© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 107430
    Prev
    Next
    INDEX
    Explanations

    kayak, canoe, SUP, paddle

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

    The neuron strongly activates on mentions of water‐based outdoor recreation—especially paddling sports such as kayak(s), SUP, rafting, etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     snowboard
    -0.95
     snowboarding
    -0.92
     golfer
    -0.91
     ski
    -0.89
     GOLF
    -0.88
     sailboat
    -0.88
     Motorcycle
    -0.88
     skier
    -0.86
     skiing
    -0.85
    walking
    -0.84
    POSITIVE LOGITS
     paddle
    2.02
     paddles
    1.64
     paddling
    1.63
    paddle
    1.57
     Paddle
    1.48
    Paddle
    1.48
     padd
    1.47
     kayak
    1.24
     SUP
    1.21
    padd
    1.17
    Activations Density 0.012%

    No Known Activations