© 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. 10-GEMMASCOPE-RES-131K
    7. 128538
    Prev
    Next
    INDEX
    Explanations

    megapixel, methylene blue, Blackpool magistrates, Penta, sequelize

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

    The neuron fires on fragments of chemical compound names (e.g. “methylene,” “penta-…” 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_10/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
    m
    -3.91
    -
    -3.55
      
    -3.52
    y
    -3.34
    …
    -3.31
    n
    -3.19
    1
    -3.16
     […]
    -3.11
     t
    -3.11
     «
    -3.08
    POSITIVE LOGITS
    蕺
    3.25
    牁
    3.19
    𓆸
    3.14
    饪
    3.14
     magasiner
    3.05
    痦
    3.00
     negatif
    2.98
    嗬
    2.95
    Ꮘ
    2.95
     kontinu
    2.88
    Activations Density 0.017%

    No Known Activations