© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129503
    Prev
    Next
    INDEX
    Explanations

    The neuron flags mentions of criminal offenses or legal charges (e.g. words indicating crimes, charges, or penalties).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    criminal charges and legal actions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Not in Any Lists

    No Comments

    Negative Logits
    Ť
    -1.45
     BORDER
    -1.39
     çat
    -1.27
     jedis
    -1.23
     mba
    -1.22
     quibus
    -1.20
    hoka
    -1.20
     cuillère
    -1.19
     regiões
    -1.18
    ⇐
    -1.18
    POSITIVE LOGITS
     your
    1.53
    Many
    1.39
    Удачи
    1.33
     or
    1.32
     бампер
    1.28
     used
    1.27
     przód
    1.27
     种
    1.25
     Your
    1.23
     За
    1.23
    Activations Density 0.052%

    No Known Activations