© 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. 14517
    Prev
    Next
    INDEX
    Explanations

    Java constructor arguments

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

    The neuron fires on constructor calls—that is, the opening parenthesis “(” immediately following a “new ClassName” expression.

    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
    </h2>
    -2.16
    </em>
    -1.80
     at
    -1.80
     E
    -1.74
     S
    -1.70
     as
    -1.68
     D
    -1.66
     still
    -1.65
    </h4>
    -1.63
     get
    -1.63
    POSITIVE LOGITS
     EDITORS
    1.71
     zapal
    1.66
     Ayuda
    1.66
    敺
    1.63
    こんな感じで
    1.62
     lorsqu
    1.60
     LECTURE
    1.59
    駸
    1.59
     rembourse
    1.59
     Eventos
    1.56
    Activations Density 0.023%

    No Known Activations