© 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. 14401
    Prev
    Next
    INDEX
    Explanations

    javascript library initialization

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

    The neuron strongly activates on parentheses punctuation—especially empty or simple “()” tokens as used in function definitions or calls.

    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
     Pfar
    -1.25
     gets
    -1.17
     края
    -1.14
     Grüßen
    -1.11
     vše
    -1.09
     androidx
    -1.07
     ਬ
    -1.05
    liothè
    -1.02
     comes
    -1.01
     says
    -0.98
    POSITIVE LOGITS
    This
    1.23
    _
    1.20
     måske
    1.14
    簓
    1.13
     menschliche
    1.12
     hilfre
    1.11
    .
    1.11
     colorear
    1.09
     ندار
    1.09
    would
    1.07
    Activations Density 0.008%

    No Known Activations