© 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. 127882
    Prev
    Next
    INDEX
    Explanations

    names and specific phrases

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

    The neuron fires on the Chinese/Japanese character “連” (and its compounds like “連続” or “連発”) or its pinyin/romanized form “lian,” i.e. tokens conveying the notion of series/continuity.

    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
    ↵↵
    -4.03
    谢谢你
    -2.70
     of
    -2.45
     traditionellen
    -2.44
     nowego
    -2.44
    嘿嘿
    -2.36
    {
    -2.36
    蠖
    -2.34
    哎呀
    -2.30
    ي
    -2.28
    POSITIVE LOGITS
    it
    2.72
    렷
    2.63
    ).
    2.50
    şiv
    2.47
    不断的
    2.41
     zoude
    2.41
    }.
    2.39
    ܜ
    2.34
    地在
    2.34
    enteri
    2.34
    Activations Density 0.000%

    No Known Activations