© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 20411
    Prev
    Next
    INDEX
    Explanations

    Johannes and unistd.h

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

    The neuron is detecting capitalized personal and place names (proper nouns) in 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
    莇
    -2.61
    哢
    -2.61
    ⃢
    -2.59
    蟜
    -2.58
    鲧
    -2.52
    尨
    -2.48
    ERON
    -2.39
    拶
    -2.36
    ÈME
    -2.34
    extField
    -2.33
    POSITIVE LOGITS
    Also
    3.14
    .
    3.06
    I
    2.91
    A
    2.91
     have
    2.88
    0
    2.83
    However
    2.80
    L
    2.78
    P
    2.52
    Other
    2.50
    Activations Density 0.054%

    No Known Activations