© 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-3-12B
    3. 31-GEMMASCOPE-2-RES-16K
    4. 15844
    Prev
    Next
    INDEX
    Explanations

    desire

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

    This neuron responds to temporal and sequence markers in narrative (words that locate or link events in time, such as “date,” “over,” “during,” “next,” “down,” “three,” “course,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    romantic or intimate language and scenarios.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    mentions and depictions of romantic love and intimate relationships, including dating, affection, and suggestive or sexual situations.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     parenting
    0.77
    孩子们
    0.77
     friendships
    0.77
     가족
    0.70
     đứa
    0.70
     child
    0.69
    Child
    0.68
     Child
    0.68
    child
    0.68
    Children
    0.67
    POSITIVE LOGITS
    💏
    1.58
     romantic
    1.50
    Romantic
    1.45
     Romantic
    1.41
     romant
    1.39
    romantic
    1.35
    💑
    1.33
     роман
    1.22
     रोमांटिक
    1.20
    💋
    1.19
    Activations Density 0.368%

    No Known Activations