© 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. 22-GEMMASCOPE-RES-131K
    4. 36228
    Prev
    Next
    INDEX
    Explanations

    code file formats

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

    The neuron responds to numbered labels or numeric references (e.g. section numbers, legal citations, figure/table 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_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
     my
    -1.69
     I
    -1.68
     you
    -1.50
     
    -1.16
    ność
    -1.12
    ૉ
    -1.09
     If
    -1.09
    stoffen
    -1.05
    领先
    -1.04
     You
    -1.02
    POSITIVE LOGITS
     ویتامین
    1.42
    はじめまして
    1.35
    こんばん
    1.34
     pidana
    1.32
     饼
    1.30
    ञ्
    1.22
    excelente
    1.20
     labdar
    1.20
     again
    1.20
     تعدى
    1.19
    Activations Density 0.000%

    No Known Activations