© 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. 10-GEMMASCOPE-RES-131K
    7. 7225
    Prev
    Next
    INDEX
    Explanations

    number one

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

    The neuron consistently activates on numeric tokens (numbers appearing in text, code, or math expressions).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    türk
    -1.41
     Before
    -1.31
     biß
    -1.30
     bisschen
    -1.28
     вси
    -1.27
     catalán
    -1.27
     föres
    -1.27
     häls
    -1.27
    他に
    -1.26
     metrów
    -1.25
    POSITIVE LOGITS
     entusias
    1.73
     vivace
    1.61
     vixion
    1.55
    卺
    1.52
    螃
    1.48
     discuter
    1.46
    飈
    1.46
    ᨆ
    1.45
     kristal
    1.45
     marcato
    1.45
    Activations Density 0.006%

    No Known Activations