© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 104644
    Prev
    Next
    INDEX
    Explanations

    Unix file paths

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

    directory paths in Unix/Linux file systems, particularly those starting with "/var" or "/etc" or "/usr".

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron detects filesystem path components (e.g. directory names like etc, var, usr, tmp) in code or commands.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     confirmado
    -1.70
    jenigen
    -1.70
     which
    -1.60
    さり
    -1.53
     tomado
    -1.53
     dubbed
    -1.53
     kebutuhan
    -1.53
    くり
    -1.52
    サ
    -1.52
    轳
    -1.49
    POSITIVE LOGITS
    you
    2.03
    warnai
    1.85
     /
    1.81
    Однако
    1.66
    における
    1.66
    橚
    1.64
    還會
    1.63
    あなたは
    1.61
    Хороший
    1.59
    這些
    1.56
    Activations Density 0.017%

    No Known Activations