© 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. 38407
    Prev
    Next
    INDEX
    Explanations

    number and percentage formatting

    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
    ."]
    -0.10
    åħļæĶ¿
    -0.10
    æĿª
    -0.09
    )":
    -0.08
    .","
    -0.08
    é¢ĩæľī
    -0.08
    ahoo
    -0.08
    å®ŀè´¨æĢ§
    -0.08
    æľ¬èº«çļĦ
    -0.08
    é¢ĩ为
    -0.08
    POSITIVE LOGITS
    \t
    0.09
    å°±è¿Ļä¹Ī
    0.08
    å°±è¿Ļæł·
    0.08
     \"
    0.08
    orate
    0.08
    åIJ¬çĿĢ
    0.08
    è¿Ļä¹Īå¤ļ
    0.08
    isOk
    0.08
     Pax
    0.08
    olley
    0.08
    Activations Density 1.085%

    No Known Activations