© 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. 34-GEMMASCOPE-RES-131K
    7. 33049
    Prev
    Next
    INDEX
    Explanations

    UI component names and suffixes

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

    The neuron fires on fragments of compound identifiers (especially CSS class names) that use hyphens or underscores.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ëren
    -0.93
    тий
    -0.90
     nawet
    -0.88
    طقة
    -0.87
     CLAR
    -0.87
     vuonna
    -0.85
     Allociné
    -0.85
    вля
    -0.85
     často
    -0.84
    telar
    -0.84
    POSITIVE LOGITS
     also
    1.10
     type
    0.98
     flush
    0.98
     without
    0.98
    without
    0.96
     జ
    0.95
     hasn
    0.91
     zonder
    0.87
    CGFloat
    0.85
     didn
    0.84
    Activations Density 0.011%

    No Known Activations