© 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. 10-GEMMASCOPE-RES-131K
    7. 128125
    Prev
    Next
    INDEX
    Explanations

    romantic partner

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

    The neuron detects mentions of a romantic relationship partner (e.g. the word “partner” in relationship contexts).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    </strong>
    -1.77
     in
    -1.73
    
    -1.63
     チップ
    -1.52
    }$
    -1.52
    のは
    -1.49
    »?
    -1.49
     принци
    -1.48
     Inglaterra
    -1.48
    
    -1.42
    POSITIVE LOGITS
     longtime
    1.84
    what
    1.78
     १
    1.76
    というわけで
    1.65
     virtudes
    1.63
    In
    1.62
    Even
    1.62
    当我
    1.60
    When
    1.59
    自分は
    1.59
    Activations Density 0.012%

    No Known Activations