© 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. 6193
    Prev
    Next
    INDEX
    Explanations

    various types

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

    The neuron strongly activates on mentions of extremist or militant groups and related conflict terms (e.g. “radical,” “Islamic,” specific groups and countries involved in Islamist conflicts”).

    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
    IRONMENT
    0.83
    収入
    0.80
    Я
    0.75
    Precio
    0.71
    úsica
    0.70
    上涨
    0.70
     Courthouse
    0.70
    Piano
    0.70
    ă
    0.69
     Biraz
    0.68
    POSITIVE LOGITS
    各大
    1.01
     various
    0.96
     соответствующих
    0.94
     các
    0.93
     các
    0.90
     artefacts
    0.86
    दारों
    0.84
    各个
    0.84
     respectivos
    0.84
    各国
    0.83
    Activations Density 0.778%

    No Known Activations