© 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. 59569
    Prev
    Next
    INDEX
    Explanations

    say the least

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

    This neuron is specialized to detect instances of the word “say,” especially in the idiomatic phrase “to say the least.”

    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
     was
    -2.42
     formado
    -2.02
    ৄ
    -1.95
    感じ
    -1.76
     Also
    -1.74
    
    -1.73
    ± 
    -1.72
     auténtica
    -1.68
    ?"
    -1.66
     determinada
    -1.66
    POSITIVE LOGITS
    ating
    2.42
    er
    1.97
    io
    1.95
    )
    1.90
    our
    1.84
    ge
    1.82
    you
    1.78
    ية
    1.75
    thus
    1.74
    とりあえず
    1.73
    Activations Density 0.008%

    No Known Activations