© 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. 115904
    Prev
    Next
    INDEX
    Explanations

    eligible, protected, qualified

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

    The neuron specializes in spotting past‐participle adjectives (–ed words) that denote status, eligibility, certification, protection, or similar formal designations.

    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
     geho
    -1.01
     praktisch
    -0.98
     akut
    -0.95
     kurzem
    -0.94
     家居
    -0.92
     chauffe
    -0.91
     ་
    -0.91
    dames
    -0.91
     bedenken
    -0.90
    Földrajz
    -0.90
    POSITIVE LOGITS
     under
    1.27
     persons
    1.24
     status
    1.23
     activities
    1.15
     specified
    1.09
    status
    1.09
     items
    1.09
     statunit
    1.05
     uses
    1.04
     areas
    1.00
    Activations Density 0.045%

    No Known Activations