© 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. 34-GEMMASCOPE-RES-131K
    7. 98518
    Prev
    Next
    INDEX
    Explanations

    infrastructure, construction, broadcasting, oceanographic

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

    The neuron strongly responds to capitalized named‐entity tokens—especially place names, institutions, and other proper nouns.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     reticulum
    -0.77
    ́ng
    -0.72
    慮
    -0.71
    Killed
    -0.71
     progressBar
    -0.71
     pēc
    -0.69
    ší
    -0.69
     līdz
    -0.69
    лізу
    -0.69
    portrait
    -0.69
    POSITIVE LOGITS
    フロント
    0.84
     остав
    0.74
     事件
    0.72
    ♢
    0.72
    Уже
    0.71
     soporte
    0.71
    thouses
    0.69
    phous
    0.68
    лте
    0.68
     三角
    0.68
    Activations Density 0.015%

    No Known Activations