© 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. 34-GEMMASCOPE-RES-131K
    7. 31728
    Prev
    Next
    INDEX
    Explanations

    directory paths in code

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

    The neuron fires strongly on code tokens used to construct and reference file paths (e.g. __DIR__, __FILE__, dirname(), require/include with “. ’/” concatenation).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     only
    -0.90
    :"",
    -0.88
     this
    -0.88
     biên
    -0.87
    :[],
    -0.84
     meticulously
    -0.84
     knew
    -0.83
    朕
    -0.82
    ുള്ള
    -0.81
     made
    -0.81
    POSITIVE LOGITS
     vijf
    1.10
     NINE
    0.98
     EIGHT
    0.94
    five
    0.94
     cinq
    0.91
     `/
    0.89
     './../
    0.89
    ..\..\
    0.88
     '..
    0.88
     STACK
    0.87
    Activations Density 0.019%

    No Known Activations