© 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. 26954
    Prev
    Next
    INDEX
    Explanations

    fill in form below

    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
    ieties
    -0.09
    eree
    -0.09
    hatt
    -0.08
    elmet
    -0.08
     ведÑĮ
    -0.08
    iasm
    -0.08
    iya
    -0.08
     deserted
    -0.08
    iat
    -0.08
    çĤ¬
    -0.08
    POSITIVE LOGITS
    éĶŃ
    0.10
    ç¼ļ
    0.10
    oni
    0.09
    æ£ĺ
    0.09
    è´Ń
    0.09
    ë§Ŀ
    0.09
     Stall
    0.09
    cks
    0.08
    Uni
    0.08
    ä½ļ
    0.08
    Activations Density 0.021%

    No Known Activations