© 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. 71826
    Prev
    Next
    INDEX
    Explanations

    testing code libraries

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

    This neuron activates on numeric literals in code (e.g., numbers in imports, version strings, or numeric constants).

    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
     on
    -1.56
     now
    -1.34
     to
    -1.27
     r
    -1.13
     incluye
    -1.08
     $
    -1.04
     itself
    -1.04
     in
    -1.01
     already
    -1.01
     were
    -0.99
    POSITIVE LOGITS
     prostitu
    1.30
     majest
    1.27
     aparat
    1.27
    ಝ
    1.24
    Komentar
    1.24
     relatie
    1.23
     automobili
    1.22
     enthou
    1.22
     parlement
    1.22
     inaug
    1.21
    Activations Density 0.008%

    No Known Activations