© 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. 8618
    Prev
    Next
    INDEX
    Explanations

    past hurt or issues

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

    The neuron fires strongly on content‐bearing (often infrequent) words—i.e. it picks out relatively rare “content” vocabulary 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
     дать
    0.77
    ⌄
    0.76
     Работа
    0.74
     Соб
    0.73
     ито
    0.71
    UIL
    0.71
    TOR
    0.71
     дали
    0.71
    SAN
    0.70
    Td
    0.70
    POSITIVE LOGITS
    a
    0.91
    ed
    0.88
     circumference
    0.86
     bulging
    0.83
     misunder
    0.82
     sedative
    0.81
     canoeing
    0.79
     invitation
    0.79
     panoramic
    0.79
     jackpot
    0.79
    Activations Density 0.000%

    No Known Activations