© 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. 129330
    Prev
    Next
    INDEX
    Explanations

    This neuron detects non‐word formatting and markup tokens (e.g. asterisks, equal signs, braces, dashes) used as section or emphasis markers in scholarly text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    scientific and medical literature

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

    scientific names and terms formatted in italics (indicated by asterisks around text).

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    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
    <bos>
    -1.03
    如此
    -0.90
    2
    -0.89
     to
    -0.84
     ultima
    -0.84
     parlano
    -0.84
    anego
    -0.82
     Empfehlungen
    -0.82
     Scopus
    -0.81
     progress
    -0.81
    POSITIVE LOGITS
    Ŀ
    1.03
    鸾
    0.96
     autobús
    0.96
    銑
    0.93
    \&
    0.93
    tament
    0.92
     chauffage
    0.91
    ilets
    0.88
    羨
    0.88
    ʧ
    0.88
    Activations Density 0.544%

    No Known Activations