© 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. 72181
    Prev
    Next
    INDEX
    Explanations

    including most

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

    The neuron fires on numeric tokens—digits or numbers used in stats, counts, measurements, scores, etc.

    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
    rollup
    -0.98
    Tél
    -0.98
    riever
    -0.90
    Aeros
    -0.90
     harganya
    -0.88
    denial
    -0.88
     rique
    -0.88
     berubah
    -0.87
    Punj
    -0.87
     comandos
    -0.87
    POSITIVE LOGITS
     including
    1.20
     głównie
    1.20
    其中
    1.17
    Includes
    1.13
     majority
    1.03
     hauptsächlich
    1.00
     predominantly
    1.00
    包含
    0.98
    Mix
    0.95
    including
    0.94
    Activations Density 0.075%

    No Known Activations