© 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. 10-GEMMASCOPE-RES-131K
    7. 60963
    Prev
    Next
    INDEX
    Explanations

    Razor page code blocks

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

    The neuron fires on programming‐language or templating directives and other snippet delimiters (e.g. “@page”, “@{”, “package(…)”, “load(…)”, “$” prompts), i.e. markers that signal the start of code or code‐related declarations.

    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
     \
    -1.91
    𖣘
    -1.90
     unwittingly
    -1.73
    珦
    -1.67
     और
    -1.61
     на
    -1.59
    別の
    -1.57
     και
    -1.56
     другим
    -1.54
     назва
    -1.52
    POSITIVE LOGITS
    consejos
    1.61
     観光
    1.58
    踯
    1.52
    勦
    1.47
     ongles
    1.46
    棡
    1.45
    owany
    1.38
    鸶
    1.38
    ↵
    1.37
    觊
    1.36
    Activations Density 0.001%

    No Known Activations