© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 10746
    Prev
    Next
    INDEX
    Explanations

    equal to

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

    The neuron fires strongly on isolated high-register, technical or formal vocabulary items—especially verbs and adjectives (“exported,” “normal,” “emphasizes,” etc.) that you’d see in legal, regulatory, or scientific prose.

    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
    ﺭ
    0.91
    er
    0.87
    AndTime
    0.81
     ভিতরে
    0.80
     νό
    0.80
    <0x8C>
    0.79
    ber
    0.77
     Эд
    0.77
    atuan
    0.76
    ffel
    0.75
    POSITIVE LOGITS
    根据
    0.92
    opausal
    0.77
    讲解
    0.75
    随着
    0.74
    单纯
    0.73
    去
    0.71
    传统的
    0.70
    })^{
    0.70
    念
    0.70
    畴
    0.70
    Activations Density 0.000%

    No Known Activations