© 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. 24-GEMMASCOPE-2-RES-16K
    7. 9006
    Prev
    Next
    INDEX
    Explanations

    subtract from

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

    The neuron fires on the “headline” or “label” words that introduce or identify structured sections (e.g. proper names, section titles, numbered‐item headings) in a document.

    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
     vitesses
    0.80
    妗
    0.80
    ੇ
    0.79
     그녀
    0.78
     meubles
    0.78
     plantes
    0.78
     రాష్ట్ర
    0.76
     \%)$
    0.75
     euthanasia
    0.75
    т
    0.75
    POSITIVE LOGITS
    and
    0.81
    op
    0.79
    ata
    0.74
    ati
    0.67
    tag
    0.67
     Railway
    0.66
    acs
    0.66
    ない
    0.66
    nach
    0.66
    aya
    0.66
    Activations Density 0.002%

    No Known Activations