© 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. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 73129
    Prev
    Next
    INDEX
    Explanations

    witness/实例 after book/names

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is selectively activating on non‐Latin (CJK) characters, i.e. Chinese or Japanese logograms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    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
     ใน
    -2.91
    ísticos
    -2.75
    URIDAD
    -2.58
    ACIÓN
    -2.53
    </i>
    -2.41
    秫
    -2.41
    籥
    -2.41
    
    -2.36
    şağı
    -2.31
    %;
    -2.27
    POSITIVE LOGITS
    </sup>
    2.88
    𞤢
    2.64
    {
    2.55
    (
    2.53
    pecabe
    2.48
    [
    2.27
    It
    2.25
    $
    2.22
    ance
    2.16
    画像は
    2.13
    Activations Density 0.001%

    No Known Activations