© 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. 13539
    Prev
    Next
    INDEX
    Explanations

    model

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

    The neuron is strongly activated by question‐mark tokens (i.e. it flags punctuation that marks questions).

    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
     BB
    0.60
    BB
    0.57
     李
    0.53
    李
    0.52
     «
    0.51
     Capital
    0.51
    E
    0.51
    {
    0.49
     長
    0.48
    W
    0.48
    POSITIVE LOGITS
    争取
    0.64
     zorgen
    0.61
    loggedin
    0.61
     conflitto
    0.61
    衝突
    0.61
     melden
    0.60
    ーー
    0.59
     controllo
    0.59
     geographies
    0.58
    чения
    0.58
    Activations Density 0.006%

    No Known Activations