© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13941
    Prev
    Next
    INDEX
    Explanations

    option choice mode

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

    This neuron fires on prominent keywords and labels—especially terms in quotes or other emphasis (e.g. menu options, field names, quoted phrases, or standout technical identifiers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     Einer
    -0.87
     作り方
    -0.85
    getPos
    -0.85
     след
    -0.85
    餉
    -0.84
     ویت
    -0.83
    锈钢
    -0.81
    Bras
    -0.81
    τικός
    -0.80
    Teknik
    -0.79
    POSITIVE LOGITS
     option
    1.85
    option
    1.32
     where
    1.16
     mode
    1.15
     opción
    1.14
     вариан
    1.05
     provoque
    1.03
     opcion
    1.02
     Option
    0.98
     wherein
    0.97
    Activations Density 0.067%

    No Known Activations