© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 98475
    Prev
    Next
    INDEX
    Explanations

    guru, spiritual leaders, holy figures

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

    This neuron strongly activates on subword tokens containing the letter sequence “ha” (e.g. the “ha” in yoga, Sadhguru, Hatha, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     inevitable
    -1.04
     religiously
    -1.02
    ә
    -0.98
    /
    -0.97
     Carvalho
    -0.97
    es
    -0.95
    Ine
    -0.94
     ancestral
    -0.94
    ck
    -0.92
     irr
    -0.91
    POSITIVE LOGITS
     mejorado
    1.14
     évêque
    1.13
    футбо
    1.13
     esconder
    1.09
     kapac
    1.05
    образи
    1.04
     konkrét
    1.03
    ᭝
    1.03
    vetica
    1.02
     alqu
    1.02
    Activations Density 0.012%

    No Known Activations