© 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. 66195
    Prev
    Next
    INDEX
    Explanations

    sentence beginnings

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

    The main thing this neuron does is spot words and phrases that signal romantic relationships or coupling (e.g., couple, dating, romance, engagement).

    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
     for
    -1.84
     as
    -1.65
     so
    -1.64
     to
    -1.61
     from
    -1.57
     other
    -1.44
     system
    -1.42
     also
    -1.42
     large
    -1.36
     S
    -1.34
    POSITIVE LOGITS
    alucía
    1.54
    睒
    1.44
     kompati
    1.43
    Takk
    1.43
    horabuena
    1.41
    两人
    1.40
     parfü
    1.40
    Recept
    1.39
    holdet
    1.39
    Kamu
    1.39
    Activations Density 0.011%

    No Known Activations