© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 8994
    Prev
    Next
    INDEX
    Explanations

    image of Black tact children events

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

    The neuron primarily fires on content words—especially verbs and adverbs—while ignoring 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
     homoc
    0.82
    ない
    0.79
     glutamate
    0.76
     únicamente
    0.75
     dinámica
    0.74
    ,
    0.73
    だけでなく
    0.72
     xenon
    0.72
     приобрета
    0.72
    жи
    0.72
    POSITIVE LOGITS
    ائیگی
    0.77
    滕
    0.75
    fti
    0.75
    श्ती
    0.75
     capitalized
    0.74
    ndal
    0.72
     Content
    0.68
    ۔
    0.67
    ಂಡ್
    0.66
     जड़ों
    0.66
    Activations Density 0.000%

    No Known Activations