© 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. 7670
    Prev
    Next
    INDEX
    Explanations

    when followed by action verbs

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

    The neuron responds most strongly to discourse-linking words that introduce conditions or clauses (e.g. “when,” “it,” “comes,” “comfortable”).

    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
    聽到
    0.73
    cj
    0.71
    pé
    0.69
    f
    0.68
     heures
    0.67
     পায়ে
    0.67
     ambayo
    0.66
    အ
    0.66
     jednego
    0.66
     конца
    0.65
    POSITIVE LOGITS
    面对
    0.80
     dealing
    0.78
    Compared
    0.77
    compared
    0.69
     విషయంలో
    0.69
     confront
    0.67
     compared
    0.65
     ánimo
    0.65
     regard
    0.64
    Account
    0.64
    Activations Density 0.075%

    No Known Activations