© 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. 10-GEMMASCOPE-RES-131K
    7. 130193
    Prev
    Next
    INDEX
    Explanations

    not sure what or how

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

    The neuron detects expressions of uncertainty, especially “not sure” followed by words like “what” or “how.”

    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
     debería
    -1.48
     naturelle
    -1.48
     stormwater
    -1.44
     Moreover
    -1.43
     marvels
    -1.42
    的問題
    -1.38
     таке
    -1.37
    ٣
    -1.35
     villains
    -1.35
    !
    -1.34
    POSITIVE LOGITS
     huset
    1.67
     who
    1.63
     saati
    1.62
     how
    1.58
    
    1.53
     even
    1.53
    lamabad
    1.52
    
    1.51
     dezelfde
    1.46
     naudoti
    1.45
    Activations Density 0.030%

    No Known Activations