© 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. 47493
    Prev
    Next
    INDEX
    Explanations

    software interface, remote, machine data, Willie Nelson

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

    The neuron identifies promotional “feature‐listing” verbs and phrases that introduce product capabilities (e.g. offers, provides, integrated, available).

    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
    izability
    -1.02
    gensen
    -0.96
    麩
    -0.93
     Algunos
    -0.92
     telewiz
    -0.92
    áky
    -0.91
     tohoto
    -0.91
    ǡ
    -0.91
    lovely
    -0.90
    bishop
    -0.90
    POSITIVE LOGITS
     optional
    1.50
     user
    1.30
     optionally
    1.19
    optional
    1.15
     powerful
    1.13
     up
    1.10
     password
    1.05
     operator
    1.04
     stored
    1.04
     proven
    1.00
    Activations Density 0.050%

    No Known Activations