© 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. 73423
    Prev
    Next
    INDEX
    Explanations

    took his own life

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

    The neuron fires on mentions of suicide or self-harm (e.g. “took his own life,” “committed suicide,” “thoughts of taking my own life”).

    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
     feind
    -0.79
     startTime
    -0.79
    TLS
    -0.74
    身に
    -0.74
     arteries
    -0.71
    返し
    -0.70
    ってきて
    -0.69
     prevenção
    -0.69
    BST
    -0.69
    TokenNameL
    -0.68
    POSITIVE LOGITS
     take
    2.34
     taking
    2.27
     took
    2.16
    Taking
    2.13
    taking
    1.96
    Take
    1.90
     Taking
    1.90
     takes
    1.84
    take
    1.81
    Took
    1.77
    Activations Density 0.021%

    No Known Activations