© 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. 34-GEMMASCOPE-RES-131K
    4. 38021
    Prev
    Next
    INDEX
    Explanations

    connections, neighbors, links

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

    This neuron consistently activates on uppercase identifiers and other capitalized tokens (e.g. class names, labels, data‐structure types).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    IMUM
    -0.90
    Özel
    -0.88
    ATIVO
    -0.81
     terakhir
    -0.79
    штей
    -0.79
     intégr
    -0.77
    aults
    -0.77
    ܛ
    -0.77
    Nolan
    -0.77
    anio
    -0.77
    POSITIVE LOGITS
     neighbors
    2.38
     neighbours
    2.13
     adjacent
    1.95
     neighbor
    1.70
     neighboring
    1.67
     successors
    1.64
     connected
    1.60
     friends
    1.50
     outgoing
    1.50
     neighbour
    1.47
    Activations Density 0.041%

    No Known Activations