© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 14646
    Prev
    Next
    INDEX
    Explanations

    ( punctuation) whipped cream

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

    The neuron fires strongly on long numeric strings (and surrounding “/” separators) that look like IDs—e.g. citation numbers, DOIs, PMIDs, Amazon product codes, etc.

    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
     cuyas
    0.85
     ceased
    0.84
     katalog
    0.83
    <0x80>
    0.81
     termasuk
    0.81
     dejó
    0.81
     día
    0.80
     hasta
    0.80
     channeled
    0.79
    ే
    0.79
    POSITIVE LOGITS
    Ꮭ
    0.73
    这
    0.71
    そんな
    0.68
     நிச்சயம்
    0.67
    𝙲
    0.67
    Η
    0.67
     Courtesy
    0.66
     SURE
    0.66
     apprent
    0.65
    众人
    0.65
    Activations Density 0.000%

    No Known Activations