© 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. 14630
    Prev
    Next
    INDEX
    Explanations

    say magic

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

    No Comments

    Negative Logits
    bench
    -18.63
     форум
    -17.75
    裕
    -17.00
    orer
    -16.88
     forums
    -16.50
     Forums
    -16.25
    贲
    -16.13
    orde
    -15.75
    Verifier
    -15.38
     Parse
    -15.31
    POSITIVE LOGITS
    雪花
    15.50
    魔法师
    15.00
    magic
    14.56
     Magical
    14.44
    魔法
    14.44
     magic
    14.00
     magician
    14.00
    ohl
    13.94
    目
    13.75
     MAGIC
    13.56
    Activations Density 0.021%

    No Known Activations