© 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. 27180
    Prev
    Next
    INDEX
    Explanations

    consistency and consistent

    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
     folds
    -0.07
    缴æİ¥åIJij
    -0.07
     outright
    -0.06
    Insn
    -0.06
     вовÑģе
    -0.06
    é§Ĩãģij
    -0.06
    æİ´
    -0.06
    kaart
    -0.06
    wak
    -0.06
    æĮij
    -0.06
    POSITIVE LOGITS
     consistency
    0.15
    ä¸Ģèĩ´
    0.13
    ä¿ĿæĮģä¸Ģèĩ´
    0.13
    ä¸Ģèĩ´çļĦ
    0.13
     consistent
    0.13
     throughout
    0.13
    consistent
    0.12
    çļĦä¸Ģèĩ´æĢ§
    0.12
     across
    0.12
    ä¸Ģèĩ´æĢ§
    0.12
    Activations Density 0.045%

    No Known Activations