© 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. 1577
    Prev
    Next
    INDEX
    Explanations

    traditional or standard

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

    The neuron strongly activates on adjectives like “traditional,” “typical,” or “usual”—i.e. words that flag something as the conventional or normative case.

    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
    柢
    0.71
    呚
    0.69
    总理
    0.66
    لم
    0.66
    オリ
    0.66
    ظ
    0.66
    inté
    0.64
    公正
    0.63
    宇
    0.63
    朳
    0.63
    POSITIVE LOGITS
    ist
    0.92
    erweise
    0.90
     hallmarks
    0.82
     típ
    0.78
     haunts
    0.77
     pitfalls
    0.76
    methods
    0.71
     paralelo
    0.69
    ী
    0.69
     drowsiness
    0.68
    Activations Density 0.047%

    No Known Activations