© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27391
    Prev
    Next
    INDEX
    Explanations

    internal variable names

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

    The neuron activates on standalone underscore tokens (i.e. “_”).

    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
     decoradas
    -2.31
    UCIÓN
    -2.19
    <bos>
    -2.17
     magnética
    -2.14
    觧
    -2.06
     kopling
    -2.05
    UCION
    -2.03
     It
    -2.03
    érienne
    -2.02
     menn
    -2.00
    POSITIVE LOGITS
    Y
    2.34
    她
    2.20
    H
    2.14
    [
    2.13
    ↵↵
    2.13
     extremely
    2.13
    昔から
    2.11
    .'
    2.11
    
    2.08
    もうすぐ
    2.05
    Activations Density 0.015%

    No Known Activations