© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 10995
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on action‐oriented verbs that introduce or signal an operation or shift in the discourse (e.g. “focuses,” “typing,” “let”).

    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
    ebo
    0.79
    বিদ্যা
    0.75
    뀌
    0.75
    ểu
    0.74
     대해서
    0.74
     eğer
    0.74
    માં
    0.71
    ंची
    0.71
    📶
    0.71
    rinde
    0.71
    POSITIVE LOGITS
    ј
    0.80
     политика
    0.73
     করিত
    0.71
     statistic
    0.71
    धाना
    0.71
     parlor
    0.70
     patriotic
    0.68
     wearer
    0.68
     surpasses
    0.67
     speculator
    0.66
    Activations Density 0.000%

    No Known Activations