© 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. 56247
    Prev
    Next
    INDEX
    Explanations

    doctorate from

    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
    ราà¸Ĭ
    -0.10
    æī§
    -0.09
    imen
    -0.09
    onen
    -0.08
    lish
    -0.08
    âĸµ
    -0.08
     Bast
    -0.08
    èħ¾
    -0.08
    rna
    -0.08
    æī§çĿĢ
    -0.08
    POSITIVE LOGITS
    ize
    0.09
    عا
    0.09
    å½Ĵå±ŀ
    0.09
    å½Ĵå±ŀäºİ
    0.09
    ENER
    0.09
    ives
    0.08
    æ®Ĭ
    0.08
    asi
    0.08
    åŬ
    0.08
    æĻķ
    0.08
    Activations Density 0.014%

    No Known Activations