© 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. 15602
    Prev
    Next
    INDEX
    Explanations

    path joining operations

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

    The neuron detects file‐system path operations, especially occurrences of “path.join” (and related dot‐access on a path object).

    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
     educativas
    -1.27
     and
    -1.23
     each
    -1.05
     territ
    -1.05
    ámide
    -1.03
    髄
    -1.03
     Bibel
    -1.02
     primaire
    -1.02
     avesse
    -1.01
     climático
    -1.00
    POSITIVE LOGITS
     Мы
    1.29
     Kommentare
    1.19
    ه‌ای
    1.18
     They
    1.14
    俚
    1.08
    aju
    1.07
     manžel
    1.04
     bustling
    1.02
    加以
    1.01
    p
    1.00
    Activations Density 0.008%

    No Known Activations