© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 65401
    Prev
    Next
    INDEX
    Explanations

    programming includes and protocols

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

    The neuron strongly activates on underscores (as used in code identifiers and include paths), i.e. it’s looking for tokens containing “_.”

    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
     hướng
    -0.82
    Coro
    -0.75
    Odon
    -0.75
     panning
    -0.75
     pobreza
    -0.75
    Mots
    -0.74
     pride
    -0.74
    iedades
    -0.74
     procedimento
    -0.74
     nghệ
    -0.72
    POSITIVE LOGITS
    heran
    0.90
    Bone
    0.85
     Leeuwen
    0.83
     Bone
    0.82
    Constructor
    0.81
    various
    0.80
    bone
    0.79
    tainable
    0.77
    Available
    0.75
    Bought
    0.75
    Activations Density 0.003%

    No Known Activations