© 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. 4965
    Prev
    Next
    INDEX
    Explanations

    followed by not

    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
    atat
    -0.09
    çĸµ
    -0.09
    NR
    -0.09
    讯
    -0.09
    ITT
    -0.08
    çĿĢæīĭ
    -0.08
    æħĪ
    -0.08
    rots
    -0.08
    eneg
    -0.08
    rawn
    -0.08
    POSITIVE LOGITS
    æľīä»»ä½ķ
    0.10
    rium
    0.09
    è¶Ł
    0.09
    etz
    0.08
    å¿ĺè®°äºĨ
    0.08
    é»į
    0.08
    æĪIJ为ä¸Ģ个
    0.08
    æīĢè°ĵçļĦ
    0.08
    é¾Ľ
    0.08
    åĪĽå®¢
    0.08
    Activations Density 0.030%

    No Known Activations