© 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. 21715
    Prev
    Next
    INDEX
    Explanations

    code object assignment

    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
    æīĭæĦŁ
    -0.06
     socket
    -0.05
    ying
    -0.05
    locator
    -0.05
    ses
    -0.05
     stalk
    -0.05
    èĢģ人家
    -0.05
    åįĬå²Ľ
    -0.05
    amping
    -0.05
    arm
    -0.05
    POSITIVE LOGITS
    ongo
    0.06
    ondre
    0.06
     esclus
    0.06
    ubi
    0.05
    ap
    0.05
    hev
    0.05
    heti
    0.05
    iev
    0.05
    afa
    0.05
    usalem
    0.05
    Activations Density 0.001%

    No Known Activations