© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 106600
    Prev
    Next
    INDEX
    Explanations

    inference and rules

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

    The neuron fires on document or section titles and headings, especially technical or domain‐specific headers.

    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
     opleiding
    -0.78
     citron
    -0.75
     Bunk
    -0.74
     NgModule
    -0.74
    prettier
    -0.73
     habitu
    -0.73
    视
    -0.72
    zeera
    -0.71
     moj
    -0.71
    setVisibility
    -0.71
    POSITIVE LOGITS
     expert
    2.28
     knowledge
    2.22
     Expert
    1.98
     Knowledge
    1.98
    Expert
    1.90
     reasoning
    1.85
     rule
    1.82
    expert
    1.82
    Knowledge
    1.74
     inference
    1.62
    Activations Density 0.052%

    No Known Activations