© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 10963
    Prev
    Next
    INDEX
    Explanations

    introducing a subject's background

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

    This neuron spotlights section headings or labels (usually uppercase words followed by a colon) that demarcate different parts of a document.

    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
    India
    -1.01
     etme
    -0.91
     will
    -0.91
    ‛
    -0.90
     restoran
    -0.89
     कृपया
    -0.87
     kesi
    -0.86
    ヶ月
    -0.85
     蓝
    -0.84
     condominio
    -0.84
    POSITIVE LOGITS
    :
    1.28
     while
    1.24
    aupun
    1.13
    غيرة
    1.09
     although
    1.05
     none
    1.04
     поскольку
    1.04
    grupp
    1.03
    reportError
    1.02
     essendo
    1.02
    Activations Density 0.041%

    No Known Activations