© 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. 282
    Prev
    Next
    INDEX
    Explanations

    The neuron identifies and highlights strong, context-significant action verbs (especially those describing strategic or political moves) in the text.

    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
     
    0.37
     pero
    0.37
    )
    0.31
     ',
    0.31
     fichier
    0.31
     trois
    0.31
    ',
    0.30
     gambar
    0.30
    !
    0.30
     beträgt
    0.30
    POSITIVE LOGITS
    䏹
    0.47
    <unused2157>
    0.43
     of
    0.38
     Historically
    0.37
     to
    0.37
    <unused2197>
    0.37
     furthering
    0.36
     לת
    0.36
    㡚
    0.34
    <unused210>
    0.33
    Activations Density 11.736%

    No Known Activations