© 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. 104619
    Prev
    Next
    INDEX
    Explanations

    prefix

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

    This neuron fires on occurrences of the word “prefix” (and to a lesser extent “suffix”) in code or configuration identifiers.

    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
    {
    -4.03
    ↵↵
    -3.78
    or
    -3.36
    [
    -3.02
    "
    -2.84
    -
    -2.81
      
    -2.81
    .
    -2.75
     .
    -2.72
    )
    -2.55
    POSITIVE LOGITS
     車場
    3.11
    鮠
    3.08
     diki
    2.84
    鏐
    2.81
    ጁ
    2.81
     sayfa
    2.78
     abiye
    2.73
     rajut
    2.59
     saada
    2.58
    ‪.‬‬
    2.53
    Activations Density 0.004%

    No Known Activations