© 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. 4688
    Prev
    Next
    INDEX
    Explanations

    equations and technical terms

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

    The neuron consistently lights up on the main content word at the end of a clause or sentence—typically the sentence-final noun or verb that carries its semantic “head.”

    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
    P
    0.72
    Repository
    0.69
     Operational
    0.69
     vices
    0.68
     действие
    0.68
    ium
    0.67
     operational
    0.67
     senses
    0.67
    ד
    0.67
     continuously
    0.66
    POSITIVE LOGITS
     स्प्रे
    0.88
     धमाकेदार
    0.85
    }_{\
    0.84
    ტის
    0.79
     सेटिंग
    0.76
    参照
    0.75
     तरल
    0.75
    andaş
    0.75
     echter
    0.74
     शुगर
    0.73
    Activations Density 0.000%

    No Known Activations