© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21013
    Prev
    Next
    INDEX
    Explanations

    what [noun] does/do

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

    The neuron strongly activates on content words that appear in headings or titles (e.g. key nouns in questions and section titles).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pneus
    -1.74
    躂
    -1.64
     手绘
    -1.64
     rente
    -1.61
     ouest
    -1.52
    ㍑
    -1.51
    ització
    -1.50
    ≂
    -1.48
     内衣
    -1.48
     fournisseurs
    -1.48
    POSITIVE LOGITS
     this
    2.08
     you
    2.06
     do
    1.78
     it
    1.76
     makes
    1.73
     obecnie
    1.62
    lardır
    1.59
     they
    1.54
     add
    1.54
     respectivamente
    1.51
    Activations Density 0.056%

    No Known Activations