© 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. 10-GEMMASCOPE-RES-131K
    7. 43941
    Prev
    Next
    INDEX
    Explanations

    names and context

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

    The neuron fires strongly on capitalized proper names—people’s last names, organizations, teams, and other named entities.

    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
    of
    -4.31
    ll
    -3.02
    The
    -2.88
     Then
    -2.77
    苙
    -2.70
     της
    -2.63
     een
    -2.61
     של
    -2.58
    咈
    -2.53
     yearn
    -2.50
    POSITIVE LOGITS
    𓆜
    2.41
     douche
    2.36
    çou
    2.36
     inderdaad
    2.34
    𖥧
    2.31
    뿅
    2.31
    蜩
    2.30
     voldoende
    2.28
    増えて
    2.28
    龊
    2.23
    Activations Density 0.009%

    No Known Activations