© 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. 34798
    Prev
    Next
    INDEX
    Explanations

    transfer of power/packets

    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
    åĮ¡
    -0.12
    çĥ©
    -0.10
    ä¸ĢåĪĢ
    -0.10
    åĽ´
    -0.09
    缩水
    -0.09
    åħ¬è®¤
    -0.09
    æİ¥è¿ij
    -0.09
    æ³Ľ
    -0.09
    éŁ¿
    -0.08
    åľį
    -0.08
    POSITIVE LOGITS
    交æİ¥
    0.27
    移交
    0.24
    æİ¥åĬĽ
    0.19
    交ä»ĺ
    0.15
    交代
    0.15
    æݥ管
    0.14
    è¿ĩ渡
    0.14
    æĹłç¼Ŀ
    0.13
    è¡Ķæİ¥
    0.13
    转移åĪ°
    0.12
    Activations Density 0.088%

    No Known Activations