© 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. 7631
    Prev
    Next
    INDEX
    Explanations

    Stable Diffusion, neural networks, gender identity

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

    The neuron selectively activates on capitalized tokens—proper nouns, acronyms, and other named or technical terms.

    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
     и
    0.67
     없고
    0.65
     और
    0.63
     και
    0.63
     এবং
    0.59
     و
    0.59
    और
    0.58
     torno
    0.56
     và
    0.55
     અને
    0.55
    POSITIVE LOGITS
     respectively
    0.62
     ஆகியவற்ற
    0.59
     alike
    0.59
     ஆகிய
    0.57
     sebagainya
    0.57
    .​​
    0.57
     \,.
    0.56
     respectivamente
    0.55
    !).
    0.53
     പ്രത്യേ
    0.53
    Activations Density 0.001%

    No Known Activations