© 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. 34-GEMMASCOPE-RES-131K
    7. 93541
    Prev
    Next
    INDEX
    Explanations

    Dear Don, Q:, e prime, runner:

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

    The neuron specializes in spotting URL‐like or code/metadata tokens—things such as path fragments, numeric IDs, config keys, or other technical strings.

    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
     for
    -0.98
     our
    -0.94
     Depending
    -0.94
     those
    -0.94
     so
    -0.93
     some
    -0.88
     its
    -0.88
     four
    -0.87
     then
    -0.87
     anaknya
    -0.87
    POSITIVE LOGITS
    如果你
    0.96
     香水
    0.93
    钾
    0.91
    「
    0.91
     MIRROR
    0.90
     максимально
    0.90
     netizen
    0.89
     viſ
    0.89
     ὑ
    0.87
     momente
    0.87
    Activations Density 0.000%

    No Known Activations