© 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. 22011
    Prev
    Next
    INDEX
    Explanations

    Joachim Kura ancourt

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

    The neuron fires strongly on proper names—place names, personal names, organizations—i.e. 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
     –
    -2.97
    ↵↵
    -2.86
    ጵ
    -2.75
    蘗
    -2.38
    *}$
    -2.36
    ating
    -2.33
    ן
    -2.30
    ｼ
    -2.25
     и
    -2.22
    jface
    -2.19
    POSITIVE LOGITS
    饪
    4.06
     Giappone
    3.09
    ൎ
    2.94
    ኧ
    2.91
    antaranya
    2.88
     ſte
    2.88
    訫
    2.84
     pagare
    2.77
     aveva
    2.73
     ندارد
    2.73
    Activations Density 0.203%

    No Known Activations