© 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. 22-GEMMASCOPE-RES-131K
    7. 72613
    Prev
    Next
    INDEX
    Explanations

    family members and counts

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

    This neuron responds to numerical quantity expressions (cardinal numbers and their descriptive modifiers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     añade
    -1.28
    もあるので
    -1.23
    你这
    -1.21
    かわい
    -1.21
    ハンドメイド
    -1.18
     supaya
    -1.16
    なので
    -1.16
    ઃ
    -1.13
     ܙ
    -1.09
    格好
    -1.09
    POSITIVE LOGITS
    ：
    1.15
    📉
    1.07
    ▽
    1.06
     as
    1.05
     husband
    1.03
    的概念
    1.02
    ánicas
    1.02
     Spagna
    1.01
     sœur
    1.01
     леса
    1.00
    Activations Density 0.001%

    No Known Activations