© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 5058
    Prev
    Next
    INDEX
    Explanations

    **marca, brand, profit, limited, defined**

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

    The neuron fires on boilerplate metadata tokens—like licensing statements, copyright attributions, “filed” in patent refs, and site‐footer phrases—i.e. non‐content/legal/attribution 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
    indelijk
    0.77
    堊
    0.76
     pâle
    0.75
     karma
    0.75
     Karma
    0.75
    剩余
    0.74
     numéro
    0.72
    こんばんは
    0.72
    表达
    0.71
     tránh
    0.70
    POSITIVE LOGITS
    iin
    0.95
    s
    0.86
    i
    0.80
     inscrib
    0.77
    ною
    0.77
    l
    0.74
    ਆਂ
    0.72
     streng
    0.71
     ove
    0.70
    लेस
    0.70
    Activations Density 0.001%

    No Known Activations