© 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. 14825
    Prev
    Next
    INDEX
    Explanations

    friends and acquaintances

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

    This neuron detects mentions of someone’s friend, particularly the phrase “a friend of mine.”

    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
    Hinweise
    -1.01
    Warna
    -1.01
     möglicherweise
    -0.98
     amables
    -0.96
     موجب
    -0.96
    مشخص
    -0.94
    の時に
    -0.94
    ṉ
    -0.93
     zufried
    -0.93
    iş
    -0.93
    POSITIVE LOGITS
     who
    2.56
     have
    1.15
     londres
    1.12
    whose
    1.06
     Jahren
    1.05
     from
    1.05
     whose
    1.05
    在我的
    1.05
    和我
    1.05
    I
    1.03
    Activations Density 0.026%

    No Known Activations