© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 67557
    Prev
    Next
    INDEX
    Explanations

    so far

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

    This neuron fires on markers of the perfect aspect—instances of “have/has/been” plus a past participle.

    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
     when
    -1.66
     AGAIN
    -1.66
     again
    -1.55
     thereupon
    -1.42
    óricos
    -1.42
    ୃ
    -1.36
     suddenly
    -1.34
     Đó
    -1.32
     preocupado
    -1.32
     bep
    -1.31
    POSITIVE LOGITS
    )
    1.50
     consegui
    1.37
     bisher
    1.34
    2
    1.31
    '
    1.26
     =
    1.24
     besucht
    1.21
    >
    1.20
    了不少
    1.20
     deixa
    1.18
    Activations Density 0.056%

    No Known Activations