© 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. 34-GEMMASCOPE-RES-131K
    4. 59818
    Prev
    Next
    INDEX
    Explanations

    duck or duck-related terms

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

    The neuron responds to tokens beginning with “duc,” such as “duck” or “duct.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    亢
    -0.73
    рым
    -0.70
    jack
    -0.68
    的下
    -0.68
    他自己
    -0.68
     mahdollis
    -0.67
     bagel
    -0.67
    กำ
    -0.67
    known
    -0.66
    不容易
    -0.66
    POSITIVE LOGITS
     Dynasty
    1.20
    Duck
    1.05
     Duck
    1.05
    billed
    1.02
    🦆
    1.02
     Dodgers
    1.00
     dynasty
    0.99
    lings
    0.88
    weed
    0.87
    鴨
    0.84
    Activations Density 0.016%

    No Known Activations