© 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. 31-GEMMASCOPE-2-RES-16K
    7. 496
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on weird formatting/artifact characters (e.g. “”), stray backslashes or escape-style quotes, and other nonstandard punctuation rather than normal 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_31_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
     heen
    0.31
     diverges
    0.30
     daleko
    0.29
     impenetr
    0.28
     ÷
    0.28
     skr
    0.28
     sublattice
    0.27
     smach
    0.27
     facsimile
    0.27
     mye
    0.27
    POSITIVE LOGITS
    玫瑰
    0.35
    练习
    0.32
    если
    0.32
    ре
    0.31
    READ
    0.31
    aremos
    0.31
    閱讀
    0.31
    暉
    0.31
    DON
    0.31
    真
    0.31
    Activations Density 0.260%

    No Known Activations