© 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. 35352
    Prev
    Next
    INDEX
    Explanations

    technical terms, nouns

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

    The neuron fires on code tokens naming core domain entities or resources (e.g. “contact(s)”, “post”, “task”, “module”), i.e. object/class names in programming contexts.

    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
     borracha
    -1.64
    corrhi
    -1.62
     leaſt
    -1.58
     chargeur
    -1.53
    に似
    -1.53
     dessinée
    -1.52
    lji
    -1.51
    cocina
    -1.50
     muñecas
    -1.48
    consejos
    -1.48
    POSITIVE LOGITS
     in
    2.06
     is
    1.91
     now
    1.71
     on
    1.63
     with
    1.63
     by
    1.57
     to
    1.54
     can
    1.51
     has
    1.50
     from
    1.45
    Activations Density 0.095%

    No Known Activations