© 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. 44575
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns and 'who/whom'

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    ’，
    -1.37
    haviours
    -1.35
     propriétaire
    -1.34
     vittoria
    -1.30
     laquelle
    -1.27
     hvilken
    -1.24
     remboursement
    -1.20
     konsumen
    -1.20
     novità
    -1.20
     ansvar
    -1.19
    POSITIVE LOGITS
     their
    2.98
     who
    2.31
     they
    2.31
     whom
    2.00
     Their
    1.88
     которых
    1.83
     их
    1.73
     They
    1.70
    Their
    1.69
     którzy
    1.67
    Activations Density 0.243%

    No Known Activations