© 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. 12731
    Prev
    Next
    INDEX
    Explanations

    and then connecting words

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

    The neuron fires strongly on the coordinating conjunction “and,” especially when it links phrases or clauses.

    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
    a
    -1.38
    C
    -1.26
    R
    -1.20
    L
    -1.19
    else
    -1.17
    H
    -1.13
    (\
    -1.13
    то
    -1.12
    M
    -1.12
    com
    -1.10
    POSITIVE LOGITS
     then
    1.43
    ようで
    1.39
     beoor
    1.30
     різних
    1.30
     jeśli
    1.30
     categorias
    1.27
     ſec
    1.26
    moteur
    1.26
    を探す
    1.26
     aprende
    1.24
    Activations Density 0.067%

    No Known Activations