© 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. 65663
    Prev
    Next
    INDEX
    Explanations

    daring actions

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

    The neuron fires on multiword threat or challenge constructions expressing opposition (e.g. “to be reckoned with,” “those who oppose them,” “if you have a problem with that”).

    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
    Мето
    -0.90
    充電
    -0.79
    ーん
    -0.78
    IORITY
    -0.77
     menjelaskan
    -0.77
    vaig
    -0.76
    atorze
    -0.76
     horrid
    -0.75
    automat
    -0.75
    mutlich
    -0.75
    POSITIVE LOGITS
     dare
    2.36
     dared
    2.22
     dares
    1.95
     attempt
    1.63
     Dare
    1.60
    dare
    1.60
    Dare
    1.55
     challenge
    1.52
     try
    1.52
    敢
    1.51
    Activations Density 0.027%

    No Known Activations