© 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. 10-GEMMASCOPE-RES-131K
    7. 11345
    Prev
    Next
    INDEX
    Explanations

    cooking until a state

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

    The neuron is detecting imperative cooking‐instruction verbs and other action words used in recipe steps (e.g., cook, fry, sizzle, microwave, melt).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    轵
    -1.78
    ⢊
    -1.48
    能不能
    -1.47
    蕺
    -1.46
     обладает
    -1.46
    べく
    -1.44
    Composición
    -1.40
    喽
    -1.40
    趼
    -1.39
    termilk
    -1.38
    POSITIVE LOGITS
    -
    1.57
         
    1.48
     Dicapai
    1.48
    The
    1.45
        
    1.42
          
    1.40
     prioriti
    1.40
    If
    1.39
    我们
    1.38
              
    1.38
    Activations Density 0.010%

    No Known Activations