© 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. 1394
    Prev
    Next
    INDEX
    Explanations

    testing or tried

    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
     Primitive
    -0.09
     primitive
    -0.09
    atum
    -0.09
    为èĩªå·±
    -0.08
    欣
    -0.08
    .foundation
    -0.08
    å®ľ
    -0.08
     Bauer
    -0.08
    fusion
    -0.08
    æĻ®
    -0.08
    POSITIVE LOGITS
    èĢĥéªĮ
    0.20
    æ£ĢéªĮ
    0.12
    æµĭè¯ķ
    0.12
     tests
    0.12
    è°ģèĥ½
    0.12
    罡
    0.11
    æ±ľ
    0.11
    羣çĪ±
    0.10
    -testing
    0.10
    .test
    0.10
    Activations Density 0.080%

    No Known Activations