© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 38268
    Prev
    Next
    INDEX
    Explanations

    code snippets or probability problems

    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
    8
    -0.09
    å°»
    -0.08
     Grim
    -0.08
    èĵ¼
    -0.08
    6
    -0.08
    earer
    -0.08
    åĴĮåıijå±ķ
    -0.08
    ием
    -0.08
    7
    -0.08
    åĬª
    -0.08
    POSITIVE LOGITS
    åħ®
    0.11
     alike
    0.09
    ãĤĿ
    0.09
    .Ptr
    0.09
    çĽ²
    0.08
     Bauer
    0.08
    æľ¬åľŁ
    0.08
    ifice
    0.08
    绸
    0.08
    çĦ¶åIJİåĨį
    0.08
    Activations Density 1.592%

    No Known Activations