© 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. 13428
    Prev
    Next
    INDEX
    Explanations

    social programs

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

    The neuron fires strongly on the final content word of each paragraph or segment, effectively marking section or sentence endings.

    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
    0
    0.87
    4
    0.78
    2
    0.74
    7
    0.65
     ર
    0.62
     Μ
    0.62
    9
    0.62
    ub
    0.62
    5
    0.62
    はもちろん
    0.61
    POSITIVE LOGITS
    இதனால்
    0.74
    变得
    0.72
    不安定
    0.70
     இதனால்
    0.68
     deswegen
    0.68
     derfor
    0.68
    變得
    0.67
     verlieren
    0.67
     subconsciously
    0.66
    Therefore
    0.64
    Activations Density 0.009%

    No Known Activations