© 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. 61959
    Prev
    Next
    INDEX
    Explanations

    Danticat, Cameron, DiCaprio, Cabrera, Bibiena

    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
    ../../
    -0.09
    ìĿį
    -0.09
    iming
    -0.08
    ../../../
    -0.08
    UGHT
    -0.08
    ëĭ¨ì²´
    -0.08
    itized
    -0.08
    æľ¬åľ°
    -0.08
    áĢī
    -0.08
    ocop
    -0.08
    POSITIVE LOGITS
    té
    0.13
    岸
    0.11
    è¿Ļæł·çļĦè¯Ŀ
    0.10
    æķijçģ¾
    0.09
    ropa
    0.09
     fri
    0.08
    ité
    0.08
    åĵĪå°Ķ
    0.08
    licate
    0.08
    staking
    0.08
    Activations Density 0.001%

    No Known Activations