© 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. 26660
    Prev
    Next
    INDEX
    Explanations

    alternativeMainCharactersRoutesBoom Boommatching processderivative (Channel3Actual Play

    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
    uges
    -0.10
    ensus
    -0.10
    åĨľè´¸
    -0.10
    éĥ
    -0.09
     Buckley
    -0.09
    çĦķ
    -0.09
    ilion
    -0.09
    ãģĿãĤĮãģ«
    -0.09
    è®°èĢħåľ¨
    -0.09
    phins
    -0.09
    POSITIVE LOGITS
    ä¹ħ
    0.08
    就好
    0.08
    åı¯æĥľ
    0.08
    æīĭæĦŁ
    0.08
    pal
    0.07
    åīįä¸ĸ
    0.07
     Clem
    0.07
    å°±æľīäºĨ
    0.07
     vs
    0.07
     gan
    0.07
    Activations Density 1.935%

    No Known Activations