© 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. 73349
    Prev
    Next
    INDEX
    Explanations

    SOAPS

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

    The neuron fires on title‐ or heading‐style text (i.e. words in title case or large header phrases).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     different
    -1.40
     many
    -1.39
    וּ
    -1.33
    logotipo
    -1.15
     პ
    -1.13
     normally
    -1.11
     other
    -1.11
     large
    -1.10
    镭
    -1.09
     huge
    -1.09
    POSITIVE LOGITS
     your
    1.63
    ׇ
    1.55
     you
    1.53
    你
    1.44
     النبات
    1.35
     چربی
    1.34
    στο
    1.30
    voer
    1.30
     Transfermarkt
    1.25
     YOU
    1.23
    Activations Density 0.054%

    No Known Activations