© 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. 42012
    Prev
    Next
    INDEX
    Explanations

    mixed or conflicting results

    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
    orium
    -0.08
    ovable
    -0.08
    æĹ¢çĦ¶æĺ¯
    -0.07
    Normally
    -0.07
    ateg
    -0.07
    acol
    -0.07
     manually
    -0.07
    容
    -0.07
    undy
    -0.07
    ä¸ĩä¸Ģ
    -0.07
    POSITIVE LOGITS
     findings
    0.13
     reports
    0.12
    å®ŀè¯ģ
    0.12
     statistically
    0.11
     reported
    0.11
     statistical
    0.11
    çłĶ究åıijçİ°
    0.10
    çłĶ究æĬ¥åijĬ
    0.10
    ç»Łè®¡æķ°æį®
    0.10
     statistics
    0.10
    Activations Density 0.322%

    No Known Activations