© 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. 65907
    Prev
    Next
    INDEX
    Explanations

    interpunctuation and sentence endings

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

    The neuron responds to emphatic, high‐intensity markers—especially long runs of punctuation (!!!, …), parentheses, and similar tokens—signaling an excited or emotionally charged tone.

    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
     while
    -1.61
     there
    -1.27
     when
    -1.13
     their
    -1.10
     during
    -1.05
     or
    -1.03
     some
    -0.97
     Attachments
    -0.97
     other
    -0.94
     berlina
    -0.92
    POSITIVE LOGITS
     colorear
    1.07
    signable
    1.05
    mostly
    1.05
     armazenamento
    1.02
    cible
    1.01
     reduzir
    1.01
    mainly
    1.00
    ynchronously
    0.99
    )$;
    0.98
    crimination
    0.95
    Activations Density 0.036%

    No Known Activations