© 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. 28359
    Prev
    Next
    INDEX
    Explanations

    nicknames in quotes before names

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

    The neuron detects personal nicknames or sobriquets set off in quotation marks within names.

    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
     elef
    -1.33
     banane
    -1.31
     Comunic
    -1.25
     ponga
    -1.25
     aves
    -1.22
     gole
    -1.20
     silla
    -1.20
     albic
    -1.19
     selva
    -1.18
     rastre
    -1.17
    POSITIVE LOGITS
     who
    1.38
    edificio
    1.23
    gettyimages
    1.20
    entendre
    1.13
    <bos>
    1.12
     hinzufügen
    1.06
    european
    1.05
    organe
    1.05
     logros
    1.05
    neko
    1.03
    Activations Density 0.009%

    No Known Activations