© 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. 129141
    Prev
    Next
    INDEX
    Explanations

    The neuron lights up on relatively rare, content-bearing words—i.e. uncommon or specialized vocabulary rather than common function words.

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

    back and since

    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
    Ii
    -1.13
    меется
    -1.12
     autén
    -1.10
    
    -1.04
     willen
    -1.00
     oe
    -0.99
    osten
    -0.99
    xv
    -0.98
    なのでしょう
    -0.98
     enfermos
    -0.96
    POSITIVE LOGITS
    特長
    1.05
     internationaux
    1.04
     political
    1.02
     two
    1.02
     Since
    0.99
     Entsche
    0.98
    書籍
    0.98
     Darum
    0.97
    年度
    0.96
     kebanyakan
    0.96
    Activations Density 0.054%

    No Known Activations