© 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. 40161
    Prev
    Next
    INDEX
    Explanations

    pointing out specific things

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

    The neuron fires on demonstrative‐lead statements—phrases that begin with “this/these” introducing or labelling something (e.g. “This is…,” “These haircuts…”).

    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
     Imágenes
    -1.00
     naszego
    -0.96
    ОТ
    -0.94
    纤
    -0.94
     Añade
    -0.94
     нашего
    -0.94
     особли
    -0.91
     Porque
    -0.91
     zosta
    -0.89
     Diseño
    -0.89
    POSITIVE LOGITS
    削減
    1.07
     WaitForSeconds
    1.05
    ]));
    
    1.02
     Vikipedi
    0.98
    setState
    0.96
     onOptions
    0.94
    NewGuid
    0.93
    lidean
    0.93
     different
    0.93
     inappropriate
    0.93
    Activations Density 0.022%

    No Known Activations