© 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. 84725
    Prev
    Next
    INDEX
    Explanations

    institution acronyms

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

    The neuron specializes in spotting sequences of uppercase letters—i.e. acronyms or organization 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_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
     so
    -1.25
     as
    -1.19
     Moreover
    -1.12
     even
    -1.12
     where
    -1.11
     out
    -1.09
     about
    -1.09
     for
    -1.09
     this
    -1.07
     only
    -1.07
    POSITIVE LOGITS
     SUCH
    1.46
     éxitos
    1.41
     takie
    1.38
     surtido
    1.34
    such
    1.31
    Szer
    1.24
     такая
    1.19
    probleme
    1.18
    תיים
    1.16
     sesiones
    1.16
    Activations Density 0.431%

    No Known Activations