© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 5491
    Prev
    Next
    INDEX
    Explanations

    objects; MOVE; bowl

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

    The neuron activates primarily on numerical content—digits, measurements, dates, quantities, or other figure‐based tokens.

    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
     AIM
    0.45
    参观
    0.45
     slits
    0.45
    通过
    0.44
     ellipt
    0.44
     egos
    0.44
     harem
    0.43
    휀
    0.43
     endast
    0.43
    
    0.42
    POSITIVE LOGITS
    탰
    0.44
    )}`;
    0.44
    sequent
    0.44
     সেনারা
    0.43
    ην
    0.43
    getahuan
    0.43
    ansive
    0.43
    <eos>
    0.42
    点を
    0.42
    пы
    0.41
    Activations Density 0.139%

    No Known Activations