© 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. 1742
    Prev
    Next
    INDEX
    Explanations

    CCD cameras and output

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

    The neuron selectively activates on isolated capitalized tokens—proper nouns, acronyms, or other uppercase identifiers.

    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
    e
    -2.77
    溦
    -2.59
    in
    -2.53
     exacting
    -2.48
    r
    -2.48
     impede
    -2.47
     insidious
    -2.41
     Coupled
    -2.39
    G
    -2.38
     et
    -2.36
    POSITIVE LOGITS
    ים
    3.47
    s
    3.30
    禤
    2.84
    堙
    2.81
    ς
    2.77
    寕
    2.69
     fashioned
    2.63
    塒
    2.61
    <bos>
    2.58
    垻
    2.56
    Activations Density 0.021%

    No Known Activations