© 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. 68823
    Prev
    Next
    INDEX
    Explanations

    words starting with Jas

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

    The neuron is triggered by any token containing the character sequence “jas,” e.g. in names or words like “Jas,” “Jasem,” or “jasmine.”

    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
    賒
    -3.41
    鳢
    -3.36
    塤
    -3.17
    涷
    -3.17
     チラシ
    -3.11
    ”),
    -3.08
    呖
    -3.06
    ဴ
    -3.05
     torres
    -3.02
    豋
    -2.97
    POSITIVE LOGITS
    1
    3.39
    8
    3.25
    5
    3.11
    4
    3.05
    2
    3.05
    3
    3.00
     
    2.88
    7
    2.84
     '
    2.73
    相當
    2.70
    Activations Density 0.003%

    No Known Activations