© 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. 34-GEMMASCOPE-RES-131K
    7. 92099
    Prev
    Next
    INDEX
    Explanations

    prepositions followed by descriptors

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

    The neuron fires on numeric and percentage tokens and related study‐result language—i.e. quantitative statements of efficacy or reduction (e.g. “reduce,” “%,” numbers) in a scientific/medical context.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     spotless
    -0.79
    september
    -0.75
    ƽ
    -0.75
    Liver
    -0.73
     Cosmetic
    -0.70
     osmotic
    -0.70
    страции
    -0.69
     لاعب
    -0.69
     Probes
    -0.69
     Appearance
    -0.68
    POSITIVE LOGITS
     droplets
    1.23
     airborne
    1.23
     droplet
    1.12
     snee
    1.07
     SARS
    1.02
     distance
    0.98
     coughs
    0.97
     aeros
    0.96
     talking
    0.95
     particles
    0.91
    Activations Density 0.015%

    No Known Activations