© 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. 9378
    Prev
    Next
    INDEX
    Explanations

    "is" or "exists" in multiple languages

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

    The neuron selectively activates on capitalized labels for domains, sectors, fields, or organizations (e.g. “health sector,” “Paranormal field,” “CrossFit,” “Biologos,” “film”).

    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
     envió
    0.59
    ぶりに
    0.57
    躋
    0.56
     svoju
    0.55
    を発表
    0.54
     Trip
    0.54
     Prz
    0.54
    を迎
    0.53
    跻
    0.53
    気分
    0.53
    POSITIVE LOGITS
     существуют
    0.71
     есть
    0.70
     используются
    0.70
     অবস্থিত
    0.66
     vorhanden
    0.66
     где
    0.62
     существует
    0.61
    есть
    0.61
     имеются
    0.60
     нет
    0.58
    Activations Density 0.204%

    No Known Activations