© 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. 34-GEMMASCOPE-RES-131K
    7. 3282
    Prev
    Next
    INDEX
    Explanations

    sentence endings, pauses, or direct speech

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

    The neuron reliably activates on punctuation marks (especially periods, commas, and quotation marks), effectively spotting sentence and dialogue boundaries.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    aka
    -1.52
     онлайн
    -1.43
     aka
    -1.22
    Spoiler
    -1.16
     également
    -1.11
     messaging
    -1.10
    meh
    -1.09
     quinoa
    -1.09
    Jika
    -1.07
     sự
    -1.04
    POSITIVE LOGITS
     now
    1.34
     well
    1.16
     evidently
    1.15
     after
    1.12
     certainly
    1.11
     that
    1.07
     \&
    1.05
     frankly
    1.04
     when
    1.04
     primarily
    1.03
    Activations Density 0.038%

    No Known Activations