© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 40290
    Prev
    Next
    INDEX
    Explanations

    conjunction followed by clause starter

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

    The neuron primarily activates on the coordinating conjunction “and.”

    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
     daños
    -1.13
     publicités
    -1.13
     perles
    -1.03
     inſ
    -1.02
     egyéb
    -0.97
    澪
    -0.97
    perfectly
    -0.97
    ρθ
    -0.96
     myndig
    -0.96
    ettür
    -0.94
    POSITIVE LOGITS
     that
    1.50
     by
    1.23
     to
    1.20
     this
    1.16
     with
    1.16
     certainly
    1.15
     for
    1.10
     starts
    1.05
     at
    1.05
     seems
    1.00
    Activations Density 0.021%

    No Known Activations