© 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. 34-GEMMASCOPE-RES-131K
    7. 116223
    Prev
    Next
    INDEX
    Explanations

    direction or rtl text

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

    The neuron strongly activates on HTML/XML markup tokens (e.g. tags like “<”, “div”, “span”, “html”, “body” and attributes like “rtl”).

    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
     حديث
    -0.83
    Rap
    -0.75
     uppercase
    -0.74
    Kelvin
    -0.73
    zesi
    -0.69
    τυ
    -0.68
    llave
    -0.68
    จาก
    -0.68
    fiés
    -0.67
     pz
    -0.66
    POSITIVE LOGITS
     direction
    2.66
     Direction
    2.31
    direction
    2.31
    Direction
    2.16
     RTL
    2.13
    rtl
    1.98
     dir
    1.84
    DIRECTION
    1.81
    RTL
    1.80
     DIRECTION
    1.80
    Activations Density 0.081%

    No Known Activations