© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 5453
    Prev
    Next
    INDEX
    Explanations

    mechanics, demand, description

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

    This neuron activates on the kind of formal, promotional or announcement‐style words you’d see in press releases or product blurbs (e.g. “states,” “report,” “free,” “inspired,” “access,” “old”).

    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
    Мі
    0.78
     обме
    0.78
    payer
    0.77
    payers
    0.76
    𝖾
    0.75
     Race
    0.73
    multip
    0.73
     Market
    0.73
    sim
    0.72
    是真的
    0.72
    POSITIVE LOGITS
     sluč
    0.81
    ように
    0.75
    숙
    0.71
            
    0.70
    anen
    0.67
    ؤال
    0.67
    ˳
    0.67
     väx
    0.66
    ås
    0.66
     tradizione
    0.66
    Activations Density 0.001%

    No Known Activations