© 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. 8378
    Prev
    Next
    INDEX
    Explanations

    harm, loss, or death

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

    The neuron fires on semantically heavy, content-bearing words (not common function words) that tend to carry the main topics or key concepts of a passage.

    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
    uyorum
    0.77
     attualmente
    0.75
    פ
    0.75
     můžete
    0.71
    currently
    0.70
    о
    0.70
    Walmart
    0.70
    z
    0.70
     सध्या
    0.69
     있는데요
    0.69
    POSITIVE LOGITS
     perished
    1.00
     করেছিল
    0.93
     Opfer
    0.89
     Fatal
    0.83
     succumbed
    0.82
     świę
    0.79
     victime
    0.78
     столь
    0.77
     furent
    0.76
     Symbol
    0.75
    Activations Density 0.103%

    No Known Activations