© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 13485
    Prev
    Next
    INDEX
    Explanations

    friend

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

    The neuron specifically detects mentions of personal relationships by firing strongly on the word “friend.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_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
    ς
    1.00
     kidnapping
    0.91
    𝘀
    0.90
     Instance
    0.89
     calligraphy
    0.89
     Appeals
    0.89
     Directors
    0.87
    musik
    0.86
     hérit
    0.86
     отец
    0.85
    POSITIVE LOGITS
    ৯
    1.09
    เพื่อน
    1.00
    يسي
    0.99
    दोस्तों
    0.95
    იმ
    0.91
    aan
    0.89
     amigas
    0.88
    划分
    0.88
     colegas
    0.88
    ichloro
    0.87
    Activations Density 0.029%

    No Known Activations