© 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. 7241
    Prev
    Next
    INDEX
    Explanations

    you're following

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

    The neuron strongly activates on the “’re” contraction (as in you’re, we’re) indicating it’s detecting the contracted form of “are.”

    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
     you
    -2.91
     We
    -2.73
    {
    -2.38
     doesn
    -2.33
     she
    -2.27
     повіт
    -2.17
     यह
    -2.11
     bunun
    -2.09
    g
    -2.09
            
    -2.08
    POSITIVE LOGITS
    }
    2.47
     upto
    2.41
     raiſ
    2.30
    ,—
    2.23
    婠
    2.20
    ——”
    2.20
    – 
    2.19
     Enero
    2.17
     westfalen
    2.16
     広場
    2.16
    Activations Density 0.019%

    No Known Activations