© 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. 34-GEMMASCOPE-RES-131K
    7. 54180
    Prev
    Next
    INDEX
    Explanations

    graph search algorithms

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

    The neuron fires on occurrences of named path‐finding algorithms (e.g. “Dijkstra,” “D* Lite,” and their variants).

    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
     splits
    -0.76
     placental
    -0.73
     Gummi
    -0.72
     Hiroshima
    -0.69
    cientes
    -0.69
    ADRID
    -0.68
     gestational
    -0.68
     derniers
    -0.68
     secre
    -0.67
     Abon
    -0.66
    POSITIVE LOGITS
     bfs
    2.14
    bfs
    2.09
     BFS
    2.06
     Dijkstra
    2.05
    BFS
    1.86
    DFS
    1.80
     DFS
    1.79
    dfs
    1.73
    dijk
    1.72
     breadth
    1.63
    Activations Density 0.059%

    No Known Activations