© 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. 6584
    Prev
    Next
    INDEX
    Explanations

    turn, curve, bend, corner

    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
    iete
    -0.10
    atur
    -0.09
    ç»Ī身
    -0.09
    å°ıæŶ
    -0.09
     Josef
    -0.08
    åı°é£İ
    -0.08
    éĨĴäºĨ
    -0.08
    azor
    -0.08
    åľ¨åºĬä¸Ĭ
    -0.08
    éħ¢
    -0.08
    POSITIVE LOGITS
    æĭIJ
    0.17
    转æĬĺ
    0.12
    转弯
    0.11
     corners
    0.11
    å¤Ħ
    0.11
     corner
    0.10
     bend
    0.10
    èĻķ
    0.10
    æĬĺ
    0.10
    isors
    0.10
    Activations Density 0.085%

    No Known Activations