© 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. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 19803
    Prev
    Next
    INDEX
    Explanations

    rather, preferred, or instead

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æľªåħį
    -0.07
    _HTML
    -0.07
    #error
    -0.07
    è¡¡
    -0.07
    åŁºæľ¬
    -0.07
    èĶļ
    -0.07
    缴æİ¥æĪĸéĹ´æİ¥
    -0.07
    æł¹æľ¬å°±æ²¡æľī
    -0.07
    oola
    -0.07
    Prem
    -0.06
    POSITIVE LOGITS
    rather
    0.11
    ä¸ºå®ľ
    0.11
    èĢĮéĿŀ
    0.11
    为佳
    0.11
     rather
    0.10
    preferred
    0.10
    èĢĮä¸įæĺ¯
    0.10
     plutôt
    0.10
     preferred
    0.09
     instead
    0.09
    Activations Density 0.256%

    No Known Activations