© 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. 10886
    Prev
    Next
    INDEX
    Explanations

    numbers in the thousands or millions

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

    The neuron activates on numeric tokens (integer constants) in code or text.

    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
     the
    -2.66
    をしました
    -2.20
     brinquedo
    -2.14
    うわ
    -2.14
    を行います
    -2.11
    に来て
    -2.09
    凄く
    -2.05
    赑
    -2.05
    所で
    -2.03
    を行いました
    -2.03
    POSITIVE LOGITS
    0
    2.98
    —
    2.00
    er
    1.95
    в
    1.87
    <sup>
    1.79
     (
    1.77
    だが
    1.75
    ”
    1.72
     of
    1.71
     —
    1.64
    Activations Density 0.014%

    No Known Activations