© 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. 13617
    Prev
    Next
    INDEX
    Explanations

    trade secrets, relations, and trade-offs

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

    This neuron strongly activates on the word “trade” (including in compounds like “trade-off,” “trade secret,” “trade union,” etc.).

    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
     propr
    0.64
     ³
    0.57
     아무
    0.56
     Lactobacillus
    0.56
     conveyor
    0.56
     ENT
    0.56
    rifuge
    0.55
     हरियाणा
    0.55
    椁
    0.55
     bain
    0.55
    POSITIVE LOGITS
    Trade
    1.19
     Trade
    1.15
     trade
    1.13
    trade
    1.11
     TRADE
    1.01
     Trades
    0.94
    TRADE
    0.93
    Trades
    0.91
    trades
    0.88
     traded
    0.88
    Activations Density 0.100%

    No Known Activations