© 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. 13853
    Prev
    Next
    INDEX
    Explanations

    inclusion criteria; temporary; estimate

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

    The neuron strongly responds to numeric tokens—especially multi‐digit numbers and decimal values—i.e. it’s detecting real‐number or statistic expressions 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
    '
    1.00
    鼗
    0.87
     Diarsipkan
    0.85
     Hardly
    0.81
    торой
    0.78
     ocorr
    0.76
    JDK
    0.75
     tuyển
    0.75
     Pemer
    0.74
    }']
    0.74
    POSITIVE LOGITS
     برا
    0.94
     الت
    0.75
     nombres
    0.75
     foss
    0.75
    آ
    0.74
    يات
    0.73
     principales
    0.73
     fach
    0.73
     stitched
    0.71
     during
    0.71
    Activations Density 0.000%

    No Known Activations