© 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. 13951
    Prev
    Next
    INDEX
    Explanations

    choice and volition

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

    The neuron flags words and phrases expressing voluntary choice or decision‐making (e.g. choose, choice, voluntarily, choosing).

    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
     After
    -1.49
     in
    -1.43
     as
    -1.40
    ්‍
    -1.36
     if
    -1.35
     until
    -1.30
     they
    -1.25
     adequate
    -1.23
    lenül
    -1.22
     “
    -1.20
    POSITIVE LOGITS
    albeit
    1.62
    𖣔
    1.59
     rafraî
    1.59
    
    1.55
    錞
    1.53
     ་
    1.50
    ⁛
    1.49
    ーーーー
    1.49
     épar
    1.48
     salmón
    1.45
    Activations Density 0.054%

    No Known Activations