© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 1329
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on numeric tokens containing decimal points (e.g. “1528.5587,” “645.4302,” “1498.5872”), i.e. it detects decimal‐formatted numbers.

    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
    したり
    0.53
    াজে
    0.50
     слабы
    0.46
    atoti
    0.46
     ходе
    0.46
     vidéos
    0.46
    ᕇ
    0.46
     または
    0.46
    alarını
    0.45
    áveis
    0.44
    POSITIVE LOGITS
     municipality
    0.98
     Municipality
    0.82
     municipio
    0.79
     Province
    0.79
     county
    0.77
     County
    0.75
     province
    0.75
     city
    0.75
     şehr
    0.72
    縣
    0.72
    Activations Density 0.026%

    No Known Activations