© 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. 69109
    Prev
    Next
    INDEX
    Explanations

    visible flaws

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

    The neuron fires on mentions of body parts—especially when they’re listed (e.g. “face, hands & feet”).

    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
    }):
    -0.98
    mband
    -0.96
    vccd
    -0.96
    ございました
    -0.94
    uomo
    -0.94
     atardecer
    -0.91
    팠
    -0.90
    byshev
    -0.89
    όρ
    -0.89
     ofrecen
    -0.89
    POSITIVE LOGITS
     visible
    1.29
     appearance
    1.09
     stark
    0.92
    dign
    0.91
     display
    0.91
     unattractive
    0.91
     contrasting
    0.90
     appearances
    0.88
    猎人
    0.87
     menampilkan
    0.86
    Activations Density 0.140%

    No Known Activations