© 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. 31-GEMMASCOPE-2-RES-16K
    7. 6538
    Prev
    Next
    INDEX
    Explanations

    numbers, months, dates

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

    The neuron selectively activates on numeric tokens—especially multipart numbers in citations or references (years, volume/issue numbers, page ranges, 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_31_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
     ২০
    0.59
    <unused82>
    0.57
     Sixteen
    0.57
    ৬৮
    0.57
    ৪২
    0.56
     ஆராய்ச்ச
    0.55
     इंटेलिज
    0.54
     ৩৬
    0.54
    ٢
    0.54
     سكس
    0.53
    POSITIVE LOGITS
    ୯
    0.59
     neun
    0.58
    9
    0.57
     vijf
    0.56
    奇
    0.56
    
    0.55
    odin
    0.55
    irregular
    0.55
     nine
    0.54
     solit
    0.54
    Activations Density 0.150%

    No Known Activations