© 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. 9496
    Prev
    Next
    INDEX
    Explanations

    The neuron consistently fires on the initial key content word of each new section or paragraph, marking the first substantive term that introduces a segment.

    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
    لاميذ
    0.77
     NOTICE
    0.73
    الى
    0.71
    ဆင့်
    0.71
    郏
    0.69
    iscus
    0.69
    Всем
    0.67
    寒い
    0.67
    ALLOW
    0.67
     slučaju
    0.66
    POSITIVE LOGITS
     uppercase
    0.84
     outliers
    0.77
     josé
    0.77
     homegrown
    0.75
     firsthand
    0.75
     অ
    0.75
     techn
    0.73
     unknown
    0.72
    Webpack
    0.71
     early
    0.71
    Activations Density 0.255%

    No Known Activations