© 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. 1403
    Prev
    Next
    INDEX
    Explanations

    The neuron is detecting the presence of soft‐hyphen or discretionary‐hyphen character sequences used to break words 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_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
    当前
    0.72
    cont
    0.71
    ​
    0.68
     என்பவர்
    0.68
    [{
    0.68
     बड़े
    0.67
    在這個
    0.65
    ='
    0.64
    ）
    0.64
    假設
    0.64
    POSITIVE LOGITS
     kinases
    0.72
     wigs
    0.68
     amma
    0.67
     nieces
    0.66
    ayutt
    0.66
     chinos
    0.66
     daerah
    0.65
     سبب
    0.65
     herbs
    0.65
     antide
    0.64
    Activations Density 0.000%

    No Known Activations