© 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. 16158
    Prev
    Next
    INDEX
    Explanations

    values with +/-

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

    This neuron specifically detects numeric measurement tokens (numbers and numeric values) in technical contexts.

    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
     汽车
    -1.21
    рно
    -1.13
     السياس
    -1.09
    klima
    -1.07
     etc
    -1.05
     pouvoit
    -1.05
     publier
    -1.05
    Popis
    -1.04
    Grup
    -1.04
     innych
    -1.02
    POSITIVE LOGITS
     for
    1.45
     دخترانه
    1.16
    pushFollow
    1.13
     almost
    1.12
    🌒
    1.10
     only
    1.09
     ajang
    1.09
     технологі
    1.08
     nearly
    1.06
     erzeugen
    1.06
    Activations Density 0.014%

    No Known Activations