© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 72657
    Prev
    Next
    INDEX
    Explanations

    mathematical sequences or variables

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

    The neuron activates whenever it encounters tokens containing digits—that is, it’s detecting numeric values.

    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
     unparalleled
    -2.98
     while
    -2.88
     although
    -2.64
     unrelenting
    -2.55
     allows
    -2.55
     supremely
    -2.52
     superbly
    -2.50
     enhances
    -2.47
     every
    -2.44
     seven
    -2.44
    POSITIVE LOGITS
    厹
    2.34
     spearheaded
    2.34
    殽
    2.31
    熜
    2.28
     Dinas
    2.23
    sometimes
    2.23
    probably
    2.23
    often
    2.22
     dando
    2.20
    several
    2.20
    Activations Density 0.020%

    No Known Activations