© 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. 15326
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly responds to low‐frequency, unusually long or domain‐specific tokens—i.e. rare words that stand out 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
     downtime
    0.87
    的一个
    0.79
     heady
    0.75
    细腻
    0.73
     filthy
    0.71
    震惊
    0.71
    𝟬
    0.71
     scouring
    0.70
    되어
    0.70
    队长
    0.69
    POSITIVE LOGITS
    t
    0.80
     Kur
    0.77
     KUR
    0.76
    Kur
    0.76
     W
    0.74
    b
    0.74
    ve
    0.74
    iour
    0.73
     Mer
    0.73
     แชร์
    0.73
    Activations Density 0.000%

    No Known Activations