© 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. 67492
    Prev
    Next
    INDEX
    Explanations

    word endings and foreign names

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

    The neuron primarily activates on non-English word fragments (i.e. tokens from foreign languages).

    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
     nieuwe
    -1.72
    だけでなく
    -1.59
    roud
    -1.59
    だけではなく
    -1.56
     ">
    -1.44
    felizmente
    -1.44
    zusch
    -1.41
     maaaring
    -1.41
    梱
    -1.40
     specif
    -1.38
    POSITIVE LOGITS
     was
    1.90
    ஂ
    1.81
     two
    1.76
     if
    1.67
     three
    1.66
     just
    1.63
     definitely
    1.62
     amusing
    1.56
     Definitely
    1.55
     one
    1.50
    Activations Density 0.081%

    No Known Activations