© 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. 42675
    Prev
    Next
    INDEX
    Explanations

    Initials and foreign names

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

    This neuron responds to author-name initials—single (or paired) uppercase letters (often with punctuation like “.” or “-”) in bibliographic entries.

    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
     outros
    -1.84
     otros
    -1.53
     Acuerdo
    -1.50
     outras
    -1.48
    other
    -1.48
    																								
    -1.41
     противополо
    -1.39
     anderen
    -1.34
    诳
    -1.32
    鲆
    -1.30
    POSITIVE LOGITS
    ’
    1.66
    '
    1.62
    囖
    1.55
     addTo
    1.41
    翼翼
    1.33
    Entrée
    1.31
     terlaris
    1.31
     frauen
    1.31
    糖尿病
    1.30
     emphat
    1.30
    Activations Density 0.023%

    No Known Activations