© 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. 48637
    Prev
    Next
    INDEX
    Explanations

    motion for rehearing denied

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

    The neuron is strongly activated by numeric tokens (e.g. years, volume/issue numbers, page ranges) in academic citations.

    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
     spaces
    -0.79
     Spaces
    -0.76
     six
    -0.73
     four
    -0.73
     pertinent
    -0.73
     Consume
    -0.73
     seven
    -0.72
    oore
    -0.71
    Spaces
    -0.70
    Aviso
    -0.69
    POSITIVE LOGITS
    ály
    0.85
     kitty
    0.82
    circledR
    0.79
    ểu
    0.78
    "");
    0.78
    UniformLocation
    0.76
    нє
    0.73
    rildi
    0.73
     doet
    0.71
    gede
    0.71
    Activations Density 0.017%

    No Known Activations