© 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. 14690
    Prev
    Next
    INDEX
    Explanations

    past working relationships and mentorship

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

    The neuron detects descriptions of sexual harassment events (e.g., “sexually harassed her when they worked together”).

    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
    ～　
    -1.04
     people
    -0.99
     everyone
    -0.97
    ornada
    -0.97
    dipl
    -0.96
    的我
    -0.94
     it
    -0.89
    Всем
    -0.88
     everybody
    -0.88
     zarządz
    -0.87
    POSITIVE LOGITS
     his
    0.91
    師
    0.90
     friendship
    0.88
    graphers
    0.88
    IBinder
    0.87
    dı
    0.87
    gemeins
    0.86
    🤜
    0.86
    friendship
    0.86
    netje
    0.85
    Activations Density 0.055%

    No Known Activations