© 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. 82490
    Prev
    Next
    INDEX
    Explanations

    weightlifting exercises

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

    The neuron activates strongly on mentions of squat variations (e.g. “back squat,” “overhead squat”) and related comparative exercise terms.

    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
    изнь
    -0.91
    Istanbul
    -0.88
     alfabe
    -0.86
    平凡
    -0.81
    rahmen
    -0.77
     wortel
    -0.76
     Niel
    -0.76
     CLAY
    -0.75
    strze
    -0.74
    Tó
    -0.74
    POSITIVE LOGITS
     lockout
    0.97
    Pause
    0.90
     Bar
    0.89
     bars
    0.87
     throughout
    0.87
     grip
    0.85
     narrower
    0.83
     racked
    0.82
     wide
    0.81
     Smith
    0.80
    Activations Density 0.020%

    No Known Activations