© 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. 39827
    Prev
    Next
    INDEX
    Explanations

    hidden + concepts or entities

    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
    èĶ¼
    -0.10
    ðŁı»
    -0.09
    das
    -0.08
    naments
    -0.08
    ikon
    -0.08
    iets
    -0.08
    ÑĪиб
    -0.08
    éĥ¡
    -0.08
    cloth
    -0.08
    imbus
    -0.08
    POSITIVE LOGITS
    annel
    0.11
    wealth
    0.11
    ized
    0.09
    วร
    0.09
    ness
    0.09
    è§Ħå¾ĭ
    0.08
    pell
    0.08
    morph
    0.08
     Chef
    0.08
    æĸĹäºī
    0.08
    Activations Density 0.014%

    No Known Activations