© 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. 39923
    Prev
    Next
    INDEX
    Explanations

    promotion and relegation

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

    The neuron activates on numeric tokens—integers, decimals, years, statistics—i.e. any standalone numbers in the text.

    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
     Erzb
    -1.09
     produite
    -1.08
    ;</
    -1.08
     réellement
    -1.05
     lesiones
    -1.02
     Audiodateien
    -1.02
    νονται
    -1.02
     fria
    -1.01
     voulu
    -1.00
     passione
    -1.00
    POSITIVE LOGITS
     this
    1.11
     deliber
    1.04
     its
    1.03
     on
    1.03
     satélite
    1.00
    mó
    0.98
     tier
    0.96
     in
    0.96
     rever
    0.95
     Besar
    0.94
    Activations Density 0.046%

    No Known Activations