© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-1.7B
    5. 26-LLAMASCOPE-2-LORSA-16K-K64
    7. 3102
    Prev
    Next
    INDEX
    Explanations

    say "box"

    unknown · unknown
    New Auto-Interp
    Top Features by Cosine Similarity
    Embeds
    IFrame
    Link
    Negative Logits
    -method
    -19.88
    method
    -19.38
    (method
    -19.00
     method
    -18.38
    xbb
    -17.75
    _method
    -17.25
    方法
    -16.88
    -Methods
    -16.63
     treatment
    -16.25
     Method
    -16.25
    POSITIVE LOGITS
     box
    43.25
     Box
    38.50
    框
    35.00
    box
    34.50
    Box
    34.50
    .box
    33.75
    _box
    33.75
    -box
    32.00
    	box
    31.38
     boxes
    30.25
    Activations Density 0.020%

    No Known Activations