© 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. 69593
    Prev
    Next
    INDEX
    Explanations

    arbitrary

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

    The neuron is highly activated by the adjective “arbitrary” (especially when it modifies technical or mathematical terms).

    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
     '
    -3.75
    {
    -3.31
    ’
    -3.09
    )
    -3.05
     The
    -2.80
    ar
    -2.55
    苤
    -2.53
     wth
    -2.39
    я
    -2.39
    cortina
    -2.36
    POSITIVE LOGITS
    衪
    3.06
    杙
    2.95
    鄚
    2.81
    鱘
    2.78
    玧
    2.78
    𖡼
    2.78
    
    2.64
    甌
    2.63
    ization
    2.56
    ～　
    2.56
    Activations Density 0.002%

    No Known Activations