© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 79735
    Prev
    Next
    INDEX
    Explanations

    code separators and comments

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

    The neuron activates on comment markers that denote metadata or section boundaries—e.g. “begin_description,” “BEGIN LICENSE BLOCK,” and “<auto‐generated>.”

    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
     for
    -1.56
    providedIn
    -1.45
     both
    -1.35
     ALL
    -1.35
     using
    -1.34
    ;
    -1.26
     it
    -1.26
     any
    -1.25
     Souver
    -1.25
     all
    -1.23
    POSITIVE LOGITS
     PREVENT
    1.39
     vaisseaux
    1.39
    ᠌
    1.39
     pietre
    1.36
     AVOID
    1.33
    
    1.32
     frontières
    1.30
     réactions
    1.26
    yordu
    1.26
    küche
    1.26
    Activations Density 0.001%

    No Known Activations