© 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. 55800
    Prev
    Next
    INDEX
    Explanations

    not followed by specific words

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

    The neuron is strongly activated by negation and negative–polarity words (e.g. “not,” “wouldn’t,” “little,” “expecting”).

    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
     lente
    -0.91
     solide
    -0.87
     intrig
    -0.86
     absolu
    -0.84
     bologna
    -0.82
    izations
    -0.81
     maioria
    -0.78
    тын
    -0.78
     bep
    -0.78
     comport
    -0.77
    POSITIVE LOGITS
     envisaged
    0.95
     any
    0.88
    aktif
    0.87
    ͈
    0.85
     packet
    0.85
    🏟
    0.84
    ENTON
    0.84
     Ти
    0.83
     Masyarakat
    0.83
     endless
    0.82
    Activations Density 0.055%

    No Known Activations