© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 537
    Prev
    Next
    INDEX
    Explanations

    The neuron consistently spikes on the initial fragment (the first syllable or prefix) of longer, multi-syllable content words, marking the word’s beginning.

    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
     Notices
    0.34
     Executive
    0.32
     Appeal
    0.32
     vou
    0.32
     Integration
    0.31
     VID
    0.31
     Linking
    0.31
     É
    0.31
     Adds
    0.30
     Assim
    0.30
    POSITIVE LOGITS
    oração
    0.35
     cheesecake
    0.33
     garlic
    0.33
     shrub
    0.33
    bathtub
    0.33
    ussels
    0.33
    apsack
    0.33
    chocolate
    0.33
     earthenware
    0.33
     tablecloth
    0.32
    Activations Density 0.000%

    No Known Activations