© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 105383
    Prev
    Next
    INDEX
    Explanations

    delicious things

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

    The neuron fires on positive taste descriptors—adjectives like “yummy,” “delish,” and “delicious” that signal something is tasty.

    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
    ternut
    -1.97
    は
    -1.83
    と同じ
    -1.77
    </th>
    -1.75
    </sub>
    -1.73
     ähnliche
    -1.73
    。」
    -1.67
     dasselbe
    -1.65
    ांकि
    -1.64
     hauptsächlich
    -1.63
    POSITIVE LOGITS
    .
    2.77
    葫
    1.63
     huile
    1.62
    mHandler
    1.59
     Dados
    1.57
     erreur
    1.55
     objectif
    1.52
     seule
    1.51
     potentiel
    1.46
     vertes
    1.46
    Activations Density 0.016%

    No Known Activations