© 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. 8755
    Prev
    Next
    INDEX
    Explanations

    broken down

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

    The neuron fires on content words that are action verbs (especially strong, dynamic verbs in past, present‐participle, or base form).

    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
    !
    0.71
     curto
    0.69
     medit
    0.66
    !“
    0.65
     langen
    0.65
     curta
    0.64
     muchachos
    0.64
    いろんな
    0.63
    .!
    0.62
     كده
    0.61
    POSITIVE LOGITS
    롭게
    0.72
    heartedly
    0.69
    ρίζ
    0.68
     Anschließend
    0.66
    б
    0.64
     हादसे
    0.63
    uje
    0.63
    izuje
    0.63
    SUFFIX
    0.62
    izes
    0.61
    Activations Density 0.663%

    No Known Activations