© 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. 22-GEMMASCOPE-RES-131K
    7. 72836
    Prev
    Next
    INDEX
    Explanations

    lists of names and publications

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

    The neuron strongly activates on capitalized proper‐noun tokens, i.e. names of publications, places, 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_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
     alfombra
    -1.08
    ̣i
    -1.07
     that
    -1.02
    правки
    -1.01
     options
    -1.00
    默认
    -0.98
    ならではの
    -0.97
    版本
    -0.97
    ]-->
    -0.96
    ìm
    -0.96
    POSITIVE LOGITS
    ־ה
    1.24
    felizmente
    1.17
    uyasha
    1.08
    各类
    1.08
    וּ
    1.07
    ▵
    1.03
    anciers
    1.02
     various
    1.01
     여러
    1.00
    וֹת
    1.00
    Activations Density 0.033%

    No Known Activations