© 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. 22-GEMMASCOPE-RES-131K
    7. 131004
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on content‐bearing keywords (mostly nouns and salient topic words) in titles or ad‐style listings, ignoring common function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    German marriage certificate translation marriage documents

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    这是一
    -0.90
    相片
    -0.89
    Rozm
    -0.88
    旄
    -0.88
    ξει
    -0.88
    énier
    -0.88
    Tento
    -0.88
    化する
    -0.86
    siehe
    -0.85
     scorching
    -0.84
    POSITIVE LOGITS
    .
    0.90
     rendition
    0.85
    他能
    0.81
    汚
    0.79
    芽
    0.77
    ation
    0.77
     воду
    0.77
     карту
    0.76
     FK
    0.75
    不肯
    0.75
    Activations Density 0.007%

    No Known Activations