© 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. 514
    Prev
    Next
    INDEX
    Explanations

    empty, null, or void values

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

    This neuron detects placeholder or default‐value identifiers (e.g. names like “empty,” “noop,” “dummy,” or “ZERO”).

    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
     flink
    -1.32
     eenig
    -1.24
     dichter
    -1.20
    honestly
    -1.15
    primarily
    -1.13
     !!!!
    -1.13
    considering
    -1.13
     xxvi
    -1.11
     lidí
    -1.10
     โดย
    -1.09
    POSITIVE LOGITS
     if
    1.58
     there
    1.39
     when
    1.34
     because
    1.32
     purposefully
    1.17
     as
    1.16
     purposely
    1.13
     in
    1.07
     nearing
    1.05
     universally
    1.04
    Activations Density 0.023%

    No Known Activations