© 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. 54518
    Prev
    Next
    INDEX
    Explanations

    connecting clauses

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

    The neuron is selectively detecting 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
    が
    -1.32
    </em>
    -1.21
     mivel
    -1.20
    <bos>
    -1.14
    '
    -1.09
    公立
    -1.09
    は
    -1.09
    lankton
    -1.08
    近年
    -1.07
     dichotomy
    -1.07
    POSITIVE LOGITS
    為に
    1.64
    vább
    1.52
    corrhi
    1.48
    筈
    1.43
    ícias
    1.42
    を探す
    1.41
     krijg
    1.40
     then
    1.40
    久し
    1.38
     maakte
    1.38
    Activations Density 0.169%

    No Known Activations