© 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. 11971
    Prev
    Next
    INDEX
    Explanations

    magical power and abilities

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

    The neuron fires strongly on words naming weapons, armor, or other combat‐related gear.

    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
     “
    0.94
     also
    0.91
     distributional
    0.89
     says
    0.86
     aussi
    0.86
    睬
    0.85
     we
    0.84
     Facebook
    0.84
    0.83
     ‘
    0.82
    POSITIVE LOGITS
     criaturas
    1.19
     teleport
    1.16
     hắn
    1.04
    ⚔
    1.04
     causando
    1.02
     combate
    1.01
     undead
    0.99
     verhindern
    0.99
     rune
    0.98
    へと
    0.96
    Activations Density 0.577%

    No Known Activations