© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 28163
    Prev
    Next
    INDEX
    Explanations

    second only to

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

    The neuron detects words and short phrases that signal ranking or ordering relationships—e.g. “next,” “to,” “after,” “behind”—in comparative statements.

    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
    -1.02
     different
    -0.97
     ingeniero
    -0.93
    همان
    -0.90
    Să
    -0.90
     ANY
    -0.88
    Índice
    -0.88
    дные
    -0.88
     (~(
    -0.86
     einfach
    -0.84
    POSITIVE LOGITS
     только
    1.09
     only
    1.02
     Only
    1.00
    nly
    0.98
     μόνο
    0.95
     magát
    0.95
     maybe
    0.93
     perhaps
    0.93
     itself
    0.93
    rebbe
    0.92
    Activations Density 0.028%

    No Known Activations