© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 110630
    Prev
    Next
    INDEX
    Explanations

    his/her and spouse or family

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

    The neuron fires on mentions of a person’s spouse, i.e. possessive pronouns (“his,” “her”) paired with kinship terms like “husband” or “wife.”

    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
    carnated
    -1.43
    螈
    -1.41
     sv
    -1.41
     что
    -1.34
     itself
    -1.32
     memahami
    -1.30
    らせる
    -1.30
     OF
    -1.27
    的主要
    -1.27
     med
    -1.26
    POSITIVE LOGITS
     этом
    1.76
     schmerz
    1.69
     are
    1.69
     halben
    1.55
     sogenannten
    1.48
     休閒
    1.47
     familiare
    1.46
     такого
    1.45
     einzigen
    1.44
     artistico
    1.44
    Activations Density 0.020%

    No Known Activations