© 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. 34-GEMMASCOPE-RES-131K
    4. 107633
    Prev
    Next
    INDEX
    Explanations

    may not use or may obtain

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

    The neuron detects occurrences of the word “may” (the modal auxiliary indicating permission).

    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
    ssä
    -0.92
     delicioso
    -0.88
     believes
    -0.85
     puts
    -0.83
     put
    -0.82
     goes
    -0.81
    лѣ
    -0.79
     does
    -0.78
     становится
    -0.77
     countless
    -0.77
    POSITIVE LOGITS
     also
    1.59
    also
    1.32
     также
    1.23
    Also
    1.20
     також
    1.09
     também
    1.08
     be
    1.02
    Alternatively
    1.02
     эту
    0.98
     أيضاً
    0.96
    Activations Density 0.014%

    No Known Activations