© 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. 16259
    Prev
    Next
    INDEX
    Explanations

    consent and sensitive topics

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

    The neuron preferentially activates on relatively rare, content‐bearing tokens (specialized nouns or verbs) rather than on common function words.

    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
    𝟖
    0.90
    Warum
    0.86
     Warum
    0.85
    
    0.80
     enn
    0.79
    Karena
    0.79
     Sekarang
    0.79
     Bitte
    0.78
    ➠
    0.78
    熵
    0.78
    POSITIVE LOGITS
     receivables
    0.77
    ırmaya
    0.75
     úd
    0.74
    ாட்டு
    0.73
     polici
    0.71
     यही
    0.70
    okedex
    0.70
     indebtedness
    0.69
     coffers
    0.69
     জম
    0.68
    Activations Density 0.001%

    No Known Activations