© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 5315
    Prev
    Next
    INDEX
    Explanations

    numbers in calculations

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

    The neuron strongly responds to numeric tokens (e.g. years, measurements, numerical citations) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    f
    0.72
    u
    0.72
    prev
    0.71
    re
    0.67
    !-
    0.65
    Explain
    0.64
    us
    0.64
    PE
    0.64
    ПО
    0.64
    v
    0.63
    POSITIVE LOGITS
     negozi
    0.77
     caffè
    0.77
     estilos
    0.75
     styles
    0.73
     accountant
    0.73
     alliances
    0.73
     snee
    0.73
     présenté
    0.71
     stile
    0.70
     gimm
    0.70
    Activations Density 0.001%

    No Known Activations