© 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. 71396
    Prev
    Next
    INDEX
    Explanations

    names and diverse languages

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

    The neuron fires on uncommon subword stems—especially the initial segments of technical or foreign‐language terms—marking the beginnings of specialized or nonstandard words.

    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
    '
    -3.22
     '
    -2.86
    </b>
    -2.75
     he
    -2.55
    "
    -2.48
    ↵
    -2.33
    霎
    -2.31
     siquiera
    -2.25
    ).}
    -2.23
    ,
    -2.19
    POSITIVE LOGITS
    鬭
    3.06
    涐
    2.91
    筚
    2.86
    Веду
    2.73
     sprüche
    2.58
    呶
    2.56
    샅
    2.55
    읭
    2.53
     anhänger
    2.53
    ègre
    2.52
    Activations Density 0.236%

    No Known Activations