© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 19503
    Prev
    Next
    INDEX
    Explanations

    code variables

    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
    doch
    -0.06
    åĻľ
    -0.06
    ÑģÑĤеÑģÑĤвенно
    -0.05
    itou
    -0.05
    istou
    -0.05
    ارة
    -0.05
     اÙĦÙħتÙħ
    -0.05
    =./
    -0.05
     Tow
    -0.05
    Incre
    -0.05
    POSITIVE LOGITS
    ugar
    0.06
    _cast
    0.06
    казаÑĤÑĮ
    0.05
    宫
    0.05
     ones
    0.05
     longer
    0.05
    hares
    0.05
    /swagger
    0.05
    ion
    0.05
    é¡ŀ
    0.05
    Activations Density 0.040%

    No Known Activations