© 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. 22005
    Prev
    Next
    INDEX
    Explanations

    xml version declaration

    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
     leftover
    -0.05
    åī©ä½Ļ
    -0.05
    remaining
    -0.05
     remaining
    -0.05
     Erb
    -0.05
     ترکÛĮ
    -0.05
    çļĦç»ĦåIJĪ
    -0.05
    ç®ĹäºĨ
    -0.05
    æĮīä½ı
    -0.05
    _remaining
    -0.05
    POSITIVE LOGITS
    åĽ¾çīĩæĪĸ
    0.07
    tsy
    0.07
    оÑĤе
    0.07
    åĽºçĦ¶
    0.06
    fér
    0.06
    ousse
    0.06
    ç§ijæĬĢ大åѦ
    0.06
    etje
    0.06
    اط
    0.06
    à¸ļาย
    0.06
    Activations Density 0.022%

    No Known Activations