© 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. 65688
    Prev
    Next
    INDEX
    Explanations

    spatial navigation descriptions

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

    The neuron detects numeric values and quantities (tokens representing numbers or measurements).

    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
     различни
    -1.27
    Cevap
    -1.15
    stwie
    -1.14
     ANSWERS
    -1.11
    polski
    -1.08
     harga
    -1.07
    REVIEW
    -1.06
     изпъл
    -1.06
    stus
    -1.05
     черен
    -1.05
    POSITIVE LOGITS
     where
    1.38
     murió
    1.15
     abandonado
    1.14
    这不是
    1.12
     Substanz
    1.09
     abandoned
    1.08
    ednesdays
    1.08
     slightly
    1.07
     marginally
    1.05
    Polecam
    1.05
    Activations Density 0.018%

    No Known Activations