© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13467
    Prev
    Next
    INDEX
    Explanations

    prosecution under

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

    The neuron strongly activates on rare or low‐frequency tokens—uncommon words, names, and unique proper nouns.

    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
    什么
    1.00
    k
    0.93
    YOU
    0.88
    你是
    0.87
    Diaz
    0.86
    你
    0.84
    didn
    0.81
    YOUR
    0.79
    你有
    0.79
    zę
    0.78
    POSITIVE LOGITS
     creative
    0.68
    仄
    0.68
     previewBuilder
    0.66
     Creative
    0.66
     juxtap
    0.66
    atu
    0.64
    ুমার
    0.63
    ensical
    0.62
    ẹ̀
    0.61
     dissemin
    0.61
    Activations Density 0.000%

    No Known Activations