© 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. 48115
    Prev
    Next
    INDEX
    Explanations

    project root

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

    The neuron strongly activates on all‐uppercase identifiers (e.g. macro or constant names) in code.

    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
    WET
    -0.82
    对着
    -0.80
     obyvateľov
    -0.79
     použit
    -0.79
    wet
    -0.75
     létre
    -0.74
    ushy
    -0.73
     Schick
    -0.73
     マルチ
    -0.73
     ihres
    -0.73
    POSITIVE LOGITS
    itosti
    0.88
    uride
    0.84
    componentWill
    0.84
    æld
    0.83
     operazioni
    0.81
    مشخصات
    0.81
     Tân
    0.80
     Trotzdem
    0.80
     Ř
    0.80
     molho
    0.79
    Activations Density 0.003%

    No Known Activations