© 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. 10-GEMMASCOPE-RES-131K
    7. 8844
    Prev
    Next
    INDEX
    Explanations

    source directory paths

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

    The neuron strongly activates on occurrences of “src” in file‐path strings (e.g., directory paths referencing a src folder).

    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
    眛
    -1.64
     travaillons
    -1.57
    ᯾
    -1.53
     ſp
    -1.47
     and
    -1.47
    bbbb
    -1.44
    犼
    -1.42
    eſ
    -1.41
    鸸
    -1.40
     Danh
    -1.38
    POSITIVE LOGITS
    3
    2.48
    4
    2.36
    9
    2.25
    7
    2.19
    Some
    2.13
    0
    2.02
    Other
    2.02
    8
    2.00
    2
    1.98
    6
    1.91
    Activations Density 0.001%

    No Known Activations