© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 16632
    Prev
    Next
    INDEX
    Explanations

    properties, states, and processes

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

    The neuron strongly activates on long noun or adjective word endings (–ability, –ity, –ing, –ion, etc.), i.e. common nominalization and derivational suffixes in technical writing.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    ኽ
    -1.12
     唯美
    -1.08
    
    -1.07
    albeit
    -1.06
     Phật
    -1.04
     clô
    -1.04
     materialen
    -1.02
    зії
    -1.02
     levando
    -1.01
     suministro
    -0.99
    POSITIVE LOGITS
     not
    1.23
     of
    1.09
     nanti
    1.03
     rusak
    1.02
     any
    0.96
     may
    0.95
     k
    0.95
    Many
    0.94
    ventajas
    0.94
     For
    0.94
    Activations Density 0.276%

    No Known Activations