© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8966
    Prev
    Next
    INDEX
    Explanations

    legal counsel and advice

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

    The neuron fires on words referring to organized academic, professional, or social programming and support initiatives (e.g. “academic,” “social,” “programming,” “support,” “law,” “school,” “navigate,” “journey”).

    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_31_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
     Vegeta
    0.59
     Steam
    0.59
     steam
    0.57
     насы
    0.54
     Serum
    0.54
     chromosphere
    0.54
     serum
    0.53
    ග
    0.53
     sphere
    0.53
    STEAM
    0.53
    POSITIVE LOGITS
     lawyers
    2.42
     lawyer
    2.38
     Lawyers
    2.36
     Lawyer
    2.36
    律师
    2.25
     attorneys
    2.16
     Attorneys
    2.09
     attorney
    2.03
    Legal
    2.03
     Legal
    2.02
    Activations Density 0.593%

    No Known Activations