© 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. 14953
    Prev
    Next
    INDEX
    Explanations

    applies to any case

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

    The neuron is tuned to formal legal/regulatory or policy-style statements—passages laying out conditions, obligations, or rules (often marked by modals like “will,” “must,” “have to,” and by specific legal terms and numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     cardin
    -1.27
     verspre
    -1.20
     utro
    -1.09
     prenn
    -1.05
     vell
    -1.02
     marqués
    -1.01
     verlie
    -1.00
     madame
    -0.99
     pantal
    -0.98
     pompe
    -0.98
    POSITIVE LOGITS
     anything
    1.70
     if
    1.48
    Any
    1.41
    Anything
    1.38
     examples
    1.30
     example
    1.25
     any
    1.18
     cases
    1.17
     ANY
    1.17
    Here
    1.16
    Activations Density 0.031%

    No Known Activations