© 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. 2532
    Prev
    Next
    INDEX
    Explanations

    colorful, stylish descriptions

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

    The neuron is detecting descriptive adjectives that qualify or emphasize the attributes of something.

    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
    ́
    0.73
    에는
    0.71
    와
    0.70
    ̀
    0.66
    ̆
    0.66
    ̉
    0.63
     $+\
    0.63
    َ
    0.63
    には
    0.61
    까지
    0.59
    POSITIVE LOGITS
     porém
    0.70
     тобто
    0.68
     albeit
    0.67
     mutta
    0.67
    然后在
    0.65
     მაგრამ
    0.63
     voire
    0.61
     pero
    0.61
    albeit
    0.61
     bersih
    0.60
    Activations Density 0.181%

    No Known Activations