© 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. 10-GEMMASCOPE-RES-131K
    7. 127884
    Prev
    Next
    INDEX
    Explanations

    agenda points

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

    This neuron responds to non-English (foreign) words or fragments, especially those containing accented or unusual characters.

    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
    乆
    -2.58
    乄
    -2.58
    ଃ
    -2.56
    
    -2.45
    
    -2.44
    <bos>
    -2.34
    genodigd
    -2.33
     diverses
    -2.33
    쨩
    -2.33
    伈
    -2.22
    POSITIVE LOGITS
     in
    3.23
     because
    2.61
    ～　
    2.38
     with
    2.34
     seven
    2.33
     despite
    2.33
     and
    2.30
    One
    2.25
     incredible
    2.23
     unusual
    2.22
    Activations Density 0.011%

    No Known Activations