© 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. 34-GEMMASCOPE-RES-131K
    7. 77904
    Prev
    Next
    INDEX
    Explanations

    is/was subject to / governed by / done

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

    The neuron is detecting occurrences of the copula “to be” (is, was, are, be) – especially where it marks passive or legal‐style constructions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    CREATED
    -1.03
    想要
    -1.02
    created
    -1.02
    κια
    -0.89
     برای
    -0.89
     heran
    -0.89
     تغییر
    -0.88
     निर्माण
    -0.86
     ħ
    -0.85
    -0.85
    POSITIVE LOGITS
     done
    2.72
     undertaken
    2.27
     performed
    2.23
     accomplished
    2.00
     conducted
    1.98
     avven
    1.89
     dilakukan
    1.85
     odby
    1.80
     انجام
    1.80
    done
    1.73
    Activations Density 0.259%

    No Known Activations