© 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. 51786
    Prev
    Next
    INDEX
    Explanations

    Looking at the activations, all the values shown are 0, which means this neuron doesn't activate for any of the tokens in these documents. The neuron does not appear to find anything in these examples.

    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
     [âĢ¦]
    -0.08
    âĢ¦âĢ¦
    -0.07
    æĪĸèĢħ说
    -0.07
    åĽºæľī
    -0.07
     presumably
    -0.07
     approximately
    -0.07
    âĢ¦.
    -0.07
     furthermore
    -0.06
    æĸ¹éĿ¢çļĦ
    -0.06
     necess
    -0.06
    POSITIVE LOGITS
    _<
    0.09
    ochrome
    0.09
     <",
    0.08
     \''
    0.08
     tiap
    0.08
    ariales
    0.08
    ()",
    0.08
    é¤ĬçĶŁ
    0.08
    ↵				↵
    0.08
    nish
    0.08
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.