© 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. 22-GEMMASCOPE-RES-131K
    7. 67411
    Prev
    Next
    INDEX
    Explanations

    variable assignments in code

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

    The neuron fires on tokens that start with a capital letter—i.e. the second‐word fragments in camelCase or PascalCase identifiers.

    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
     schli
    -1.30
    側から
    -1.15
    postolic
    -1.14
    有不少
    -1.13
     zustand
    -1.11
    事です
    -1.11
     the
    -1.08
    aigu
    -1.08
     stadt
    -1.05
    キレイ
    -1.04
    POSITIVE LOGITS
    damit
    1.15
     =
    1.14
    ☎️
    1.02
    Estoy
    1.00
    сумка
    0.98
    📞
    0.97
    俳優
    0.94
     funktioner
    0.93
    Τα
    0.93
     лютого
    0.92
    Activations Density 0.012%

    No Known Activations