© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2904
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on software‐API documentation or code‐comment blocks (e.g. function signatures, parameter/return descriptions), distinguishing those technical doc passages from ordinary prose.

    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
    ból
    0.96
    sth
    0.82
    figs
    0.79
     व्हेल
    0.78
    если
    0.77
    :~
    0.76
    fisher
    0.75
    né
    0.75
    vět
    0.74
    salmon
    0.74
    POSITIVE LOGITS
    ↵↵
    1.09
    ↵↵↵
    1.04
    </h2>
    1.00
    <0x0D>
    0.99
    ↵
    0.96
     \\
    0.87
     }}$
    0.86
    <0x0B>
    0.84
    0.83
    0.82
    Activations Density 0.080%

    No Known Activations