© 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. 21686
    Prev
    Next
    INDEX
    Explanations

    table formatting characters

    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
    èģĮ
    -0.07
    áĥIJáĥ®
    -0.07
    werken
    -0.06
    agné
    -0.06
    Certain
    -0.06
    æīĵéĢłæĪIJ为
    -0.06
    mêmes
    -0.06
    idler
    -0.06
    dik
    -0.06
    inasikan
    -0.06
    POSITIVE LOGITS
    w
    0.06
    2
    0.05
    3
    0.05
    iv
    0.05
     tonel
    0.05
    çİĭçļĦ
    0.05
    夫åIJĽ
    0.05
     t
    0.05
    åĽĽæĸ¹
    0.05
    Âł
    0.05
    Activations Density 0.010%

    No Known Activations