© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34060
    Prev
    Next
    INDEX
    Explanations

    table separators and headers

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

    The neuron fires on multi-digit numeric tokens (i.e. 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
     so
    -1.15
     momentos
    -1.12
     风景
    -1.10
    кожа
    -1.07
     Although
    -1.07
     dieses
    -1.05
    鳢
    -1.03
    сение
    -1.02
    industan
    -1.02
     as
    -1.01
    POSITIVE LOGITS
     choisir
    1.27
     expliquer
    1.17
     trabalha
    1.13
     réaliser
    1.13
     milj
    1.08
    んじゃ
    1.04
     remercier
    1.02
     effectuer
    0.99
    なども
    0.98
     vêtements
    0.96
    Activations Density 0.005%

    No Known Activations