© 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. 9794
    Prev
    Next
    INDEX
    Explanations

    line separator divider

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

    The neuron responds to mentions of separators or dividing lines—words like “rule,” “Divider,” “line,” “border,” and related horizontal/vertical grid‐line terms.

    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
     to
    -1.33
     names
    -1.10
     images
    -1.02
    ;
    -1.02
     někter
    -0.97
     sicheren
    -0.97
     verkla
    -0.96
     eighties
    -0.96
     richtigen
    -0.94
     boy
    -0.93
    POSITIVE LOGITS
     medicamento
    1.13
     cerim
    1.11
    atação
    1.08
     inaugu
    1.07
    kowej
    1.06
    𝐗
    1.05
     satisfactor
    1.05
     disser
    1.03
     novelas
    1.01
     cortado
    1.01
    Activations Density 0.005%

    No Known Activations