© 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. 11089
    Prev
    Next
    INDEX
    Explanations

    DL, PL, CL, ML, NL, SL abbreviations

    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
    ulant
    -0.09
    ä¸Ģè·¯
    -0.09
    ials
    -0.09
    lesi
    -0.08
     bst
    -0.08
    iets
    -0.08
    oir
    -0.08
    ÙĬض
    -0.08
    ç°¸
    -0.08
    ulate
    -0.07
    POSITIVE LOGITS
    ateral
    0.13
    raries
    0.13
    loyd
    0.11
    ambda
    0.11
    æĪ¬
    0.11
    amma
    0.11
    imited
    0.10
    ening
    0.10
    IFT
    0.09
    ifetime
    0.09
    Activations Density 0.202%

    No Known Activations