© 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. 13030
    Prev
    Next
    INDEX
    Explanations

    configuration files

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

    The neuron fires strongly on mentions of a “descriptor” (especially in the context of reusing or defining a shared descriptor in build or assembly configurations).

    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
    はなく
    -1.11
     women
    -1.11
    ’
    -1.10
     cultural
    -1.09
     rheinland
    -1.07
     food
    -1.07
     Item
    -1.05
    ^
    -1.05
     -
    -1.01
     первом
    -1.01
    POSITIVE LOGITS
     file
    1.19
    파일
    1.16
     docteur
    1.13
    ִי
    1.12
    Выбор
    1.11
     (*.
    1.09
    ítico
    1.09
     Docteur
    1.07
     Pilihan
    1.07
    íticos
    1.07
    Activations Density 0.083%

    No Known Activations