© 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. 130789
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on comparative and analogical language—phrases that liken or compare one person, thing, or time to another.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    historical comparisons

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     correos
    -0.95
     souvent
    -0.92
    (_)
    -0.92
     ২০১
    -0.91
    後來
    -0.91
     недавно
    -0.90
     contenedores
    -0.90
    gestaltung
    -0.90
     atuais
    -0.89
    枋
    -0.89
    POSITIVE LOGITS
     historically
    1.35
     historic
    1.30
     classic
    1.23
     Historically
    1.23
     ancient
    1.19
     previous
    1.18
     old
    1.14
     historical
    1.11
     früher
    1.11
     anciens
    1.10
    Activations Density 0.108%

    No Known Activations