© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 3497
    Prev
    Next
    INDEX
    Explanations

    words ending in -able, -ive, -age, -ess

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

    Neuron 4 fires on the last substantive content word right before a segment or paragraph break.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    ↵↵
    0.85
     although
    0.70
     Region
    0.69
    <0xC2>
    0.69
     Gör
    0.68
     ,
    0.67
     Although
    0.67
     Global
    0.67
    ↵
    0.67
     International
    0.66
    POSITIVE LOGITS
    ifiable
    1.02
    therners
    0.94
    iguing
    0.93
    givings
    0.91
    eled
    0.90
    opically
    0.90
    therly
    0.90
    istically
    0.89
    duled
    0.89
    ocentric
    0.89
    Activations Density 1.069%

    No Known Activations