© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 100400
    Prev
    Next
    INDEX
    Explanations

    incapacitation through suffocation or strangulation

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

    The neuron detects terms referring to asphyxiation or suffocation (methods of causing death by cutting off air), such as strangulation, choking, drowning, and asphyxiation.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    
    -1.64
     ⓒ
    -1.60
    芘
    -1.55
    家で
    -1.55
     入荷
    -1.53
     TORONTO
    -1.52
     問題
    -1.50
     Cacao
    -1.49
     概要
    -1.48
    ustomed
    -1.47
    POSITIVE LOGITS
    }
    1.62
     one
    1.59
    ти
    1.49
     atmospheric
    1.45
     multifaceted
    1.42
     strangle
    1.41
     by
    1.41
    ка
    1.37
    ByPrimaryKey
    1.37
    ación
    1.36
    Activations Density 0.016%

    No Known Activations