© 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. 14406
    Prev
    Next
    INDEX
    Explanations

    pauses and conjunctions

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

    The neuron fires on sentence-initial, capitalized tokens—that is, words (often pronouns or emphatic terms) that begin a new sentence.

    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
    و
    0.63
     Abub
    0.63
    只好
    0.62
    Conv
    0.60
     ignorant
    0.60
    చ్చి
    0.60
     Conv
    0.59
     Съ
    0.58
    ैर
    0.57
     હાલ
    0.56
    POSITIVE LOGITS
    aVal
    0.68
    ።
    0.62
    нк
    0.61
    BeerItem
    0.59
     opgenomen
    0.59
     jpg
    0.58
    poque
    0.58
    κες
    0.58
    uerpo
    0.57
    .[/
    0.57
    Activations Density 0.177%

    No Known Activations