© 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. 34-GEMMASCOPE-RES-131K
    7. 60717
    Prev
    Next
    INDEX
    Explanations

    dear advice columns

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

    The neuron is picking up high‐level section headers and salutations—especially “Dear X,” openings and other similarly formatted headings.

    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
     Vimeo
    -0.79
     conditioner
    -0.78
     report
    -0.78
     PPT
    -0.74
    äs
    -0.73
     conditioners
    -0.73
     Conditioner
    -0.71
    汇报
    -0.71
     OTP
    -0.71
     Scooby
    -0.71
    POSITIVE LOGITS
     advice
    2.70
    advice
    2.45
     Advice
    2.33
    Advice
    2.31
     column
    2.17
     Dear
    1.91
    Dear
    1.83
     Ask
    1.77
     columns
    1.75
     Column
    1.75
    Activations Density 0.078%

    No Known Activations