© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 112847
    Prev
    Next
    INDEX
    Explanations

    checking or performing actions

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

    This neuron detects action‐oriented verbs—especially those used as function or method names—in code.

    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
     lauded
    -2.05
     serene
    -2.02
     quintessential
    -1.96
     花纹
    -1.94
     Dinas
    -1.93
    
    -1.86
     acclaimed
    -1.84
    ⏑
    -1.84
     delightfully
    -1.82
     strikingly
    -1.77
    POSITIVE LOGITS
    Three
    2.50
    Some
    2.48
    Not
    2.44
    Because
    2.42
     Some
    2.39
     three
    2.39
     four
    2.38
    Without
    2.36
    Two
    2.34
     five
    2.31
    Activations Density 0.104%

    No Known Activations