© 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. 17215
    Prev
    Next
    INDEX
    Explanations

    new baby, sacredness, commit murder

    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
    ØŃÙĬ
    -0.07
    stal
    -0.06
     جÙĨب
    -0.06
    çī¹
    -0.06
    ä¸įåĬ¨æijĩ
    -0.06
    amb
    -0.06
    alist
    -0.06
    entang
    -0.06
    åIJĦå¸Ĥ
    -0.06
     sny
    -0.06
    POSITIVE LOGITS
    å®ŀåĪĻ
    0.05
    perse
    0.05
    ì¹ĺê²Į
    0.05
    áu
    0.05
     eigentlich
    0.05
     perci
    0.05
    erto
    0.04
    isure
    0.04
    erkers
    0.04
    urses
    0.04
    Activations Density 0.011%

    No Known Activations