© 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. 128532
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on the path‐separator “/” when it’s immediately followed by a purely numeric segment (i.e. a slash introducing a digit‐only directory or filename)

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    path separators or file types

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    pantalon
    -1.50
     verborgen
    -1.41
     eenig
    -1.33
     brachten
    -1.32
     gedachten
    -1.29
     nimmer
    -1.29
     maler
    -1.27
     dichter
    -1.26
    licitations
    -1.26
     eenige
    -1.24
    POSITIVE LOGITS
     observa
    1.25
     complementar
    1.24
    معرفی
    1.24
     activa
    1.23
    begin
    1.16
     retir
    1.14
     persua
    1.14
     paralel
    1.13
     motiva
    1.13
     prece
    1.13
    Activations Density 0.015%

    No Known Activations