© 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. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 19139
    Prev
    Next
    INDEX
    Explanations

    ending a thought with "well"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    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
    æķħåIJį
    -0.09
     hence
    -0.08
    æķħ
    -0.08
    éģĤ
    -0.08
    æīĢ以她
    -0.08
    æīĢ以ä»ĸ
    -0.07
    åĽłèĢĮ
    -0.07
    æīĢ以ä»ĸ们
    -0.07
     thus
    -0.07
     lắm
    -0.07
    POSITIVE LOGITS
    èĻ½çĦ¶åľ¨
    0.09
    å®ĥä¸İ
    0.09
    顾åIJįæĢĿä¹ī
    0.09
    å®ĥåı¯ä»¥
    0.08
    å®ĥä¸įä»ħ
    0.08
    ä¼ĹæīĢåij¨çŁ¥
    0.08
    å®ĥåľ¨
    0.08
    å¦Ĥæŀľå¤§å®¶
    0.08
    ...',
    0.08
    å®ĥ
    0.08
    Activations Density 0.088%

    No Known Activations