© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129946
    Prev
    Next
    INDEX
    Explanations

    Explanation of neuron 4 behavior: the main thing this neuron does is find promotional superlatives and hype language.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    trade event

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     children
    -1.71
     Erziehung
    -1.48
     hermosas
    -1.45
     people
    -1.45
     examples
    -1.37
     my
    -1.35
     without
    -1.35
     what
    -1.29
     using
    -1.28
     school
    -1.27
    POSITIVE LOGITS
    MQTT
    1.31
    身形
    1.29
     preceding
    1.28
    ֶּ
    1.27
     marché
    1.27
     ۳۰
    1.24
    [+]
    1.23
     меда
    1.20
     EXHIB
    1.19
    šin
    1.18
    Activations Density 0.024%

    No Known Activations