© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 4180
    Prev
    Next
    INDEX
    Explanations

    requests for support or improvement

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

    the neuron detects terms referring to legal or governmental procedural actions and compliance (e.g. court orders, investigations, reviews, violations).

    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
     (=
    0.78
    +
    0.70
    においては
    0.68
    においても
    0.65
     ture
    0.63
    ='',
    0.62
     এরূপ
    0.61
    ="",
    0.61
     हूँ
    0.59
     quite
    0.59
    POSITIVE LOGITS
     murderous
    0.82
     bloodshed
    0.77
    ساعدة
    0.76
     warrantless
    0.76
     botched
    0.72
     killings
    0.68
     carnage
    0.68
     ousted
    0.68
     Obamacare
    0.68
     हत्या
    0.67
    Activations Density 0.014%

    No Known Activations