© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 84250
    Prev
    Next
    INDEX
    Explanations

    By Microsoft

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

    The neuron activates on numeric tokens and number-like expressions (e.g. digits, prices, and percentages).

    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
    isiin
    -1.48
     my
    -1.48
     three
    -1.47
     school
    -1.37
     five
    -1.34
     мой
    -1.29
     family
    -1.28
     winter
    -1.27
     nine
    -1.25
     four
    -1.24
    POSITIVE LOGITS
    
    1.43
    Kapcsolódó
    1.41
     设计
    1.38
     serangga
    1.38
     bugün
    1.31
    न्त
    1.30
    謚
    1.30
     overzicht
    1.30
     绘画
    1.30
     illustratie
    1.30
    Activations Density 0.086%

    No Known Activations