© 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. 13837
    Prev
    Next
    INDEX
    Explanations

    news headlines and software updates

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

    The neuron fires on numeric literals (sequences of digits) in code or documentation.

    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
     tämän
    -1.25
     balcón
    -1.23
    полнение
    -1.14
    摂取
    -1.11
    izadores
    -1.11
    hater
    -1.08
     permitirá
    -1.07
    agosto
    -1.06
    enged
    -1.05
     campamento
    -1.05
    POSITIVE LOGITS
    我們
    1.20
    ArgumentParser
    1.11
    我们
    1.07
    fraid
    1.05
     cắt
    0.99
    หรือ
    0.96
    HttpEntity
    0.95
    \
    0.93
     هدف
    0.93
    ència
    0.91
    Activations Density 0.018%

    No Known Activations