© 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. 12739
    Prev
    Next
    INDEX
    Explanations

    key-value pairs and programming constructs

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

    The neuron strongly activates on code constructs that iterate over dictionary key–value pairs (e.g. “foreach(var item in dictionary)” or “for k, v in d.items()”).

    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
     już
    -1.23
     already
    -1.20
     to
    -1.15
    énération
    -1.12
     У
    -1.07
     potrafi
    -0.95
     уже
    -0.94
     αλλά
    -0.93
     или
    -0.93
     Процитовано
    -0.91
    POSITIVE LOGITS
    Etimología
    1.27
     beca
    1.20
    Татьяна
    1.09
    所谓的
    1.04
    Ṫ
    1.03
    Наталья
    1.02
    ributing
    1.02
    Aceptar
    1.00
     raras
    1.00
     becau
    0.99
    Activations Density 0.004%

    No Known Activations