© 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. 31-GEMMASCOPE-2-RES-16K
    7. 10889
    Prev
    Next
    INDEX
    Explanations

    creative writing, blog posts, concise syntax

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

    The neuron responds to descriptive adjectives that highlight appealing qualities or sensory attributes (e.g. “green,” “rolling,” “vibrant,” “plush,” “friendly”).

    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_31_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
    埸
    0.79
     ইংরাজ
    0.70
    ritional
    0.68
     chromatographic
    0.67
     interment
    0.65
     ವಿಶೇಷ
    0.65
     sofistic
    0.64
    দিগের
    0.64
    umbnail
    0.62
     ಕೂದಲ
    0.62
    POSITIVE LOGITS
    ​
    0.68
    ​​​​
    0.66
     लिखकर
    0.64
     để
    0.63
    â
    0.63
    [/
    0.62
    ,​
    0.62
    ​​
    0.61
    https
    0.59
     allowing
    0.59
    Activations Density 0.184%

    No Known Activations