© 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. 10885
    Prev
    Next
    INDEX
    Explanations

    programming variable declarations

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

    This neuron detects tokens that introduce or declare variables or types in code (e.g., float, local, List, string, Intent, Dim, var).

    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
     canadensis
    -1.13
    堰
    -1.11
    fordshire
    -1.07
    ㎎
    -1.05
    chuhe
    -1.05
    😵
    -1.05
    signia
    -1.02
    申し上げます
    -1.00
    时间的
    -0.98
    rester
    -0.98
    POSITIVE LOGITS
     another
    1.53
     homicidio
    1.18
     spéciaux
    1.13
     soluções
    1.12
     further
    1.11
    って何
    1.08
    Sì
    1.08
    另一
    1.06
    Allora
    1.05
     síntesis
    1.05
    Activations Density 0.008%

    No Known Activations