© 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. 31-GEMMASCOPE-2-RES-16K
    7. 1646
    Prev
    Next
    INDEX
    Explanations

    combat sports

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

    The neuron strongly activates on mentions of boxers’ names and their fight outcomes or match results.

    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
     harp
    0.62
     স্থাপত্য
    0.62
     dood
    0.61
     calving
    0.60
     bibliographic
    0.60
     kolon
    0.59
     decou
    0.59
    桔
    0.59
     architectural
    0.58
     bookmarks
    0.57
    POSITIVE LOGITS
     boxing
    1.82
     UFC
    1.75
     Boxing
    1.75
     fighters
    1.74
    Boxing
    1.71
     Mayweather
    1.69
     boxers
    1.68
     MMA
    1.66
     fighter
    1.66
     boxer
    1.64
    Activations Density 0.196%

    No Known Activations