© 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. 57953
    Prev
    Next
    INDEX
    Explanations

    improved, increased, maximum, better

    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
    çĶ¨æĿ¥
    -0.10
    ahn
    -0.10
    OCK
    -0.09
    æľīç§į
    -0.09
    çļĦæĵįä½ľ
    -0.09
    erin
    -0.09
    å·§
    -0.09
     lượt
    -0.09
    æī©
    -0.09
    æĺ¯çĶ¨æĿ¥
    -0.09
    POSITIVE LOGITS
    æĽ´å¥½çļĦ
    0.10
    çŃĸ
    0.10
    ear
    0.09
     underst
    0.09
    .Fields
    0.08
    EAR
    0.08
     earlier
    0.08
    æĽ´å¥½åľ°
    0.08
    Council
    0.08
    .Dialog
    0.08
    Activations Density 0.059%

    No Known Activations