© 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. 106986
    Prev
    Next
    INDEX
    Explanations

    sentence terminators followed by conjunctions

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

    The neuron activates strongly on colon characters (":"), marking punctuation that introduces or separates clauses or sections.

    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
     poft
    -1.05
    嗽
    -1.02
    fährt
    -1.02
    ansicht
    -1.01
     kwe
    -1.01
     opinie
    -0.99
     francus
    -0.98
     châssis
    -0.98
     souverain
    -0.98
     kaupung
    -0.98
    POSITIVE LOGITS
     It
    2.81
     There
    2.28
     This
    2.28
     The
    2.13
     If
    1.91
     In
    1.80
     At
    1.80
     Although
    1.77
     We
    1.77
     Some
    1.74
    Activations Density 0.098%

    No Known Activations