© 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. 48054
    Prev
    Next
    INDEX
    Explanations

    nerd and geek related terms

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

    The neuron activates on labels of “nerdy” identity—i.e. words like “nerd,” “nerds,” “nerdy,” and “geek.”

    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
    ⤥
    -2.06
    焄
    -2.02
    鈰
    -1.99
    𖤍
    -1.95
    unicaciones
    -1.92
    冷却
    -1.91
    瓒
    -1.90
    咈
    -1.89
    獃
    -1.84
     deliciosa
    -1.76
    POSITIVE LOGITS
    0
    2.36
     Перейти
    2.00
     tne
    1.95
    色んな
    1.88
    きち
    1.79
    麪
    1.77
     dritte
    1.73
    色々と
    1.71
     работой
    1.70
     nerd
    1.67
    Activations Density 0.013%

    No Known Activations