© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 3264
    Prev
    Next
    INDEX
    Explanations

    Dungeons and Dragons

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

    The neuron fires most strongly on the first “meaty” content word that opens a new section or paragraph—i.e. it marks the beginning of a new block of substantive text.

    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
    ä
    0.84
    ada
    0.78
     fermented
    0.77
    ੰਜ
    0.77
    ý
    0.73
    ovací
    0.71
     넣
    0.70
    服用
    0.70
     garlic
    0.69
     digested
    0.68
    POSITIVE LOGITS
    ال
    0.84
     Utara
    0.83
    難しい
    0.76
     podej
    0.76
     نهاية
    0.75
     ইতিমধ্যে
    0.74
    CAUSE
    0.71
    ството
    0.70
     fallacy
    0.70
     hitherto
    0.70
    Activations Density 0.000%

    No Known Activations