© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 38798
    Prev
    Next
    INDEX
    Explanations

    msg, vals, Grav

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

    The neuron strongly activates on the token “msg” (e.g. parts of identifiers or attributes named msg).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     we
    -2.50
     Have
    -2.45
    {~
    -2.42
    ieteur
    -2.27
    </u>
    -2.23
    {
    
    -2.19
     groundbreaking
    -2.14
    with
    -2.13
    but
    -2.13
    </th>
    -2.13
    POSITIVE LOGITS
    一旁
    2.34
     последу
    2.30
     &
    2.13
    虓
    2.08
     hisz
    1.97
     hele
    1.91
    颡
    1.88
     experimente
    1.87
     thc
    1.86
    Of
    1.82
    Activations Density 0.001%

    No Known Activations