© 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. 73863
    Prev
    Next
    INDEX
    Explanations

    events happening after or during actions

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

    The neuron responds strongly to numeric tokens or digit sequences (e.g., years, IDs, or counts).

    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
    BOTH
    -1.28
     lada
    -1.28
     kapital
    -1.24
     Soal
    -1.19
     klassi
    -1.18
     både
    -1.16
     इस्त
    -1.16
     Kontrol
    -1.16
    fahrer
    -1.15
     Lunes
    -1.15
    POSITIVE LOGITS
     during
    1.95
     when
    1.88
     after
    1.68
     later
    1.63
    その後
    1.43
     setelah
    1.37
     luego
    1.28
    when
    1.27
     afterwards
    1.26
     helping
    1.24
    Activations Density 0.006%

    No Known Activations