© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130193
    Prev
    Next
    INDEX
    Explanations

    This neuron specifically detects the word “having” when used to introduce a participial clause.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    having + past participle

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     was
    -2.31
     بارد
    -2.25
    ﹔
    -2.22
    𓆜
    -2.11
    戗
    -2.02
     robusto
    -2.02
    ፑ
    -1.98
    ﮦ
    -1.95
    gencias
    -1.94
    Ꮵ
    -1.93
    POSITIVE LOGITS
     been
    2.36
     to
    2.16
     удалить
    2.03
     Focusing
    2.02
    しまった
    1.98
    ものに
    1.93
    幸好
    1.91
    
    1.81
     Reveals
    1.80
    もの
    1.78
    Activations Density 0.025%

    No Known Activations