© 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. 62662
    Prev
    Next
    INDEX
    Explanations

    included in

    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
    asher
    -0.10
    ä¸ĵä¸ļçŁ¥è¯Ĩ
    -0.09
    oyer
    -0.09
    antis
    -0.09
    oti
    -0.09
     BASIS
    -0.08
    bero
    -0.08
    æīĭä¸Ĭ
    -0.08
    ODO
    -0.08
    ocked
    -0.08
    POSITIVE LOGITS
    èĢĥéĩı
    0.12
    è¦ģèĢĥèĻij
    0.09
    å½Ģ
    0.09
    åĽĬ
    0.09
    cludes
    0.09
    æĪijè¦ģ
    0.08
    Clark
    0.08
    putation
    0.08
    coverage
    0.08
    ilk
    0.08
    Activations Density 0.055%

    No Known Activations