© 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. 23846
    Prev
    Next
    INDEX
    Explanations

    wireless and wired network protocols

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

    The neuron strongly activates on sequences of digits (e.g. standard numbers, model numbers, version numbers) in technical contexts.

    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
    adaptiveStyles
    -0.71
    }`}>
    -0.69
     切
    -0.66
    guise
    -0.66
     darken
    -0.66
     лук
    -0.66
     crescita
    -0.65
     infecciones
    -0.64
     bese
    -0.64
     Hof
    -0.64
    POSITIVE LOGITS
     nodes
    1.25
     Zig
    1.23
     node
    1.15
    RF
    1.05
    nodes
    1.05
    Zig
    1.01
     RF
    0.98
    Nodes
    0.94
    Rf
    0.92
     Nodes
    0.91
    Activations Density 0.036%

    No Known Activations