© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43100
    Prev
    Next
    INDEX
    Explanations

    will be

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

    This neuron selectively activates on quantitative content—numbers, percentages, and other numeric or count‐based expressions.

    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
     and
    -1.73
     From
    -1.47
     from
    -1.47
     in
    -1.38
     jepang
    -1.35
     at
    -1.32
    рона
    -1.28
     Offisielt
    -1.28
     fondu
    -1.24
    сные
    -1.23
    POSITIVE LOGITS
     its
    1.63
    goes
    1.45
    áreas
    1.39
    zdroj
    1.38
    ίη
    1.38
    características
    1.35
    especially
    1.35
    ばい
    1.34
     ihrer
    1.34
     strongest
    1.32
    Activations Density 0.101%

    No Known Activations