© 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. 11527
    Prev
    Next
    INDEX
    Explanations

    should probably advise

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

    The neuron fires on hedging advice phrased as “should probably.”

    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
    )
    -1.71
     such
    -1.71
     \%$,
    -1.65
    -1.60
            
    -1.52
     “
    -1.49
           
    -1.49
     after
    -1.43
     since
    -1.35
     with
    -1.33
    POSITIVE LOGITS
    
    2.00
     ビキニ
    1.88
     レスト
    1.67
     グラビア
    1.66
    ＇
    1.62
    寕
    1.60
     手绘
    1.58
    ícias
    1.57
    嫪
    1.57
     チラ
    1.56
    Activations Density 0.013%
    No Known Activations