© 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. 22-GEMMASCOPE-RES-131K
    7. 9592
    Prev
    Next
    INDEX
    Explanations

    html input elements

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

    The neuron fires on HTML markup tokens—particularly the angle‐bracket delimiters and attribute/value fragments of HTML tags.

    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
     matematika
    -1.02
     onLoad
    -1.00
     politika
    -0.97
     "!
    -0.97
     to
    -0.95
    ༼
    -0.94
     laki
    -0.94
     игрушки
    -0.93
     diskusi
    -0.93
     translateY
    -0.93
    POSITIVE LOGITS
    Mng
    1.02
     ditto
    0.97
     completing
    0.94
    不已
    0.91
     trenger
    0.91
    Belo
    0.88
    着脸
    0.88
    镜子
    0.87
     kohdetta
    0.87
     Replaced
    0.86
    Activations Density 0.002%

    No Known Activations