© 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. 31152
    Prev
    Next
    INDEX
    Explanations

    numbers followed by questions

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

    The neuron is specialized to pick out legal-document section headings (like the “Appeal” heading) in text.

    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
    豊富
    -0.69
    glDisable
    -0.68
     Contempor
    -0.68
     Reun
    -0.65
    дження
    -0.65
     választ
    -0.64
    achs
    -0.63
    ހ
    -0.63
    Gover
    -0.63
    Mater
    -0.63
    POSITIVE LOGITS
    pocket
    0.71
     radio
    0.71
    寨
    0.68
     merah
    0.68
     oxy
    0.65
     PP
    0.63
     Brendan
    0.63
     bal
    0.63
     Private
    0.62
    radio
    0.62
    Activations Density 0.061%

    No Known Activations