© 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. 54498
    Prev
    Next
    INDEX
    Explanations

    the environment

    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
    åĩ«
    -0.08
    åij
    -0.08
    åħ³éĶ®æŶåĪ»
    -0.08
    Unnamed
    -0.08
    omics
    -0.08
    äºĭæķħåıijçĶŁ
    -0.08
     anom
    -0.07
     mere
    -0.07
     fug
    -0.07
    çī¹å®ļ
    -0.07
    POSITIVE LOGITS
    涯
    0.10
     Burnett
    0.09
    yas
    0.09
    èĭ±æł¼
    0.09
    ä¹Ŀ大
    0.09
    溯æºIJ
    0.09
    yles
    0.09
    icina
    0.09
    alic
    0.09
    /libs
    0.09
    Activations Density 0.411%

    No Known Activations