© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 87139
    Prev
    Next
    INDEX
    Explanations

    alliance for

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

    The neuron activates on names of formal groups or programs—especially “Alliance,” “Coalition,” or “Partnership” titles that use “for” (or similar prepositions) to introduce their focus area.

    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
    Huff
    -0.89
    PCL
    -0.82
    buat
    -0.79
    palo
    -0.79
    COT
    -0.79
    :${
    -0.75
    saber
    -0.75
     wür
    -0.75
    Refreshing
    -0.74
    Saber
    -0.74
    POSITIVE LOGITS
    促销
    0.84
    也不
    0.81
     fornis
    0.75
    cauciones
    0.75
     ám
    0.73
    mbor
    0.71
     Exp
    0.71
    显然
    0.71
    Nonetheless
    0.71
     fringes
    0.70
    Activations Density 0.015%

    No Known Activations