© 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. 13253
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on unusual standalone numeric or numeric‐like tokens (e.g. long floats or stray 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
    ない
    0.86
     thờ
    0.75
    тек
    0.73
    なければ
    0.70
    陶
    0.68
    、「
    0.66
     okre
    0.66
    脚注
    0.63
     nasc
    0.63
    칭
    0.63
    POSITIVE LOGITS
     Иногда
    0.82
     гидро
    0.81
     годов
    0.81
     боль
    0.80
     испо
    0.80
     инструменты
    0.80
     девя
    0.80
    ádza
    0.79
     आयाम
    0.79
     Мы
    0.78
    Activations Density 0.000%

    No Known Activations