© 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. 10340
    Prev
    Next
    INDEX
    Explanations

    unique or strange descriptions

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

    This neuron activates on subjective evaluative adjectives or adverbs that signal something unusual, surprising, or noteworthy (e.g. “strange,” “weird,” “interesting,” “curiously,” “amazing”).

    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
    mathscr
    0.79
    きましょう
    0.75
    ლ
    0.68
     Bs
    0.68
     EARL
    0.68
     הבי
    0.67
     যাবত
    0.67
    ❤️❤️
    0.66
    DependentUpon
    0.66
     ۸
    0.66
    POSITIVE LOGITS
     paradoxical
    0.89
    ان
    0.88
     strange
    0.87
    sville
    0.86
     bizarre
    0.85
    竟
    0.84
     встречается
    0.84
     abil
    0.83
     Remarkably
    0.83
     соотно
    0.83
    Activations Density 0.411%

    No Known Activations