© 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. 92931
    Prev
    Next
    INDEX
    Explanations

    particularly, especially, notably

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

    The neuron primarily detects numeric data tokens (e.g., numbers, years, counts, and statistics).

    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
    今度は
    -1.13
     teknis
    -1.12
     素描
    -1.10
    
    -1.07
    Jis
    -1.04
     vegano
    -1.03
    놋
    -1.02
    菒
    -1.02
    蚰
    -1.02
    Viitteet
    -1.02
    POSITIVE LOGITS
     particularly
    1.59
     besonders
    1.48
     especially
    1.30
     özellikle
    1.23
    特に
    1.16
     Additionally
    1.16
     Particularly
    1.12
     inoltre
    1.10
     especialmente
    1.09
     notably
    1.05
    Activations Density 0.058%

    No Known Activations