© 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. 70631
    Prev
    Next
    INDEX
    Explanations

    wag or swag

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

    This neuron is specialized to detect occurrences of the letter sequence “wag” (in particular the token “WAG” or forms like “wag”/“swag”).

    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
    ⼥
    -2.86
    珦
    -2.86
    嵛
    -2.86
    尛
    -2.84
    篥
    -2.84
    轂
    -2.81
    諴
    -2.78
    蠵
    -2.66
     nieuwe
    -2.61
     detenido
    -2.61
    POSITIVE LOGITS
    -
    3.75
    --
    3.19
    ?
    3.16
    of
    2.97
    ir
    2.80
    -"
    2.75
    m
    2.69
    y
    2.63
    =
    2.53
    ol
    2.47
    Activations Density 0.001%

    No Known Activations