© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27322
    Prev
    Next
    INDEX
    Explanations

    XML/JSON/plist structure

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

    The neuron is primarily attuned to numeric literals or number‐only tokens (e.g. standalone numbers in XML/JSON/code).

    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
    uransi
    -1.19
    skog
    -1.07
     ideen
    -1.07
     durs
    -0.99
     Conseil
    -0.97
    арак
    -0.97
     voetballer
    -0.97
    خرید
    -0.94
     élégante
    -0.94
     volna
    -0.93
    POSITIVE LOGITS
     देखा
    1.08
    ハツ
    1.05
    ⅕
    1.04
    Première
    1.02
     mastic
    0.98
    ⅗
    0.94
     mathe
    0.94
     bion
    0.94
    ⅘
    0.93
    Bande
    0.91
    Activations Density 0.001%

    No Known Activations