© 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. 110667
    Prev
    Next
    INDEX
    Explanations

    testimony

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

    The main thing this neuron does is detect mentions of formal testimony (e.g., “testimony” or “testified”).

    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
    ?"
    -3.61
     '
    -3.16
    熥
    -2.83
    تیجه
    -2.77
    ጺ
    -2.75
     ['
    -2.58
    !"
    -2.52
     और
    -2.41
    liehen
    -2.41
    dzić
    -2.38
    POSITIVE LOGITS
    .
    3.00
    佮
    2.75
    ↵
    2.59
    涐
    2.45
     prohibido
    2.36
    (“
    2.28
    {
    2.27
     “
    2.27
    焌
    2.25
     取り付け
    2.23
    Activations Density 0.002%

    No Known Activations