© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 58908
    Prev
    Next
    INDEX
    Explanations

    Java reflection code snippets

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

    This neuron fires on Java reflection–related type constructs, especially references to Type and its subinterfaces (e.g. GenericArrayType, ParameterizedType, TypeVariable) and imports from java.lang.reflect.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     will
    -1.88
    」
    -1.51
     continue
    -1.48
    ）
    -1.41
     ”
    -1.40
     continues
    -1.35
    ità
    -1.34
     स्थान
    -1.30
     ``
    -1.28
    愈
    -1.27
    POSITIVE LOGITS
    气质
    1.37
     triomphe
    1.32
    middag
    1.22
     koste
    1.22
    acakt
    1.21
     essais
    1.20
    ALIAN
    1.20
    (^_^
    1.19
     însă
    1.18
    Etape
    1.18
    Activations Density 0.314%

    No Known Activations