© 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. 21625
    Prev
    Next
    INDEX
    Explanations

    annoying, irritating, frustrated

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

    The neuron fires strongly on words that express annoyance or irritation (e.g., “annoying,” “irritating,” “pisses off”).

    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
     a
    -2.77
     kanske
    -1.65
     nocturna
    -1.61
    ၞ
    -1.49
     With
    -1.48
     multitude
    -1.47
    kinci
    -1.39
     剤
    -1.38
    ns
    -1.37
     –
    -1.36
    POSITIVE LOGITS
     всички
    1.88
    妧
    1.79
     MILLION
    1.77
    medriver
    1.72
     wszystkie
    1.66
    閦
    1.63
     obligación
    1.63
    
    1.60
     pudieron
    1.59
    争议
    1.59
    Activations Density 0.024%

    No Known Activations