© 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. 10-GEMMASCOPE-RES-131K
    4. 75421
    Prev
    Next
    INDEX
    Explanations

    willing, unwilling, prepared, would

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

    This neuron detects constructions where a desire/volition word (e.g. willing, unwilling, prepared, refused, like, accept, bear, submit) is immediately followed by the infinitive marker “to.”

    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
    //
    -1.78
     will
    -1.69
     pelaksanaan
    -1.48
     españolas
    -1.45
    !!!
    -1.43
    はい
    -1.43
    ဴ
    -1.41
    x
    -1.40
     aisladas
    -1.40
    歳の
    -1.40
    POSITIVE LOGITS
    THERE
    1.60
    ufficient
    1.56
     to
    1.55
     cependant
    1.50
    麁
    1.48
    however
    1.45
    bies
    1.45
    simplified
    1.40
     Anyone
    1.39
    previously
    1.38
    Activations Density 0.005%

    No Known Activations