© 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. 104100
    Prev
    Next
    INDEX
    Explanations

    reply or comment indicators

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

    The neuron fires on colon punctuation marks—especially the “:” in reply headers like “Re: …”—i.e. it detects reply‐style colons.

    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
    “
    -1.50
    ‘
    -1.45
    ”,
    -1.42
    {/*
    -1.41
    /*
    -1.40
    /*
    -1.37
     ‘
    -1.36
     Instagram
    -1.34
     '
    -1.33
     You
    -1.33
    POSITIVE LOGITS
    Breite
    1.62
    刪
    1.57
     Infór
    1.48
     either
    1.45
    Größe
    1.39
    suddenly
    1.34
    Şi
    1.33
    assistance
    1.31
    bangun
    1.30
    年輕
    1.29
    Activations Density 0.044%

    No Known Activations