© 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. 40026
    Prev
    Next
    INDEX
    Explanations

    Scarlet/Scarlett/Velvet names

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

    The neuron is triggered by distinctive capitalized single-word proper names (i.e. named entities).

    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
    茑
    -2.66
    ⸫
    -2.47
    
    -2.47
    聍
    -2.45
    ޏ
    -2.42
    鲠
    -2.36
    ৣ
    -2.36
    諼
    -2.31
    ㌧
    -2.30
    ሏ
    -2.30
    POSITIVE LOGITS
    0
    3.38
    2
    3.30
    6
    3.28
    4
    3.23
    3
    3.16
    7
    2.98
    9
    2.81
     This
    2.80
    e
    2.78
    8
    2.73
    Activations Density 0.005%

    No Known Activations