© 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. 65871
    Prev
    Next
    INDEX
    Explanations

    on line

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

    The neuron activates on numeric tokens—especially numbers appearing in parentheses or as line‐number/file‐reference codes in warnings, error messages, and metadata.

    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
     of
    -1.35
    ;%%
    -1.31
    ✨:
    -1.27
    зде
    -1.16
    gsub
    -1.13
    ᐈ
    -1.12
    endforeach
    -1.11
    脛
    -1.09
    痒
    -1.09
     альбом
    -1.08
    POSITIVE LOGITS
     сказал
    1.14
     цьому
    1.09
    這裡
    1.09
     сьогодні
    1.04
     deoarece
    1.03
    donde
    1.02
     weihnachten
    1.02
     lác
    0.99
    嚒
    0.99
     crescita
    0.98
    Activations Density 0.001%

    No Known Activations