© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37901
    Prev
    Next
    INDEX
    Explanations

    here

    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
    ãģĵãģĵãģ§
    -0.11
    åľ¨è¿ĻéĩĮ
    -0.11
    åľ¨æѤ
    -0.11
    æĿ¥è¿ĻéĩĮ
    -0.10
    åľ¨åħ¶
    -0.10
    ä¸Ńæľī
    -0.10
    åľ¨è¿Ļ个
    -0.10
    æŃ¤é¡¹
    -0.10
    è¿Ļç¬Ķ
    -0.10
    è¿Ļä¸ĢåĪĩ
    -0.10
    POSITIVE LOGITS
    abouts
    0.24
    -même
    0.13
    after
    0.12
    èĢģ人家
    0.11
    /on
    0.11
    wards
    0.10
    /out
    0.10
    /to
    0.10
    ìłĢ
    0.09
    OrUpdate
    0.09
    Activations Density 0.083%

    No Known Activations