© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 59511
    Prev
    Next
    INDEX
    Explanations

    relative clauses describing nouns

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

    The neuron strongly activates on English relative‐clause markers (and nearby words) like “that,” “which,” and “who,” effectively identifying the introduction of a relative clause.

    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
     its
    -2.13
     recargable
    -1.92
     terrestre
    -1.66
     eléctrica
    -1.62
     atriz
    -1.57
     inalámbrica
    -1.55
     politika
    -1.53
     cantora
    -1.52
     capa
    -1.50
    Its
    -1.48
    POSITIVE LOGITS
     使っ
    1.70
    \},
    1.50
     Unless
    1.47
     semuanya
    1.43
     إليها
    1.42
     vostro
    1.38
     incomplète
    1.38
    
    1.36
     bassi
    1.35
    になりたい
    1.34
    Activations Density 0.010%

    No Known Activations