© 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. 73613
    Prev
    Next
    INDEX
    Explanations

    Draw followed by numbers

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

    The neuron activates on numeric tokens—especially dates, years, and other multi‐digit number sequences.

    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
     flesta
    -1.45
     också
    -1.45
     profonde
    -1.30
     illustre
    -1.27
     discuter
    -1.25
     grises
    -1.24
    也会
    -1.23
     élas
    -1.21
     velg
    -1.20
     damen
    -1.20
    POSITIVE LOGITS
     et
    1.04
     Anreg
    1.03
    ó
    1.02
     finally
    1.01
    ur
    1.00
     June
    1.00
     закона
    0.98
     T
    0.98
     lavorato
    0.97
    向了
    0.96
    Activations Density 0.014%

    No Known Activations