© 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. 90397
    Prev
    Next
    INDEX
    Explanations

    what's wrong with

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

    The neuron activates on instances of the phrase “wrong with,” especially in questions like “What’s wrong with…?”

    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
    In
    -1.85
    Very
    -1.77
    9
    -1.75
    What
    -1.61
    Dimensioni
    -1.58
    确实是
    -1.57
    One
    -1.55
    بيل
    -1.52
     of
    -1.52
     It
    -1.48
    POSITIVE LOGITS
    鲼
    1.84
    ⚣
    1.74
     projet
    1.70
    忖
    1.70
    茈
    1.64
    姈
    1.59
     departement
    1.59
    mindedness
    1.59
    梛
    1.54
     THOMAS
    1.52
    Activations Density 0.005%

    No Known Activations