© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 6835
    Prev
    Next
    INDEX
    Explanations

    pronouns and conjunctions like 'that'

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

    This neuron activates on sentence-level discourse transition words or phrases (e.g. “alternatively,” “this way,” “in any case,” “therefore,” “for some of us,” etc.).

    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
    ט
    0.71
    end
    0.70
    thus
    0.69
    한편
    0.66
    MAC
    0.64
    fill
    0.63
    cussion
    0.62
    Rx
    0.62
    buscador
    0.62
    équipement
    0.62
    POSITIVE LOGITS
    我们可以
    0.94
     nej
    0.93
    我們可以
    0.92
     we
    0.91
    我們先
    0.91
     furono
    0.91
    我們會
    0.90
    我们在
    0.85
    你有
    0.85
     you
    0.85
    Activations Density 0.034%

    No Known Activations