© 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. 10-GEMMASCOPE-RES-131K
    7. 12677
    Prev
    Next
    INDEX
    Explanations

    code structure delimiters

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

    The neuron fires on programming identifiers that are method or property names in code (e.g. model, objects, stream, ToList, GetUserId).

    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
    ascin
    -1.73
    )
    -1.72
     dann
    -1.58
     Those
    -1.55
    ématiques
    -1.52
    そして
    -1.52
    6
    -1.52
     also
    -1.49
    ]
    -1.48
    sprint
    -1.45
    POSITIVE LOGITS
    してました
    1.66
    咈
    1.48
     начинают
    1.47
     hábitat
    1.45
    髎
    1.41
     remota
    1.39
    が見えて
    1.39
    峮
    1.38
    谂
    1.37
     pesada
    1.36
    Activations Density 0.003%

    No Known Activations