© 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. 69873
    Prev
    Next
    INDEX
    Explanations

    rule definition or usage

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

    The neuron fires on XML‐ or HTML‐style tag names (e.g. elements like <rewrite>, <rules>, <rule>, <match>, <action>, etc.).

    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
     mas
    -0.76
    シオ
    -0.75
     though
    -0.75
    личие
    -0.71
    じて
    -0.71
     خوا
    -0.71
     pilgrims
    -0.70
     meanings
    -0.70
     trous
    -0.70
    тере
    -0.69
    POSITIVE LOGITS
    Rule
    1.02
     Rule
    0.99
     rule
    0.90
     getRule
    0.83
     RULE
    0.82
    getContentPane
    0.78
    ลักษ
    0.78
    (){}
    0.76
     regola
    0.75
    RULE
    0.74
    Activations Density 0.043%

    No Known Activations