© 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. 22-GEMMASCOPE-RES-131K
    7. 67389
    Prev
    Next
    INDEX
    Explanations

    positive appearance descriptors

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

    The neuron fires on enthusiastic reaction statements—especially compliments or exclamations praising how something looks or asking excitedly “how” about it.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ַּ
    -1.34
    afety
    -1.09
    ۪
    -1.07
    ͏
    -1.02
    hores
    -1.01
    pected
    -1.01
    ynthesis
    -1.01
    ₒ
    -1.01
     élections
    -1.01
    Lifecycle
    -1.00
    POSITIVE LOGITS
     looks
    1.50
    どうやって
    1.17
     nicely
    1.14
     выглядит
    1.13
    thető
    1.05
     still
    1.04
     tuyo
    1.03
     contrasts
    1.02
     awfully
    1.00
     muszą
    0.96
    Activations Density 0.011%

    No Known Activations