© 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. 13132
    Prev
    Next
    INDEX
    Explanations

    settings and options

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

    The neuron fires on UI command and option keywords—terms that name actions or settings (e.g. resize, crop, flip, rotate, save, backup, bold, italic, etc.).

    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
    ösen
    -1.17
    haha
    -1.12
    ans
    -1.11
    ῖς
    -1.10
    менту
    -1.07
    queryInterface
    -1.07
    嘢
    -1.07
     моего
    -1.05
    albeit
    -1.05
    嘅
    -1.05
    POSITIVE LOGITS
     or
    1.88
     etc
    1.40
     if
    1.40
     various
    1.32
     and
    1.31
     for
    1.23
     🎉
    1.21
     hidupnya
    1.20
     sekarang
    1.13
     jule
    1.12
    Activations Density 0.064%

    No Known Activations