© 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. 130758
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on numeric literals (digit tokens) in code/JSON contexts, i.e. on numbers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    license or conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     with
    -1.65
     from
    -1.63
     before
    -1.60
     during
    -1.55
     only
    -1.53
     when
    -1.52
     in
    -1.51
     on
    -1.49
     for
    -1.49
     first
    -1.45
    POSITIVE LOGITS
     aislados
    1.10
     schuldig
    1.09
    向けの
    1.09
     ajouter
    1.08
     insegn
    1.07
     masterful
    1.06
     seleccionados
    1.06
     údaj
    1.05
     penser
    1.05
    orni
    1.05
    Activations Density 0.099%

    No Known Activations