© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73108
    Prev
    Next
    INDEX
    Explanations

    elements, return, x

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

    The neuron fires on programming control-flow keywords and symbols that introduce or delimit code blocks (e.g. “else,” “do,” “for,” braces, etc.).

    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
     with
    -1.41
     all
    -1.08
     three
    -1.01
    Three
    -0.98
    Those
    -0.98
     allow
    -0.96
     new
    -0.95
     czyli
    -0.95
     first
    -0.93
     using
    -0.93
    POSITIVE LOGITS
    比べると
    1.14
     irmãos
    1.10
     میں
    1.10
    きれいな
    1.09
     commode
    1.06
     zodia
    1.06
     membros
    1.05
    耵
    1.05
     cinematogra
    1.03
    踭
    1.02
    Activations Density 0.005%

    No Known Activations