© 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. 33352
    Prev
    Next
    INDEX
    Explanations

    experimental analysis and applications

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

    The neuron is selectively activated by tokens that are numeric (i.e. digits or multi‐digit numbers).

    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
     Aseg
    -1.30
    normally
    -1.23
     Puedes
    -1.23
     Incluye
    -1.21
     ſeveral
    -1.20
     effectué
    -1.20
     Condiciones
    -1.19
     мужская
    -1.19
     Pues
    -1.17
    isst
    -1.16
    POSITIVE LOGITS
     and
    2.05
    <eos>
    1.42
    ></
    1.39
     will
    1.28
     it
    1.18
    ↵
    1.18
     görüntüleri
    1.16
     görüntüsü
    1.15
    </h1>
    1.13
    Legende
    1.10
    Activations Density 0.089%

    No Known Activations