© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 79572
    Prev
    Next
    INDEX
    Explanations

    arrangements, logistics, explained

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

    The neuron is keyed to punctuation and other strong boundary markers—especially sentence‐final periods and commas—i.e. it signals when a sentence or clause boundary is reached.

    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
     where
    -1.16
     especially
    -1.13
     in
    -1.13
     joins
    -0.91
    뿐
    -0.90
     Reihenfolge
    -0.88
    
    -0.86
    ан
    -0.85
    त्व
    -0.85
    !
    -0.85
    POSITIVE LOGITS
     arrangements
    1.27
     logistics
    1.16
    安排
    1.13
     arrangement
    1.13
     Vieira
    1.09
    hinter
    1.09
    udesta
    1.08
    explained
    1.06
     histoire
    1.05
     zwi
    1.05
    Activations Density 0.020%

    No Known Activations