© 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. 22-GEMMASCOPE-RES-131K
    7. 73838
    Prev
    Next
    INDEX
    Explanations

    opposing or rival positions

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

    The neuron fires on numerical tokens—dates, statistics, case numbers, percentages, page/volume figures, and other digits sequences.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
    /$
    -0.91
    *$
    -0.89
     сахара
    -0.85
     defiance
    -0.85
     lisää
    -0.85
    ")
    -0.85
     loops
    -0.83
    dshaw
    -0.83
    inWeight
    -0.82
    idiary
    -0.82
    POSITIVE LOGITS
     opposing
    1.78
     rival
    1.47
     opponents
    1.29
     opposite
    1.27
     competing
    1.27
    Conversely
    1.22
     opponent
    1.19
     противополо
    1.18
    opposite
    1.14
    Opposite
    1.14
    Activations Density 0.123%

    No Known Activations