© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-1.7B
    5. 27-LLAMASCOPE-2-LORSA-16K-K64
    7. 15787
    Prev
    Next
    INDEX
    Explanations

    do not say th

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

    No Comments

    Negative Logits
    -th
    -62.00
    Th
    -60.50
    -Th
    -56.00
    (th
    -55.25
     Th
    -55.25
    	th
    -54.75
     TH
    -53.25
    TH
    -52.00
    _th
    -51.50
    <th
    -51.50
    POSITIVE LOGITS
    迁
    15.56
    ocator
    15.13
    榉
    15.00
    struk
    14.50
     Giul
    14.50
    经开
    14.31
    枧
    14.25
    菅
    14.00
    龙门
    13.94
    外援
    13.88
    Activations Density 5.041%

    No Known Activations