© 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. Qwen3-1.7B
    5. 27-LLAMASCOPE-2-LORSA-16K-K64
    7. 13958
    Prev
    Next
    INDEX
    Explanations

    say "grammar"

    unknown · unknown
    New Auto-Interp
    Top Features by Cosine Similarity
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    茶园
    -20.00
    产品质量
    -18.88
    _allocation
    -18.88
    .setCode
    -18.50
    Allocation
    -17.25
    .addProperty
    -17.00
    遗产
    -16.88
    财产
    -16.88
     PropertyValue
    -16.75
    氨基酸
    -16.75
    POSITIVE LOGITS
    grammar
    21.00
     Grammar
    17.25
    语
    16.63
     conjug
    16.38
     grammar
    15.69
    造
    15.25
    struct
    15.13
    verbs
    15.06
     ké
    14.94
     gramm
    14.75
    Activations Density 0.441%

    No Known Activations