© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 99812
    Prev
    Next
    INDEX
    Explanations

    uncanny ability, UML Model, jarring grit

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

    The neuron fires on standout tokens like acronyms, proper names, and other unusually “strong” vocabulary terms.

    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
    ↵
    -3.22
    
    -2.84
     ursprüng
    -2.80
     verbess
    -2.75
    鸩
    -2.66
     You
    -2.55
    贗
    -2.53
     Then
    -2.50
     We
    -2.50
     wichtigste
    -2.38
    POSITIVE LOGITS
    橚
    3.36
    ᯾
    3.11
    ارتمان
    2.89
    龊
    2.80
    
    2.77
    ing
    2.75
    ycznego
    2.72
    蟶
    2.70
     bedeckt
    2.55
    苤
    2.55
    Activations Density 0.014%

    No Known Activations