© 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. 102265
    Prev
    Next
    INDEX
    Explanations

    anywhere, somewhere, everywhere

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

    This neuron activates on adverbial location words containing “where” (e.g. somewhere, anywhere, everywhere).

    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
    🫃
    -2.44
    𓆏
    -2.42
    🫅
    -2.38
    𓁹
    -2.30
    🫴
    -2.22
     editados
    -2.19
     preparados
    -2.13
     enviados
    -2.09
    蹀
    -2.08
    鲥
    -2.03
    POSITIVE LOGITS
    in
    2.98
    f
    2.56
    v
    2.38
    end
    2.23
    on
    2.23
    s
    2.06
    il
    1.96
     else
    1.95
    n
    1.95
    m
    1.91
    Activations Density 0.020%

    No Known Activations