© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 59597
    Prev
    Next
    INDEX
    Explanations

    turn off / turn something off

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

    The neuron fires when it detects the idiomatic phrase “turn a blind eye (to …),” especially the words “turn,” “blind,” and “eye.”

    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
     enlaces
    -0.85
    アンス
    -0.81
     Nishi
    -0.79
    leness
    -0.77
    йн
    -0.75
    bereiche
    -0.73
     recomiendo
    -0.72
     personalizada
    -0.71
    conta
    -0.71
    ën
    -0.71
    POSITIVE LOGITS
     turn
    2.75
    Turn
    2.58
     turning
    2.42
    Turning
    2.42
    turn
    2.41
     Turn
    2.34
    turning
    2.33
     Turning
    2.27
     turned
    2.20
    TURN
    2.19
    Activations Density 0.019%

    No Known Activations