© 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. 13679
    Prev
    Next
    INDEX
    Explanations

    (and connecting phrases)

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

    The neuron fires on the conjunction “and” when it appears immediately before a parenthesized or inline numeric count (e.g. “(and 968 samples)”).

    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
     gustado
    -2.13
    🫃
    -2.08
     både
    -2.06
     ſou
    -2.03
    熜
    -1.99
     間取り
    -1.96
     ホルダー
    -1.93
    archiwizowane
    -1.89
     mikrofon
    -1.87
    
    -1.85
    POSITIVE LOGITS
     possibly
    2.66
    This
    2.44
     then
    2.41
     consequently
    2.31
     even
    2.27
     hence
    2.22
     thus
    2.20
     perhaps
    2.14
     more
    2.14
     for
    2.14
    Activations Density 0.014%

    No Known Activations