© 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. 10-GEMMASCOPE-RES-131K
    7. 110985
    Prev
    Next
    INDEX
    Explanations

    being threatened

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

    The neuron activates on words built from the root “threat,” e.g. “threatened,” “threatening,” and “threatens.”

    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
    釭
    -2.83
    嵙
    -2.69
    ᚔ
    -2.67
    琑
    -2.66
     mencionado
    -2.66
    琹
    -2.56
    かけた
    -2.53
     anún
    -2.53
    وانید
    -2.48
    ⢊
    -2.48
    POSITIVE LOGITS
    ↵
    3.45
    -
    3.11
    '
    2.89
    1
    2.63
    "
    2.55
    lossians
    2.41
     öffentliche
    2.31
     ursprünglich
    2.31
    will
    2.30
     Houſe
    2.27
    Activations Density 0.004%

    No Known Activations