© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13387
    Prev
    Next
    INDEX
    Explanations

    programming language keywords and identifiers

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

    This neuron fires on programming-language type declarations and related syntax (e.g. class/variable types like HashMap, Node, int, double, array brackets, and generic “<…>” markers).

    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
     opět
    -1.17
     novamente
    -1.13
     “(
    -1.09
    着了
    -1.09
     Certaines
    -1.09
    Öl
    -1.09
     Aucune
    -1.08
     postIndex
    -1.08
     někter
    -1.07
    filme
    -1.05
    POSITIVE LOGITS
    %;
    1.27
     my
    1.27
     zelo
    1.18
    1.17
     nera
    1.15
     веке
    1.11
     Լ
    1.07
    
    1.07
     current
    1.06
     gamla
    1.06
    Activations Density 0.009%

    No Known Activations