© 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
    2. Qwen3.5-9B Base
    3. 15-QWENSCOPE-RES-64K
    4. 29540
    Prev
    Next
    INDEX
    Explanations

    transition words or phrases in mathematical problem-solving contexts, particularly "however," "for," "I thought," and similar expressions that introduce a new approach or consideration.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jrosseruk
    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
    zio
    -0.07
    zl
    -0.07
    OrCreate
    -0.06
    ç»ıå®¡æŁ¥
    -0.06
    å°¤åħ¶åľ¨
    -0.06
    "...
    -0.06
     Tow
    -0.06
    òm
    -0.06
    ×ķר×ļ
    -0.06
    elize
    -0.06
    POSITIVE LOGITS
    ä½łæľīä»Ģä¹Ī
    0.09
    丽ä¸Ŀ
    0.08
    åľ¨è¿Ļç§įæĥħåĨµä¸ĭ
    0.07
    oir
    0.07
    æĽ´åºĶ该
    0.07
    åĪĿä¸ŃçĶŁ
    0.07
    bomb
    0.07
    å¸ĮæľĽå¤§å®¶èĥ½å¤Ł
    0.07
    æıIJä¾ĽæĽ´å¤ļçļĦ
    0.07
    å¦Ĥæŀľèĥ½å¤Ł
    0.06
    Activations Density 0.000%

    No Known Activations