© 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. 13893
    Prev
    Next
    INDEX
    Explanations

    Strong emotional exclamations

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

    This neuron fires on phrasing that looks like a personal plea or request for help or information (e.g. “sir,” “please,” “my admission,” “Any one know…?”).

    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
     &/
    0.84
    /
    0.80
     (&
    0.78
     Thoreau
    0.75
     PowerPoint
    0.75
     René
    0.74
     Wright
    0.73
     Merton
    0.73
     Miller
    0.73
     Roger
    0.73
    POSITIVE LOGITS
    если
    1.09
    aaaaaaaa
    0.88
    😥
    0.88
    😭😭
    0.85
    😍😍
    0.85
    😘
    0.85
    Kalau
    0.84
    ！！！！
    0.83
    呵呵
    0.82
     думаю
    0.82
    Activations Density 0.074%

    No Known Activations