© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 93775
    Prev
    Next
    INDEX
    Explanations

    Q: question prefix

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

    The neuron strongly activates on web‐address tokens—especially “.com” domains (e.g. “Agoda.com”).

    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
     iub
    -0.85
    denken
    -0.83
     Trader
    -0.82
    ంది
    -0.82
    妧
    -0.82
     обви
    -0.81
     druga
    -0.80
     delicacy
    -0.80
     ulong
    -0.80
     bym
    -0.79
    POSITIVE LOGITS
     Flight
    1.06
     even
    0.92
    Flight
    0.89
     let
    0.86
     détermination
    0.84
     because
    0.84
     separated
    0.83
    nejšie
    0.82
     not
    0.82
    ligere
    0.81
    Activations Density 0.002%

    No Known Activations