© 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. 9398
    Prev
    Next
    INDEX
    Explanations

    UI attributes and alignment

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

    The neuron fires strongly on tokens that are part of code or resource identifiers (e.g. variable/function names, CSS class names, file- or drawable-resource names, domain labels) rather than ordinary text words.

    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
     orna
    -1.18
     ananas
    -1.18
     poros
    -1.16
    niest
    -1.16
     automobili
    -1.16
     conster
    -1.14
    fotografie
    -1.14
     lavander
    -1.13
    caping
    -1.12
     expuls
    -1.12
    POSITIVE LOGITS
    ėjimas
    1.11
     Now
    1.00
     fortæller
    0.99
    可能會
    0.98
     llevará
    0.96
     заказать
    0.96
    Mr
    0.96
     Теперь
    0.95
    相片
    0.94
    .
    0.94
    Activations Density 0.001%

    No Known Activations