© 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. 10-GEMMASCOPE-RES-131K
    7. 60430
    Prev
    Next
    INDEX
    Explanations

    Japanese and multilingual words

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

    The neuron is activated by uppercase-letter acronyms or initialisms (e.g. UE, PZs, ITF) in the text.

    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
    </h2>
    -2.75
    \
    -2.53
    medriver
    -2.48
     beinahe
    -2.41
    one
    -2.41
    さぁ
    -2.36
    i
    -2.34
     reichlich
    -2.33
    「
    -2.30
    닌
    -2.20
    POSITIVE LOGITS
     l
    2.92
     algunas
    2.73
     algún
    2.69
     algunos
    2.66
     großen
    2.56
    玙
    2.56
    聍
    2.50
     какой
    2.48
     может
    2.44
    phans
    2.42
    Activations Density 0.055%

    No Known Activations