© 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. 36198
    Prev
    Next
    INDEX
    Explanations

    password, request, view, build, opencv, HB, Caps, shower, US, HP, birthday, channel

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

    The neuron activates strongly on all‐capitalized tokens (acronyms or uppercase abbreviations).

    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
     with
    -2.09
     our
    -1.88
     this
    -1.70
     you
    -1.54
     your
    -1.50
     áreas
    -1.50
     one
    -1.48
     that
    -1.46
     in
    -1.46
     baratas
    -1.44
    POSITIVE LOGITS
    v
    1.49
    extremely
    1.44
    is
    1.43
    are
    1.42
    does
    1.41
    con
    1.40
    P
    1.38
    ss
    1.34
    particularly
    1.32
    T
    1.30
    Activations Density 0.069%

    No Known Activations