© 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. 128460
    Prev
    Next
    INDEX
    Explanations

    The neuron is primarily triggered by numeric tokens (digits and numbers) in structured/data‐like contexts (e.g. table or code metadata).

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

    Predict what

    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
     will
    -1.67
     you
    -1.20
    챕
    -1.09
    مدیر
    -1.07
     sains
    -1.05
     fernández
    -1.05
    sah
    -1.05
    github
    -1.04
    toxicity
    -1.03
    stoffen
    -1.02
    POSITIVE LOGITS
     Unglück
    1.34
     schwed
    1.28
     всичко
    1.26
     extraer
    1.26
     męska
    1.22
    ülen
    1.22
     claramente
    1.21
     всички
    1.21
     observaciones
    1.20
     probablemente
    1.19
    Activations Density 0.004%

    No Known Activations