© 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-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 116308
    Prev
    Next
    INDEX
    Explanations

    leaks and leakage

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

    The neuron activates on mentions of “leak” (and its forms like leaking, leaks, leakage) across contexts.

    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
     Bac
    -0.88
     ash
    -0.82
     incin
    -0.75
     Acquisition
    -0.73
     zah
    -0.72
     reflexiones
    -0.71
    擄
    -0.69
     elbise
    -0.69
    灰
    -0.69
    Bac
    -0.68
    POSITIVE LOGITS
     leak
    4.84
     leaks
    4.53
     leakage
    4.28
     leaking
    4.13
    leak
    3.86
    Leak
    3.63
     Leak
    3.58
     leaked
    3.56
     Leakage
    3.34
     Leaks
    3.19
    Activations Density 0.123%

    No Known Activations