© 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. 27346
    Prev
    Next
    INDEX
    Explanations

    Objective-C, Java, Scala programming constructs

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

    The neuron fires strongly on the identifier “Serializable,” i.e. it detects mentions of the Serializable interface or keyword in code.

    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
    četně
    -1.13
    
    -1.06
     Поэтому
    -1.05
    pageIndex
    -0.96
     abbastanza
    -0.96
    creds
    -0.95
    选择的
    -0.92
     significantly
    -0.89
    嫱
    -0.88
    MUS
    -0.87
    POSITIVE LOGITS
    <bos>
    1.30
    plin
    1.03
    </h3>
    1.02
     igång
    1.01
    algebras
    1.00
    綿
    1.00
     ン
    0.99
    
    0.99
    攬
    0.99
    otu
    0.99
    Activations Density 0.001%

    No Known Activations