© 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. 105618
    Prev
    Next
    INDEX
    Explanations

    otherwise followed by contrast

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

    The neuron specifically detects the adverb “otherwise.”

    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
     asesinado
    -1.42
    どのような
    -1.41
     condenado
    -1.40
    ブログ村
    -1.40
     obno
    -1.38
    鱨
    -1.38
     lihat
    -1.34
    istika
    -1.34
    噘
    -1.31
    莚
    -1.31
    POSITIVE LOGITS
    *
    1.92
    m
    1.79
    is
    1.77
    。
    1.68
    rs
    1.66
    [
    1.63
    w
    1.58
    та
    1.57
    ta
    1.56
    st
    1.55
    Activations Density 0.014%

    No Known Activations