© 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. 21105
    Prev
    Next
    INDEX
    Explanations

    In Black

    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
    ä¸įä½ľ
    -0.06
    è§ģ
    -0.05
     lata
    -0.05
    æ±°
    -0.05
    vet
    -0.05
     multiplic
    -0.04
    建æĸ°
    -0.04
     bearer
    -0.04
    à§ĩà¦ķà§ĩ
    -0.04
    ou
    -0.04
    POSITIVE LOGITS
    ÑıÑģÑĮ
    0.08
     bref
    0.07
    æĭ¬
    0.06
    asters
    0.06
    stel
    0.06
     officielle
    0.06
     Pulitzer
    0.06
    缸æĴŀ
    0.06
     intérieure
    0.06
    jern
    0.06
    Activations Density 0.003%

    No Known Activations