© 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. 31-GEMMASCOPE-2-RES-16K
    7. 11886
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on subword fragments created by line wraps or hyphenation—that is, pieces of words split across lines.

    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
     रोजिक
    0.54
     ाट
    0.49
    <unused2179>
    0.48
    <unused182>
    0.47
    <unused2175>
    0.47
    <unused2162>
    0.46
    ଃ
    0.46
     Diện
    0.45
     recebe
    0.45
     demiş
    0.45
    POSITIVE LOGITS
    5
    0.52
    1
    0.49
    4
    0.49
    7
    0.49
    2
    0.49
    3
    0.49
    0
    0.46
    8
    0.46
    ).
    0.46
    Sp
    0.46
    Activations Density 0.055%

    No Known Activations