© 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. 34-GEMMASCOPE-RES-131K
    7. 63273
    Prev
    Next
    INDEX
    Explanations

    pw, PW, PwC

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

    The neuron responds to any token containing the letter sequence “pw,” whether in acronyms like PW, company names like PwC, or code/library names like pwntools.

    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
     cox
    -0.74
    
    -0.71
     ▶
    -0.70
    (">
    -0.69
     Mixture
    -0.69
    āciju
    -0.69
     techni
    -0.69
    вости
    -0.68
    CollectionUtils
    -0.68
    Fabio
    -0.67
    POSITIVE LOGITS
     pw
    0.93
    تم
    0.78
     pew
    0.78
    pw
    0.78
     queried
    0.75
    conditional
    0.73
     Pew
    0.72
    wardship
    0.71
    MainActivity
    0.71
    PW
    0.70
    Activations Density 0.016%

    No Known Activations