© 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. 22-GEMMASCOPE-RES-131K
    7. 14704
    Prev
    Next
    INDEX
    Explanations

    technical or academic text introductions

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

    The neuron fires whenever it encounters a standalone multi‐digit number token.

    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
     who
    -1.12
     Uniwers
    -1.02
     КУ
    -0.98
    EMPO
    -0.98
    kami
    -0.97
    НОЕ
    -0.97
     Instyt
    -0.96
    ⌀
    -0.96
    ISIONS
    -0.96
    ∅
    -0.94
    POSITIVE LOGITS
    ratory
    1.09
    Waff
    1.07
    Mä
    1.04
    ただし
    1.03
     berbagai
    1.02
    Geld
    1.02
    三大
    1.02
    Transk
    1.02
    Gek
    1.01
    小さい
    0.98
    Activations Density 0.003%

    No Known Activations