© 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. 6045
    Prev
    Next
    INDEX
    Explanations

    grooming, stats, tended, starts, body

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

    The neuron responds to uncommon, low-frequency tokens (i.e. rare content words rather than common function words).

    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
    mpi
    0.91
    м
    0.85
    م
    0.84
    ifying
    0.82
    ومي
    0.80
    m
    0.78
    ic
    0.75
    ni
    0.74
    ere
    0.73
     aching
    0.72
    POSITIVE LOGITS
     Watercolor
    0.88
    捒
    0.84
    фесси
    0.83
     CHALL
    0.81
     Edin
    0.80
     Edna
    0.79
     EGR
    0.79
     Vans
    0.77
     Баскетбол
    0.77
     Aroma
    0.77
    Activations Density 0.000%

    No Known Activations