© 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. 22-GEMMASCOPE-RES-131K
    7. 75085
    Prev
    Next
    INDEX
    Explanations

    international career

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

    The neuron fires on richer, low‐frequency content words—i.e. descriptive or emphatic nouns and adjectives conveying strong or specific meaning.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     since
    -1.38
     in
    -1.37
     it
    -1.27
     my
    -1.20
     for
    -1.16
    orothy
    -1.16
     so
    -1.15
     our
    -1.15
     only
    -1.14
     comienza
    -1.14
    POSITIVE LOGITS
     DIFFERENCE
    1.22
     PROB
    1.20
    ｍ
    1.13
     minimale
    1.12
     RELATIVE
    1.10
     lokale
    1.09
     keten
    1.09
     vinyle
    1.07
     formules
    1.06
    supervised
    1.05
    Activations Density 0.024%

    No Known Activations