© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 3256
    Prev
    Next
    INDEX
    Explanations

    important

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

    The neuron activates on adversative discourse markers—specifically the conjunction “But” signaling a contrast or shift.

    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
     certainly
    0.86
     rarely
    0.86
     admittedly
    0.86
     clearly
    0.76
     대부분
    0.76
     oftentimes
    0.76
     obviously
    0.75
     Certainly
    0.74
     sicuramente
    0.72
    clearly
    0.72
    POSITIVE LOGITS
     WAIT
    0.84
     trotz
    0.84
     말고
    0.84
     wait
    0.83
     meski
    0.81
    Jangan
    0.79
    Wait
    0.78
     především
    0.77
    wait
    0.74
    為什麼
    0.71
    Activations Density 0.154%

    No Known Activations