© 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. 57707
    Prev
    Next
    INDEX
    Explanations

    causation and relation

    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
    æł¹æį®ä¸įåIJĮ
    -0.10
    .toHexString
    -0.09
    æĶ¾çľ¼
    -0.09
    éĢ¸
    -0.08
    åĽ´è§Ĥ
    -0.08
     tÃł
    -0.08
     live
    -0.08
    åħ³éĶ®æĺ¯
    -0.08
    ÙĦس
    -0.08
    -live
    -0.08
    POSITIVE LOGITS
    åĽłæŀľ
    0.17
     contribution
    0.13
     caus
    0.12
     attrib
    0.12
     causal
    0.11
    tribution
    0.11
    è´¡çĮ®
    0.11
    èĤ¯å®ļæĺ¯
    0.11
    åĬŁåĬ³
    0.10
     Contribution
    0.10
    Activations Density 0.111%

    No Known Activations