© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 425
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on extremely frequent “stop” tokens—common function words like a, I, it, and, but, he, are, etc.

    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
     wholeheartedly
    0.44
     enshrined
    0.37
    。」
    0.36
    🕖
    0.36
    ⏯
    0.34
     אַ
    0.33
    甚
    0.32
     unmistak
    0.32
    🕔
    0.32
     burdened
    0.32
    POSITIVE LOGITS
    erdapat
    0.51
    utilizzo
    0.49
    ım
    0.47
    t
    0.47
    yh
    0.47
    ルの
    0.47
    x
    0.47
    liquer
    0.46
    n
    0.46
    ueba
    0.46
    Activations Density 0.000%

    No Known Activations