© 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. 41715
    Prev
    Next
    INDEX
    Explanations

    thresholds and related concepts

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

    The neuron primarily detects occurrences of the word “threshold” (and its variants) in technical or measurement contexts.

    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
     нашел
    -2.11
    ꦩ
    -1.98
     dipin
    -1.90
    issez
    -1.90
    힛
    -1.86
     عصر
    -1.80
     Inſ
    -1.80
    都说
    -1.80
    面對
    -1.78
    ねば
    -1.78
    POSITIVE LOGITS
    '
    2.41
     conversely
    2.05
     Threshold
    1.99
     thresholds
    1.88
    ﮦ
    1.87
    .'
    1.84
    ttemberg
    1.77
     threshold
    1.77
    ’
    1.77
     that
    1.75
    Activations Density 0.007%

    No Known Activations