© 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. 10-GEMMASCOPE-RES-131K
    7. 104435
    Prev
    Next
    INDEX
    Explanations

    end of clause separator

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

    The neuron strongly activates on punctuation tokens (commas, periods, colons, semicolons, ellipses, emoticon separators, etc.), i.e. it detects punctuation marks.

    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
    .......
    -2.25
     undoubtedly
    -2.25
    e
    -2.25
     &
    -2.05
     були
    -2.02
     =
    -1.95
     indicates
    -1.94
    ....
    -1.92
    Going
    -1.88
    There
    -1.86
    POSITIVE LOGITS
    苤
    2.44
    archiwizowane
    2.36
     molto
    2.23
    矬
    2.14
    März
    2.14
    簠
    2.08
    蚨
    2.06
    鶫
    2.05
     avec
    2.03
     molti
    2.03
    Activations Density 0.025%

    No Known Activations