© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 20588
    Prev
    Next
    INDEX
    Explanations

    look up

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

    The neuron strongly responds to the token “up,” especially when it appears in phrasal constructions like “look up” or “looking up.”

    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
    ]<<"
    -1.50
    I
    -1.27
    T
    -1.24
    ("
    -1.19
     donc
    -1.18
    getWidth
    -1.14
     must
    -1.12
    May
    -1.11
    ).
    -1.11
      
    -1.11
    POSITIVE LOGITS
    ★☆
    1.38
    ACCIÓN
    1.20
    ↑↑
    1.20
     традиции
    1.18
    Geografía
    1.18
    Personajes
    1.17
     ウィメンズ
    1.16
    𝖻
    1.16
    部に
    1.16
     azz
    1.15
    Activations Density 0.009%

    No Known Activations