© 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. 129225
    Prev
    Next
    INDEX
    Explanations

    This neuron identifies tokens that are standalone numeric values (digits or multi‐digit numbers).

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

    A or Primary followed by specific nouns

    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
     they
    -1.45
     zitate
    -1.21
    鶺
    -1.20
     ahu
    -1.17
     évé
    -1.17
     púb
    -1.16
     STUART
    -1.16
    nelles
    -1.15
     ligado
    -1.14
    cestors
    -1.14
    POSITIVE LOGITS
     quizá
    1.27
     Also
    1.23
     nuevamente
    1.22
     prácticamente
    1.22
    ご連絡
    1.20
     There
    1.18
     также
    1.13
     W
    1.12
    M
    1.11
    Instead
    1.10
    Activations Density 0.000%

    No Known Activations