© 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. 38624
    Prev
    Next
    INDEX
    Explanations

    universal

    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
    riel
    -0.10
    nees
    -0.08
    azo
    -0.08
    æł¹æľ¬
    -0.08
    ereal
    -0.08
    Kel
    -0.08
    ATS
    -0.08
    è½´
    -0.08
    ıè¿°
    -0.08
    Maps
    -0.08
    POSITIVE LOGITS
    universal
    0.13
     ÑĥнивеÑĢÑģ
    0.13
     Universal
    0.13
     universal
    0.12
    Universal
    0.12
    iversal
    0.12
    éĢļçĶ¨
    0.11
    -purpose
    0.10
    è¦ĨçĽĸ
    0.10
    retch
    0.09
    Activations Density 0.060%

    No Known Activations