© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6328
    Prev
    Next
    INDEX
    Explanations

    understand urges and desires

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

    This neuron specializes in detecting words that express internal desires, motivations, or urges.

    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_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
     अपमान
    0.71
     불편
    0.65
     awkwardly
    0.65
    尴尬
    0.64
    🖕
    0.64
     complained
    0.63
     инвали
    0.61
     تۆ
    0.58
     complains
    0.57
     complaints
    0.57
    POSITIVE LOGITS
     temptation
    2.47
     temptations
    2.31
     tempted
    2.02
     tempting
    1.92
     Tempt
    1.88
     tempt
    1.77
     lure
    1.63
    誘
    1.61
     লোভ
    1.60
     allure
    1.52
    Activations Density 0.825%

    No Known Activations