© 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. 128520
    Prev
    Next
    INDEX
    Explanations

    The neuron broadly activates on numeric tokens—i.e. standalone digits or number literals in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    tokens after symbols or numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     preparing
    -1.10
     przecież
    -0.99
     GEOLOGY
    -0.97
    ).
    -0.95
     undertaken
    -0.94
    hidro
    -0.93
     assessment
    -0.92
     Size
    -0.92
     texted
    -0.91
     that
    -0.90
    POSITIVE LOGITS
     havet
    1.05
    autres
    1.04
    descri
    1.00
    fahrene
    1.00
    札
    1.00
    became
    0.97
     çeşitli
    0.96
     рисун
    0.95
     kebanyakan
    0.91
     starte
    0.91
    Activations Density 0.001%

    No Known Activations