© 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. 55656
    Prev
    Next
    INDEX
    Explanations

    NSFW or unsafe language

    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
    oser
    -0.10
    æľĢå¿«
    -0.09
    ниÑĨа
    -0.09
     strongest
    -0.08
    langs
    -0.08
    è¿ĺ好
    -0.08
    åݲ
    -0.08
    è¯Ģ
    -0.08
     RID
    -0.08
    满æĢĢ
    -0.08
    POSITIVE LOGITS
    abee
    0.10
    èĶ½
    0.09
    æĿµ
    0.09
    æľīä»»ä½ķ
    0.09
    æ»ŀ
    0.08
    åħ³åĪĩ
    0.08
    itution
    0.08
     Wolfe
    0.08
    çĥ¯
    0.08
    éĹ¾
    0.08
    Activations Density 0.002%

    No Known Activations