© 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. 10-GEMMASCOPE-RES-131K
    4. 48119
    Prev
    Next
    INDEX
    Explanations

    Type, Section, Class followed by identifiers

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

    The neuron activates on numeric or alphanumeric tokens used to label sections, types, or other numbered designations (e.g. “Type I,” “Section 110.5,” part/Rule numbers).

    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
    $',
    -1.28
     who
    -1.27
     effect
    -1.21
     تصریح
    -1.20
    ",
    -1.20
     sebagainya
    -1.20
     değil
    -1.16
     thing
    -1.14
    qJ
    -1.14
     link
    -1.13
    POSITIVE LOGITS
    回到
    1.20
    colección
    1.17
    ьере
    1.12
    pentru
    1.12
    Contenu
    1.11
    êtement
    1.11
    pueden
    1.11
     mesmas
    1.11
     amic
    1.09
    idopsis
    1.09
    Activations Density 0.088%

    No Known Activations