© 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. 112476
    Prev
    Next
    INDEX
    Explanations

    chapter or number

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

    This neuron selectively activates on numeric literals (including integers and decimals) appearing in code or comments.

    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
     industrias
    -1.60
     sogen
    -1.52
     erstes
    -1.48
     forty
    -1.45
     three
    -1.41
     neuem
    -1.37
     pierws
    -1.37
     пять
    -1.31
     be
    -1.30
     five
    -1.30
    POSITIVE LOGITS
     karde
    1.39
    踽
    1.38
     hiz
    1.31
     engel
    1.30
     huk
    1.30
     adik
    1.28
     yeti
    1.27
     tekno
    1.27
     igen
    1.25
     kalp
    1.25
    Activations Density 0.059%

    No Known Activations