© 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. 21520
    Prev
    Next
    INDEX
    Explanations

    titles and labels

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

    The neuron fires on tokens consisting of multi‐letter uppercase sequences (e.g. acronyms or proper‐name fragments written in all caps).

    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
     撮っ
    -2.23
    ーーーー
    -1.96
    鄚
    -1.94
    庒
    -1.91
     خلال
    -1.91
    でしたが
    -1.91
     當
    -1.86
     土産
    -1.81
     italiens
    -1.80
    Дети
    -1.80
    POSITIVE LOGITS
     same
    2.63
     there
    2.20
     by
    2.17
     has
    2.09
    <em>
    2.00
     according
    1.99
    U
    1.96
    P
    1.95
     they
    1.95
     siempre
    1.86
    Activations Density 0.000%

    No Known Activations