© 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. 24-GEMMASCOPE-2-RES-16K
    7. 14241
    Prev
    Next
    INDEX
    Explanations

    call tag, compress, rules, side, Phase, Tie

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron activates strongly on the final token of each input segment, i.e. it is detecting segment or sentence boundaries.

    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
    ō
    0.71
    ₁(
    0.68
    ًا
    0.67
     of
    0.66
    َة
    0.65
     ()
    0.64
     denotes
    0.63
     (
    0.62
    および
    0.61
     conductance
    0.60
    POSITIVE LOGITS
    Personally
    0.96
     이런
    0.86
     Personally
    0.83
     lavorare
    0.82
     vivere
    0.80
     prestigio
    0.80
     vivre
    0.78
     travailler
    0.78
     सच्चे
    0.75
     bisogna
    0.75
    Activations Density 0.000%

    No Known Activations