© 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. 64999
    Prev
    Next
    INDEX
    Explanations

    whether you need

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

    The neuron detects correlative choice structures—phrases that present alternatives with “whether … or ….”

    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
    neither
    -2.05
    probably
    -1.84
    mainly
    -1.77
    
    -1.70
    basically
    -1.66
     carlo
    -1.65
    both
    -1.65
    certainly
    -1.62
     praticamente
    -1.61
    perhaps
    -1.60
    POSITIVE LOGITS
     or
    3.13
     our
    2.59
     this
    2.41
     there
    2.27
     another
    2.19
     we
    2.13
     your
    1.98
     encompassing
    1.92
     at
    1.89
     extraordinarily
    1.83
    Activations Density 0.067%

    No Known Activations