© 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. 84798
    Prev
    Next
    INDEX
    Explanations

    names, programming, and multiple languages

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

    The neuron strongly activates on unusually long individual tokens, regardless of their part of speech or language.

    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
    </em>
    -2.61
    艽
    -2.58
    偁
    -2.45
     \
    -2.42
    ୌ
    -2.41
    DockStyle
    -2.28
    🛇
    -2.28
    ]
    -2.28
    「
    -2.25
    ଝ
    -2.25
    POSITIVE LOGITS
    in
    3.64
     好看
    2.47
    適切
    2.38
    ⡑
    2.33
     trochu
    2.30
    になってしまう
    2.28
    Stap
    2.20
     influenced
    2.17
    がっ
    2.16
     pratica
    2.13
    Activations Density 0.006%

    No Known Activations