© 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. 84480
    Prev
    Next
    INDEX
    Explanations

    html element identifiers

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

    The neuron is looking for CSS class name tokens (i.e. class attribute values) in markup.

    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
    érés
    -1.07
    OIR
    -1.05
     levering
    -1.02
     faulting
    -1.02
     gourmand
    -1.01
     campa
    -1.01
     grafting
    -0.99
     Atención
    -0.98
     thép
    -0.97
     decrypted
    -0.96
    POSITIVE LOGITS
     wanted
    1.17
     first
    1.16
     now
    1.05
     ohne
    1.02
     input
    1.01
     websites
    1.00
     yellow
    0.96
     really
    0.96
     type
    0.93
     second
    0.92
    Activations Density 0.009%

    No Known Activations