© 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. 10-GEMMASCOPE-RES-131K
    7. 19377
    Prev
    Next
    INDEX
    Explanations

    negated past participles

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

    The neuron detects negation constructions, especially the word “not” when used to negate verbs.

    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
     eccl
    -2.69
    
    -2.28
     precau
    -2.23
    
    -2.19
     conster
    -2.19
     leopardo
    -2.17
    emplares
    -2.17
     caos
    -2.17
    viembre
    -2.14
     видит
    -2.14
    POSITIVE LOGITS
    .
    2.22
    4
    2.16
    <sup>
    2.14
     in
    2.05
    2
    2.03
    𝐲
    2.02
    8
    2.00
    <u>
    1.96
    ↵↵↵
    1.93
    н
    1.90
    Activations Density 0.146%

    No Known Activations