© 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. 55722
    Prev
    Next
    INDEX
    Explanations

    fetal

    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
    icut
    -0.10
    eous
    -0.09
    Sock
    -0.09
    ì±Ħ
    -0.08
    outs
    -0.08
    ä½İæĪIJæľ¬
    -0.08
    oded
    -0.08
     اÙĦعÙħ
    -0.08
    uru
    -0.08
    ç»Ļ人
    -0.08
    POSITIVE LOGITS
    娩
    0.10
    çĶ¨åĬĽ
    0.10
    å±Ī
    0.09
    åħĴ
    0.09
    åĺ»åĺ»
    0.09
    åĸı
    0.09
    éħ¿
    0.09
    éķ¿å¤§äºĨ
    0.09
    åįľ
    0.08
    æľºæ²¹
    0.08
    Activations Density 0.025%

    No Known Activations