© 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. 128416
    Prev
    Next
    INDEX
    Explanations

    This neuron responds to standalone numeric literals (large numbers) embedded in the text.

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

    license, responsibility, position, library

    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
     Haare
    -0.96
    en
    -0.90
    jącej
    -0.87
     bahkan
    -0.85
    віду
    -0.85
    Tennis
    -0.84
     krótk
    -0.84
    L
    -0.83
    s
    -0.83
    女人的
    -0.82
    POSITIVE LOGITS
    Programma
    0.98
    __[
    0.97
    sdag
    0.91
    PLACEMENT
    0.87
    Etimo
    0.85
     Ghan
    0.85
     laten
    0.84
    üdis
    0.83
    λίου
    0.83
    qued
    0.83
    Activations Density 0.010%

    No Known Activations