© 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. 10-GEMMASCOPE-RES-131K
    7. 71322
    Prev
    Next
    INDEX
    Explanations

    pronunciation of acronyms

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

    The neuron strongly activates on uppercase acronyms and abbreviations (often appearing with parentheses or punctuation) in technical or scientific 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_10/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
     with
    -2.22
     from
    -1.97
     and
    -1.96
     his
    -1.80
     on
    -1.76
     its
    -1.73
     will
    -1.66
     at
    -1.65
     their
    -1.64
     all
    -1.61
    POSITIVE LOGITS
     relâche
    1.52
     CUCET
    1.48
     hés
    1.46
     interrom
    1.42
     maniere
    1.41
    Another
    1.39
     censura
    1.39
     rigide
    1.38
     multicolore
    1.38
    あなたが
    1.37
    Activations Density 0.276%

    No Known Activations