© 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. 10-GEMMASCOPE-RES-131K
    4. 20327
    Prev
    Next
    INDEX
    Explanations

    gratitude and conditions

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

    The neuron lights up on multi‐word idiomatic or conversational phrases (e.g. “just being…,” “no harm done as I…,” “look closer and you’ll…”)

    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
     that
    -1.78
     kokos
    -1.64
     karton
    -1.64
     bied
    -1.62
     mieli
    -1.59
    értel
    -1.48
     eduk
    -1.48
    然後
    -1.45
     mogli
    -1.45
     balkon
    -1.44
    POSITIVE LOGITS
    {~
    1.51
    не
    1.38
     ample
    1.36
     more
    1.35
    已是
    1.34
     before
    1.30
     still
    1.30
     terlihat
    1.30
    Arbe
    1.29
    3
    1.29
    Activations Density 0.031%

    No Known Activations