© 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. 23847
    Prev
    Next
    INDEX
    Explanations

    lists of names or items

    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
    ç³Ł
    -0.10
    人们çļĦ
    -0.10
    eref
    -0.09
    åıijæĮ¥
    -0.09
    æķ°çĽ®
    -0.09
    åıij票
    -0.09
    osos
    -0.08
    å¥ı
    -0.08
    abbrev
    -0.08
    ç½¢
    -0.08
    POSITIVE LOGITS
    å¤ļ人
    0.09
    æľīåºı
    0.08
    éĥĵ
    0.08
    èļģ
    0.08
    pill
    0.08
     Reed
    0.08
    çŃ¾åIJį
    0.08
    汤
    0.08
    åĦ¿
    0.08
     ÑĥÑģили
    0.07
    Activations Density 0.051%

    No Known Activations