© 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. 128581
    Prev
    Next
    INDEX
    Explanations

    The neuron detects sentence‐ending punctuation (periods, question marks, etc.) that mark the ends of sentences.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    period followed by specific words

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     insofar
    -1.17
     außerdem
    -1.12
     exponer
    -1.08
    σιο
    -1.08
     variously
    -1.07
     unseres
    -1.06
    くなって
    -1.02
    μος
    -1.02
    就已经
    -1.02
    stücke
    -1.01
    POSITIVE LOGITS
     somit
    1.03
     females
    1.02
     there
    1.02
     predomin
    1.01
    ekk
    1.00
     defensa
    0.99
    [{
    0.98
    revet
    0.97
     Majority
    0.95
    次は
    0.94
    Activations Density 0.067%

    No Known Activations