© 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. 12228
    Prev
    Next
    INDEX
    Explanations

    numbers and punctuation

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

    The neuron’s primary function is to detect numeric sequences (e.g. years, volume/issue numbers, page ranges) in citation entries.

    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
     కార్యక్ర
    0.61
    èrement
    0.60
    ových
    0.59
     =>{
    0.58
    urro
    0.58
     mediums
    0.57
    ów
    0.57
    arelli
    0.56
     actives
    0.56
    íamos
    0.55
    POSITIVE LOGITS
     doi
    0.70
    doi
    0.66
    张
    0.58
    unpublished
    0.56
    ARTICLE
    0.56
    শিদ
    0.55
    Comput
    0.54
     Indien
    0.54
     Gera
    0.54
     OCLC
    0.54
    Activations Density 0.065%

    No Known Activations