© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 87013
    Prev
    Next
    INDEX
    Explanations

    version numbers or years

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

    The neuron strongly activates on isolated code‐style tokens (identifiers like “map,” “name,” “Python,” etc.) and numeric literals, i.e. short alphanumeric pieces you’d see in code or 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_34/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
     wikipagina
    -0.75
    ftime
    -0.72
     risers
    -0.71
    coord
    -0.70
    και
    -0.68
     tổ
    -0.68
    DAYS
    -0.67
     числе
    -0.67
    acity
    -0.66
     trả
    -0.66
    POSITIVE LOGITS
     Fork
    0.67
    Fork
    0.65
    realtime
    0.63
     delicious
    0.62
     guy
    0.62
     spielt
    0.61
     yummy
    0.61
    reusable
    0.60
     Foy
    0.58
    𝘂
    0.58
    Activations Density 0.080%

    No Known Activations