© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 40432
    Prev
    Next
    INDEX
    Explanations

    licensed software and code

    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
    Negative Logits
    ally
    -0.11
    -most
    -0.09
    æ²
    -0.09
    ัà¸Ļ
    -0.08
    deo
    -0.08
    nal
    -0.08
     Funk
    -0.08
    åħħåĪĨåĪ©çĶ¨
    -0.08
    енно
    -0.08
    NER
    -0.08
    POSITIVE LOGITS
    楮
    0.10
    DIST
    0.09
     COPYRIGHT
    0.09
    ç¼°
    0.09
    æĥķ
    0.09
    åıªæľīä¸Ģ个
    0.08
    åĩºå¢ĥ
    0.08
    èĥ¥
    0.08
    emen
    0.08
    éĢłæĪIJçļĦ
    0.08
    Activations Density 0.029%

    No Known Activations