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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 65803
    Prev
    Next
    INDEX
    Explanations

    parentheses and associated punctuation

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

    The neuron detects textual emphasis cues and markup (e.g. phrases like “emphasis added/mine,” bold‐ or underscore‐style formatting markers).

    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
    зка
    -0.92
    当初
    -0.91
     schlechte
    -0.91
    лежки
    -0.86
     WHEEL
    -0.81
    because
    -0.81
    Atual
    -0.80
     LATER
    -0.79
     harapan
    -0.79
     kompeti
    -0.79
    POSITIVE LOGITS
    Hilarious
    0.91
     ihrer
    0.91
    Ehh
    0.88
     $+$
    0.87
    ufac
    0.85
    わけではない
    0.84
    ̗
    0.83
     ihren
    0.82
    FTFY
    0.82
    RENT
    0.82
    Activations Density 0.029%

    No Known Activations