© 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. 10-GEMMASCOPE-RES-131K
    7. 105799
    Prev
    Next
    INDEX
    Explanations

    chap, pamph, jap, caps

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

    The neuron strongly activates on the fragmented tokens of the word “chapbook,” i.e. the sequence “chap” followed by “book.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    pAd
    -3.09
     legislación
    -2.94
    聍
    -2.81
    ⣞
    -2.78
    霂
    -2.77
    lgari
    -2.72
    xnn
    -2.72
     pomocí
    -2.69
     ciudadanía
    -2.67
    诹
    -2.67
    POSITIVE LOGITS
    g
    3.72
    (
    3.67
    al
    3.61
    7
    3.58
    In
    3.58
    x
    3.52
    9
    3.48
    You
    3.45
    h
    3.42
    [
    3.41
    Activations Density 0.001%

    No Known Activations