© 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. 1159
    Prev
    Next
    INDEX
    Explanations

    richest country, War/Terror

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

    The neuron identifies section-heading words “pros” and “cons,” i.e. labels marking lists of advantages versus disadvantages.

    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
    iendo
    0.78
    س
    0.77
    aient
    0.75
    특별시
    0.75
    iov
    0.73
     arrondie
    0.73
     conformément
    0.73
    ić
    0.71
    iye
    0.71
    sd
    0.69
    POSITIVE LOGITS
    我相信
    0.80
    ਓ
    0.73
    佬
    0.69
    を取
    0.67
    を実行
    0.67
    を中心
    0.66
    不上
    0.66
    を取得
    0.66
    š
    0.65
     bashing
    0.64
    Activations Density 16.419%

    No Known Activations