© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 130213
    Prev
    Next
    INDEX
    Explanations

    names and titles

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

    The neuron fires strongly on Chinese-style list markers and ordinals (e.g. the enumeration comma “、” and numbered section headings like “一、”, “二、”, “三、”).

    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
    孓
    -2.36
     has
    -2.11
     sucedió
    -2.00
    顰
    -1.98
    讶
    -1.95
    ığı
    -1.95
     "";
    -1.94
     sucede
    -1.91
     норма
    -1.90
    ⎽
    -1.89
    POSITIVE LOGITS
    寕
    3.70
    沨
    2.75
    悭
    2.52
    纐
    2.52
     uſed
    2.44
    樒
    2.44
     erneu
    2.41
    宬
    2.38
    ,
    2.25
     brimming
    2.25
    Activations Density 0.001%

    No Known Activations