© 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. 22-GEMMASCOPE-RES-131K
    7. 129139
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on tokens containing numeric digits.

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

    numbers and punctuation

    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
     who
    -1.23
     you
    -1.13
     like
    -1.03
     saisi
    -1.02
     soyez
    -1.02
     dann
    -1.00
    ográfico
    -0.99
     alat
    -0.99
    anız
    -0.99
     whoſe
    -0.98
    POSITIVE LOGITS
     Unterscheidung
    1.24
     époux
    1.10
    both
    1.05
     orçamento
    1.02
     Anordnung
    1.01
     volontà
    1.00
     mehrm
    0.98
     dismant
    0.96
     dauern
    0.96
    accesso
    0.96
    Activations Density 0.000%

    No Known Activations