© 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. 55699
    Prev
    Next
    INDEX
    Explanations

    conditional questions

    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
    éĢĤæŶ
    -0.10
    erer
    -0.09
    esium
    -0.08
    .sep
    -0.08
    ulls
    -0.08
    æĺİæĻ°
    -0.08
    моÑĤ
    -0.08
    inq
    -0.08
    revision
    -0.08
    erring
    -0.07
    POSITIVE LOGITS
    å«Į
    0.10
    æĺ¯ä»Ģä¹Īåij¢
    0.09
    ç¼ĩ
    0.09
    åīįä¸ī
    0.09
    éĢīæĭ©
    0.09
    å¿ĥ头
    0.09
    DEBUG
    0.08
    éģ¸æĵĩ
    0.08
    âĦ
    0.08
    éĢīåıĸ
    0.08
    Activations Density 0.023%

    No Known Activations