© 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. 93118
    Prev
    Next
    INDEX
    Explanations

    biological and chemical terms

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

    technical terms in academic or programming contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron responds strongly to uncommon or specialized tokens—rare, technical, or domain‐specific words rather than common vocabulary.

    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
     even
    -1.22
     after
    -1.12
     also
    -1.10
    生命的
    -1.00
     still
    -0.98
     would
    -0.95
     now
    -0.94
     dessutom
    -0.94
    丹药
    -0.93
     before
    -0.90
    POSITIVE LOGITS
    ække
    1.04
     imprimé
    1.00
     überprüft
    0.99
     estacionamento
    0.96
     saisi
    0.94
     discer
    0.93
    明日は
    0.92
     Efek
    0.91
     ใน
    0.91
     Início
    0.91
    Activations Density 0.026%

    No Known Activations