© 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. 10266
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on standalone numeric tokens—especially floating‐point numbers—whenever it encounters explicit decimal or measurement values 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
    蠡
    0.79
    viz
    0.67
     Фор
    0.67
    buf
    0.66
    note
    0.64
    dtypes
    0.64
    format
    0.64
    ൫
    0.63
    形式
    0.63
    eux
    0.63
    POSITIVE LOGITS
     civilian
    0.82
     perturbations
    0.77
     пришлось
    0.77
     прогнозы
    0.76
     corazón
    0.75
     harapan
    0.74
     siguió
    0.73
     kämp
    0.73
     volle
    0.73
     timeliness
    0.72
    Activations Density 0.001%

    No Known Activations