© 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. 35358
    Prev
    Next
    INDEX
    Explanations

    through the

    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
    еÑģÑĤв
    -0.10
    ensus
    -0.09
    ç»ıè¿ĩ
    -0.09
    	TokenName
    -0.09
    tons
    -0.09
    chluss
    -0.09
    ulton
    -0.09
    itation
    -0.08
    ейÑģÑĤв
    -0.08
    ij
    -0.08
    POSITIVE LOGITS
    ogh
    0.15
    bred
    0.12
    puts
    0.11
    è¿Ļç§įæĸ¹å¼ı
    0.11
    ought
    0.11
    put
    0.11
    -hole
    0.10
    -out
    0.10
    lined
    0.10
    out
    0.10
    Activations Density 0.126%

    No Known Activations