© 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. 31-GEMMASCOPE-2-RES-16K
    7. 8260
    Prev
    Next
    INDEX
    Explanations

    for everyone, everybody, universal

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

    The neuron strongly activates on abstract, high-level terms expressing societal values and rights—words like “freedom,” “global,” “democratic,” “accessible,” “equity,” and similar concepts.

    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
     عاي
    0.56
    慣
    0.53
     exhale
    0.52
     Dorey
    0.51
    绾
    0.49
    ರೇ
    0.48
    移
    0.46
    醒
    0.46
    firestore
    0.46
    枣
    0.46
    POSITIVE LOGITS
     everyone
    2.16
     everybody
    2.14
     Everyone
    2.11
    everyone
    2.05
     iedereen
    2.03
     Everybody
    2.02
    Everyone
    2.00
     herkese
    2.00
     herkes
    1.96
    Everybody
    1.95
    Activations Density 0.475%

    No Known Activations