© 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. 104382
    Prev
    Next
    INDEX
    Explanations

    source and target

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

    This neuron fires on common programming parameter names—especially “target” (and related identifiers like “source,” “destination,” or “input”) 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_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
     astounding
    -1.27
     horrid
    -1.19
     splendid
    -1.14
     wonderful
    -1.13
     superb
    -1.09
     marvelous
    -1.07
    notre
    -1.04
     overarching
    -1.04
    Hotels
    -1.02
     amazingly
    -1.02
    POSITIVE LOGITS
     if
    1.38
    target
    1.32
     specified
    1.30
     cible
    1.29
     destination
    1.21
     should
    1.20
     what
    1.18
    ప్ర
    1.16
     areas
    1.15
     target
    1.14
    Activations Density 0.037%

    No Known Activations