© 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. 7042
    Prev
    Next
    INDEX
    Explanations

    end or ends

    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
    è¿ĺä¸įå¤Ł
    -0.10
    èĮģ
    -0.10
    å©Ĭ
    -0.10
    伫
    -0.09
    京
    -0.09
    æ·¤
    -0.09
    باÙĦ
    -0.09
    èij«èĬ¦
    -0.09
    éģ®
    -0.09
    åı¯è§Ĩ
    -0.08
    POSITIVE LOGITS
    quis
    0.08
    触åıĬ
    0.08
    æĦ¿
    0.08
    è¡Ķæİ¥
    0.08
    ToFront
    0.08
    á»İ
    0.08
     part
    0.08
    ãģªãĤĬ
    0.08
    éĶī
    0.08
    èµ°åħ¥
    0.08
    Activations Density 0.107%

    No Known Activations