© 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. 17173
    Prev
    Next
    INDEX
    Explanations

    let / object / what / which

    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
    å°ıåŃ©åŃIJ
    -0.10
    åĢļ
    -0.09
    æĤ£
    -0.08
    coop
    -0.08
     unreasonable
    -0.08
    骨干
    -0.08
    çļĦæ¯Ķä¾ĭ
    -0.08
    IBC
    -0.08
    trs
    -0.08
    ths
    -0.08
    POSITIVE LOGITS
    è¦ģä¹Ī
    0.09
    retch
    0.09
    æĹ¥åĨħ
    0.09
    产åĩº
    0.09
    ected
    0.09
    classic
    0.08
    产åĢ¼
    0.08
    èģĶåIJĪ
    0.08
    交代
    0.08
    atra
    0.08
    Activations Density 0.003%

    No Known Activations