© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26289
    Prev
    Next
    INDEX
    Explanations

    initialisms and abbreviations

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

    The neuron fires on multi-character uppercase or capitalized tokens—e.g. brand names, acronyms, and proper‐noun abbreviations.

    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
     habían
    -1.02
     or
    -1.00
     prétend
    -0.96
     meningkat
    -0.96
     duidelijk
    -0.95
    婶
    -0.95
    niad
    -0.93
     and
    -0.93
     dám
    -0.91
     who
    -0.90
    POSITIVE LOGITS
    itarianism
    0.99
    Välkommen
    0.99
     susi
    0.95
    Приготовление
    0.95
    called
    0.91
    pflege
    0.91
    Heraus
    0.91
    cycles
    0.91
    caused
    0.90
     попыта
    0.88
    Activations Density 0.366%

    No Known Activations