© 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. 31-GEMMASCOPE-2-RES-16K
    7. 1705
    Prev
    Next
    INDEX
    Explanations

    happening after 'going' or 'interesting'

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

    The neuron strongly activates on question‐style expressions—especially the “what is/was … going on/happening” pattern.

    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
     keltiradi
    0.51
    ραι
    0.50
    BACH
    0.50
     ['*.
    0.49
     ['<%=
    0.49
     থাকবেন
    0.48
     spacerItem
    0.48
    ↩
    0.48
    як
    0.47
    უალ
    0.47
    POSITIVE LOGITS
     happening
    2.42
     happenings
    1.88
     проис
    1.77
     occurring
    1.70
     happened
    1.63
     aconte
    1.59
     occuring
    1.55
     terjadi
    1.48
    發生
    1.48
    发生了
    1.46
    Activations Density 0.309%

    No Known Activations