© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 84546
    Prev
    Next
    INDEX
    Explanations

    golden hour, golden triangle, golden age

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

    This neuron fires on title‐style, multiword proper nouns in uppercase/lowercase title case—e.g. “Golden Hour,” “Golden Triangle,” “Big Lie,” “Hundred Days,” etc.

    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
     brachten
    -1.80
     and
    -1.67
     abundancia
    -1.59
    看起來
    -1.58
     chuť
    -1.55
     ГА
    -1.52
    ordes
    -1.52
    廝
    -1.51
    ピール
    -1.51
    biendo
    -1.51
    POSITIVE LOGITS
     of
    1.66
    ps
    1.56
     any
    1.53
    id
    1.48
    OfThe
    1.42
     time
    1.40
     falsa
    1.38
     мето
    1.37
    </b>
    1.34
     comunica
    1.34
    Activations Density 0.129%

    No Known Activations