© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 68774
    Prev
    Next
    INDEX
    Explanations

    scientific notation and math expressions

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

    The neuron mainly responds to numeric tokens and numeric expressions (e.g. digits, decimals, and scientific‐notation numbers).

    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
     of
    -1.58
     you
    -1.31
     peut
    -1.30
     they
    -1.29
    
    -1.14
     on
    -1.13
     mapas
    -1.13
     baloncesto
    -1.13
     aparato
    -1.10
    óny
    -1.09
    POSITIVE LOGITS
    3
    1.20
    aneity
    1.20
    なー
    1.18
    有点
    1.16
    了一种
    1.12
    agerie
    1.10
    4
    1.09
     geliefert
    1.05
     mówią
    1.05
    преки
    1.02
    Activations Density 0.003%

    No Known Activations