© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8691
    Prev
    Next
    INDEX
    Explanations

    file paths and directory names

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

    The neuron fires on filesystem path components, especially absolute directory names (e.g. leading “/”, folder names like “srv,” “mediawiki,” “Documents,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     introductory
    0.61
    hető
    0.55
    꾀
    0.54
     базо
    0.53
     Lipp
    0.52
    தந்திர
    0.52
     durability
    0.52
     mainline
    0.51
    俏
    0.51
     lifting
    0.50
    POSITIVE LOGITS
    OneDrive
    1.00
    Dropbox
    0.98
    Documents
    0.96
    Projects
    0.93
    Downloads
    0.88
    Desktop
    0.87
     Documents
    0.85
     Projects
    0.80
     OneDrive
    0.79
    AppData
    0.78
    Activations Density 0.079%

    No Known Activations