© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 83863
    Prev
    Next
    INDEX
    Explanations

    assistant attorney general

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

    The neuron activates on mentions of legal officers’ formal titles and roles (e.g. “Attorney General,” “Assistant Attorney General,” etc.).

    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
     AHA
    -0.77
    orderid
    -0.76
     cbs
    -0.75
     coroner
    -0.74
    Psico
    -0.72
    Dataset
    -0.72
     Attorney
    -0.72
    ۓ
    -0.71
    叕
    -0.71
    phological
    -0.71
    POSITIVE LOGITS
     Bureau
    0.85
     Coy
    0.81
     supervising
    0.78
     supervis
    0.77
     older
    0.77
     fra
    0.77
    Coy
    0.76
    ekte
    0.75
     Assistant
    0.75
     bureau
    0.75
    Activations Density 0.095%

    No Known Activations