© 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. 56050
    Prev
    Next
    INDEX
    Explanations

    DO NOT comments

    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
    åľ¨åĨħçļĦ
    -0.10
    åħ¬å®īå±Ģ
    -0.10
    éĺ´è°ĭ
    -0.09
    河边
    -0.09
    æ½ľåĬĽ
    -0.09
    odore
    -0.09
    éļıæŶéļı
    -0.08
    ducted
    -0.08
    JOB
    -0.08
    ä¼ģä¸ļ家
    -0.08
    POSITIVE LOGITS
    yth
    0.09
    cape
    0.08
    paced
    0.08
    enberg
    0.08
    weise
    0.08
    Massage
    0.08
    ä¹Ŀå·ŀ
    0.07
    裹
    0.07
    aza
    0.07
    ay
    0.07
    Activations Density 1.024%

    No Known Activations