© 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. 10-GEMMASCOPE-RES-131K
    7. 65600
    Prev
    Next
    INDEX
    Explanations

    addendum after edit

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

    The neuron detects in‐post edit markers (e.g. “Edit:”, “EDIT:”, “edit:”) and related revision headings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ’
    -1.90
     AND
    -1.59
     lavish
    -1.54
     using
    -1.48
     If
    -1.48
     our
    -1.45
     under
    -1.45
     sebelum
    -1.44
     gemeins
    -1.41
     упаковка
    -1.41
    POSITIVE LOGITS
     pertinente
    1.82
    gencias
    1.82
     crappy
    1.74
     représenter
    1.65
     spesifik
    1.62
     egenskaper
    1.62
    鮡
    1.60
     素描
    1.60
     abiye
    1.59
     caricatura
    1.59
    Activations Density 0.014%

    No Known Activations