© 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. 104340
    Prev
    Next
    INDEX
    Explanations

    technical or problem contexts

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

    This neuron is sensitive to relatively long, multisyllabic content words (i.e. higher-register or technical vocabulary).

    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
     entire
    -1.42
    Grupos
    -1.37
    2
    -1.36
     дитя
    -1.32
    Visite
    -1.31
    Fourni
    -1.29
     rimu
    -1.27
    Facile
    -1.27
    Parfait
    -1.25
     infal
    -1.23
    POSITIVE LOGITS
    %"),
    1.86
    ège
    1.45
    %");
    1.42
    ":[{
    1.35
    addCriterion
    1.34
    ęcie
    1.28
    itting
    1.21
    ๕
    1.21
    %")
    1.20
    ;");
    1.19
    Activations Density 0.330%

    No Known Activations