© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 18743
    Prev
    Next
    INDEX
    Explanations

    weather permitting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _traffic
    -0.10
    æįĤ
    -0.10
    èĨĽ
    -0.10
    以éĺ²
    -0.09
    é«ĺä»·
    -0.09
    åĻ±
    -0.09
    Outlet
    -0.09
    åłµå¡ŀ
    -0.09
    é«ĺå³°
    -0.09
    çļĦç¡®
    -0.09
    POSITIVE LOGITS
    å®ļçĤ¹
    0.10
    缮çĿ¹
    0.09
    éĹ»
    0.09
     MOD
    0.09
     thousands
    0.08
    MOD
    0.08
    èĤ¥
    0.08
     prov
    0.07
    Locator
    0.07
    çļĤ
    0.07
    Activations Density 0.069%

    No Known Activations