© 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. 105797
    Prev
    Next
    INDEX
    Explanations

    read the full article

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

    The neuron activates on numeric tokens—anything forming part of numbers, measurements, or other digit‐based figures.

    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
    姽
    -1.64
     paſſ
    -1.57
     diriger
    -1.54
     faciliter
    -1.52
     alimenta
    -1.50
    ianas
    -1.49
    儇
    -1.48
    ਗ਼
    -1.48
     Eſ
    -1.48
    similar
    -1.45
    POSITIVE LOGITS
     may
    1.84
     that
    1.80
     at
    1.68
    How
    1.59
    :
    1.58
    Some
    1.55
    .(
    1.52
     until
    1.48
     which
    1.46
     have
    1.42
    Activations Density 0.021%

    No Known Activations