© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 10404
    Prev
    Next
    INDEX
    Explanations

    religious, provincial, biblical, retinal, viral, parochial

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

    medical or scientific terminology with suffixes like "atic," "ysmal," "tic" or "phobic".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron selectively activates on descriptive adjectives (i.e. words bearing adjectival endings like –al, –ic, –ous, –y, etc.).

    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
     caráter
    -1.39
     These
    -1.23
     between
    -1.23
    tuigen
    -1.19
     will
    -1.18
    ,:);
    -1.18
     Einblick
    -1.14
     also
    -1.13
     equilíbrio
    -1.13
     verdadeiro
    -1.12
    POSITIVE LOGITS
     ór
    1.34
    ✨
    1.21
     bize
    1.20
     bizi
    1.20
     verden
    1.20
     that
    1.17
            
    1.16
    
    1.16
     liten
    1.15
     orientale
    1.14
    Activations Density 0.264%

    No Known Activations