
    -0.05
    POSITIVE LOGITS
    ç¶
    0.06
    inch
    0.06
    voll
    0.06
    oy
    0.05
    æļ´
    0.05
    abia
    0.05
    ç¹
    0.05
    ckt
    0.05
    jem
    0.05
    é£Ł
    0.05
    Activations Density 0.311%

    No Known Activations