© 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. 104152
    Prev
    Next
    INDEX
    Explanations

    nth degreesdimensionalsquareorderconnections

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

    Looking at the activations, this neuron strongly activates on the dollar sign `$` followed by single letters (especially `p`, `i`, `n`, `R`, `N`, `V`, `I`) that appear to be mathematical variables in LaTeX notation. The neu

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron is primarily detecting numeric tokens (numbers or numeric values) in the 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_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
    .
    -2.69
    </strong>
    -2.38
    1
    -2.22
    2
    -2.14
    3
    -2.06
    7
    -2.00
    5
    -1.97
    0
    -1.96
     powinny
    -1.95
    Leírás
    -1.94
    POSITIVE LOGITS
     to
    2.75
     \
    2.14
    翃
    2.09
    轳
    2.06
    蓠
    2.03
    mpagne
    2.02
    UCTION
    2.02
    ichtete
    1.97
    dhism
    1.94
    憷
    1.94
    Activations Density 0.027%

    No Known Activations