© 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. 24-GEMMASCOPE-2-RES-16K
    7. 837
    Prev
    Next
    INDEX
    Explanations

    The neuron is triggered by parentheses tokens and other markers that denote wire-service metadata or dateline numbers (e.g. “( 876.8491”, “( 863.553”), i.e. tokens used for article routing/metadata rather than regular text.

    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_24_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
     trifling
    0.54
     sacar
    0.50
     παρά
    0.49
     наличии
    0.49
     نہ
    0.48
    薨
    0.48
     собственно
    0.47
    হইবে
    0.47
     نه
    0.47
    🧾
    0.46
    POSITIVE LOGITS
    更多
    0.59
     SOURCE
    0.56
    M
    0.56
    Source
    0.56
     Source
    0.56
    Pat
    0.55
     QUESTIONS
    0.55
     FOLLOW
    0.54
    P
    0.53
    Mat
    0.53
    Activations Density 0.000%

    No Known Activations