© 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. 28035
    Prev
    Next
    INDEX
    Explanations

    human figures, skeletons, androids

    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
    ä¸ĢåĿĹ
    -0.11
    åIJ¬çĿĢ
    -0.10
    é¦Ī
    -0.10
    éĹ»
    -0.10
    èľĥ
    -0.09
    éĹŃ
    -0.09
    egan
    -0.09
     boss
    -0.09
    éģ¥è¿ľ
    -0.08
    -tip
    -0.08
    POSITIVE LOGITS
    ç«Ļç«ĭ
    0.11
    çĻ½è¡£
    0.10
    身ä¸Ĭ
    0.10
    ç«ĻçĿĢ
    0.10
    aint
    0.10
    çĶ°
    0.09
    身影
    0.09
    ç«Ļéķ¿
    0.09
    ç©¿è¡£
    0.09
    人身
    0.09
    Activations Density 0.109%

    No Known Activations