© 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. 22-GEMMASCOPE-RES-131K
    7. 130127
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on numerals or numeric tokens (e.g. counts, measurements, or figures).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    and followed by a number or specific word

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     nogal
    -0.86
     manier
    -0.86
    🗒
    -0.84
     maillot
    -0.83
    бие
    -0.83
    ншот
    -0.82
    IRQ
    -0.82
    LAM
    -0.81
    rather
    -0.80
     SOA
    -0.79
    POSITIVE LOGITS
     vairāk
    0.84
     including
    0.82
     pozycji
    0.82
     savu
    0.78
     virgen
    0.75
    tispiece
    0.74
     Visitors
    0.74
    records
    0.73
     two
    0.73
     leaving
    0.73
    Activations Density 0.070%

    No Known Activations