© 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. 23105
    Prev
    Next
    INDEX
    Explanations

    Äģ followed by vowels, l, r, or cis

    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
    ting
    -0.10
     Wake
    -0.09
    اÙĨ
    -0.09
    ansible
    -0.09
    åħ³éĶ®æŶåĪ»
    -0.08
     Hamp
    -0.08
    ub
    -0.08
    aine
    -0.08
    wake
    -0.08
    èĨĿ
    -0.08
    POSITIVE LOGITS
    ãĤĤãģ£ãģ¨
    0.11
    媳
    0.11
    棣
    0.11
    æīĢæīĢ
    0.11
    INGS
    0.10
    湫
    0.10
    ãģĹãģ¾ãģĨ
    0.10
    ãģ¯ãģĺ
    0.10
    å§ĭ建
    0.09
     Bundy
    0.09
    Activations Density 0.088%

    No Known Activations