© 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. 21528
    Prev
    Next
    INDEX
    Explanations

    web apps and projects

    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
    hom
    -0.06
     hom
    -0.06
    çݯä¿Ŀç§ijæĬĢæľīéĻIJåħ¬åı¸
    -0.05
    endale
    -0.05
    u
    -0.05
    å¾®èĸĦ
    -0.05
    ieval
    -0.05
    ln
    -0.05
    便èĥ½
    -0.05
     contemplate
    -0.05
    POSITIVE LOGITS
    éģ·
    0.06
    -transitional
    0.06
    ductive
    0.06
    ê³Ħ
    0.06
    pressive
    0.05
     migrations
    0.05
    Å¡ek
    0.05
    Å¡k
    0.05
     вп
    0.05
    qus
    0.05
    Activations Density 0.023%

    No Known Activations