© 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. 38909
    Prev
    Next
    INDEX
    Explanations

    yoga poses

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

    The neuron strongly activates on the character sequence “ha” (for example in “Ha­­tha”) wherever it appears.

    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
     which
    -1.52
     a
    -1.48
     for
    -1.45
     their
    -1.45
     could
    -1.44
     did
    -1.42
     if
    -1.32
     from
    -1.29
     whose
    -1.28
    S
    -1.26
    POSITIVE LOGITS
    FUENTES
    1.38
     النبات
    1.35
    </sup>
    1.35
    NYA
    1.32
     ܐ
    1.29
    躏
    1.29
    BEAUTY
    1.29
     Aérea
    1.29
    Yoga
    1.28
    妯
    1.28
    Activations Density 0.027%

    No Known Activations