© 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. 8648
    Prev
    Next
    INDEX
    Explanations

    understanding the

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

    The neuron fires most strongly on unusual or low‐frequency tokens—that is, it picks out rare words (including numbers) in the 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
    ‌ی
    0.71
    يل
    0.70
     부동산
    0.68
    ב
    0.68
    िका
    0.64
     Мар
    0.64
    t
    0.64
    لي
    0.64
    ל
    0.64
    0.64
    POSITIVE LOGITS
    যদিও
    0.81
    ecause
    0.77
    elasticsearch
    0.76
    millimeters
    0.75
     iterates
    0.75
     aberr
    0.74
    clarinet
    0.74
     iteratively
    0.74
     षटके
    0.72
     seventies
    0.72
    Activations Density 0.000%

    No Known Activations