© 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. 4385
    Prev
    Next
    INDEX
    Explanations

    differences and distributions

    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
     ÙĪبعد
    -0.10
    ç¨Ĺ
    -0.09
     Bail
    -0.08
    ãĥĬãĥ³
    -0.08
    shal
    -0.08
    ëĭĪê¹Į
    -0.08
    HAL
    -0.08
    izon
    -0.08
    å°±åĥı
    -0.08
    ä»Ģä¹ĪæĦıæĢĿ
    -0.08
    POSITIVE LOGITS
    çļĦåİŁåĽł
    0.14
    ä¹ĭæīĢ以
    0.13
    æĺ¯åĽłä¸º
    0.12
    è¿Ļä¸ĢåĪĩ
    0.12
    å¦ĤæѤ
    0.12
    é¢ī
    0.11
    åİŁåĽłæĺ¯
    0.11
    çļĦä¸ĢåĪĩ
    0.10
    è¿Ļæĺ¯åĽłä¸º
    0.10
    å¿«éĢŁå¢ŀéķ¿
    0.10
    Activations Density 0.235%

    No Known Activations