© 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. 10-GEMMASCOPE-RES-131K
    7. 43752
    Prev
    Next
    INDEX
    Explanations

    dependency on which

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

    The neuron activates strongly on occurrences of “which” in relative‐clause phrases (e.g. “on which …”).

    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
    ’
    -2.47
    geſ
    -2.13
    0
    -1.95
    e
    -1.95
    The
    -1.88
    6
    -1.88
     you
    -1.88
     have
    -1.82
    2
    -1.77
     was
    -1.74
    POSITIVE LOGITS
     behalf
    2.13
     maž
    1.88
    愎
    1.87
    髎
    1.85
     mayoría
    1.85
     merits
    1.84
     протяжении
    1.84
    ⮚
    1.74
    tedrale
    1.74
     svůj
    1.74
    Activations Density 0.012%

    No Known Activations