© 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. 18047
    Prev
    Next
    INDEX
    Explanations

    function definitions

    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
    inherits
    -0.08
    urin
    -0.07
    toff
    -0.06
     defens
    -0.06
     Paral
    -0.06
    éĤı
    -0.06
    kofen
    -0.06
    à¹Ģà¸Ĺศ
    -0.06
    äºĨä¸Ģè·³
    -0.06
    istring
    -0.06
    POSITIVE LOGITS
    åĪĿ
    0.06
    oz
    0.05
     involved
    0.05
    av
    0.05
    ^{
    0.05
    z
    0.05
    è¿Ļç§įåģļæ³ķ
    0.05
    br
    0.04
    C
    0.04
     -
    0.04
    Activations Density 0.043%

    No Known Activations