© 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. 25316
    Prev
    Next
    INDEX
    Explanations

    initials and names

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

    The neuron responds to isolated uppercase single letters (e.g. initials or single‐letter citations) in bibliographic or legal reference lists.

    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
     on
    -2.33
     at
    -2.30
     from
    -2.27
     in
    -2.05
     all
    -1.96
     for
    -1.91
     not
    -1.89
     without
    -1.88
     before
    -1.86
     one
    -1.78
    POSITIVE LOGITS
    küche
    1.63
     faciliter
    1.58
     quintessential
    1.48
     flamboyant
    1.47
     Dinas
    1.47
     horrid
    1.44
     corriger
    1.42
     overal
    1.42
    などを
    1.42
     ſie
    1.42
    Activations Density 0.003%

    No Known Activations