© 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. 14596
    Prev
    Next
    INDEX
    Explanations

    compared to or with

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

    The neuron fires on comparative constructions, notably the word “compared” when introducing “compared X to/with Y.”

    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
    ’,
    -1.82
    ,”
    -1.80
    ,’
    -1.78
    
    -1.70
     residente
    -1.59
    ”,
    -1.58
     informativo
    -1.53
    鱟
    -1.49
    .’
    -1.47
     …”
    -1.47
    POSITIVE LOGITS
     From
    1.79
    呦
    1.75
     añade
    1.69
    TE
    1.65
     Our
    1.62
     '
    1.60
     These
    1.55
     Then
    1.51
     Want
    1.51
    ('
    1.48
    Activations Density 0.006%

    No Known Activations