© 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. 68562
    Prev
    Next
    INDEX
    Explanations

    initials and related phrases

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

    The neuron fires on numeric tokens or digit sequences (e.g., timestamps, IDs, or standalone numbers).

    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
    odore
    -1.29
     другие
    -1.27
    今月
    -1.25
     tegens
    -1.24
    昨日は
    -1.23
    久し
    -1.20
    あなたが
    -1.20
     new
    -1.20
    先週
    -1.20
    問合せ
    -1.19
    POSITIVE LOGITS
     Grüßen
    1.39
    🩻
    1.34
    咴
    1.23
    ですから
    1.17
    ですので
    1.16
    ınca
    1.16
    🫅
    1.15
     herido
    1.14
    もあった
    1.14
     again
    1.14
    Activations Density 0.001%

    No Known Activations