© 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. 7565
    Prev
    Next
    INDEX
    Explanations

    by its nature

    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
    彪
    -0.11
    å¿ĥ头
    -0.09
    thro
    -0.09
    åĨįä¹Łæ²¡æľī
    -0.09
    è¿ĩåİ»çļĦ
    -0.09
    ADOS
    -0.09
    çIJ¬
    -0.09
    åħ¶ä¸Ńä¹ĭä¸Ģ
    -0.09
    æĹłæķĮ
    -0.09
    æ°´åĪĨ
    -0.09
    POSITIVE LOGITS
    6
    0.10
    endor
    0.09
    7
    0.09
    3
    0.08
    9
    0.08
    8
    0.08
    2
    0.08
    5
    0.08
    whatever
    0.08
    åĿIJæłĩ
    0.08
    Activations Density 0.010%

    No Known Activations