© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10415
    Prev
    Next
    INDEX
    Explanations

    useful

    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
     usual
    -0.10
    åŁºæľ¬æĥħåĨµ
    -0.10
    aji
    -0.09
    æīĢåģļçļĦ
    -0.09
    å¿ĥ缮
    -0.08
    åĽŀäºĭ
    -0.08
    è´¨éĩıåĴĮ
    -0.08
    овÑĭй
    -0.08
    绽
    -0.08
    ched
    -0.08
    POSITIVE LOGITS
     ÃŃch
    0.16
    NESS
    0.15
    -looking
    0.12
    FUL
    0.11
    lest
    0.11
    ness
    0.11
    ities
    0.11
    fully
    0.11
    /help
    0.10
     dÃłng
    0.10
    Activations Density 0.030%

    No Known Activations