© 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. 60618
    Prev
    Next
    INDEX
    Explanations

    start-up

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

    The neuron fires on the token “up” when it appears as part of hyphenated compound words (e.g. line-up, start-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
    s
    -2.78
    which
    -2.69
    獁
    -2.48
     lakini
    -2.45
    t
    -2.30
    т
    -2.28
     mika
    -2.28
    n
    -2.23
    StoryboardSegue
    -2.22
    鄚
    -2.20
    POSITIVE LOGITS
     otras
    2.52
     gracioso
    2.39
     felices
    2.34
     determinadas
    2.30
     uzman
    2.28
     ajouté
    2.23
     inspiré
    2.20
     hábiles
    2.20
    握手
    2.19
     dichas
    2.17
    Activations Density 0.026%

    No Known Activations