© 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. 57453
    Prev
    Next
    INDEX
    Explanations

    numbered lists or ranges

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

    The neuron is specialized to detect formal grant identifiers, especially NIH-style project and parenthesis‐enclosed grant numbers (e.g. “Parent R01…” and “PA-10-067”).

    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
     Madi
    -0.79
     Belles
    -0.75
     Recognizing
    -0.73
     {[
    -0.71
     Presidency
    -0.71
     μας
    -0.70
     Ottoman
    -0.69
     Refriger
    -0.69
     naso
    -0.68
    ppan
    -0.68
    POSITIVE LOGITS
     Fog
    0.84
    Fog
    0.82
     NIH
    0.80
    PHS
    0.80
     Bethesda
    0.78
    zaak
    0.78
     PI
    0.78
    hans
    0.78
     transition
    0.74
    lange
    0.73
    Activations Density 0.151%

    No Known Activations