© 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. 26083
    Prev
    Next
    INDEX
    Explanations

    heard of / that / it / about

    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
    atron
    -0.10
    çĤ¹å¤ļ
    -0.09
    åľĥ
    -0.09
    éĢ¶
    -0.09
    po
    -0.09
    _ATTRIB
    -0.08
    keh
    -0.08
    lesh
    -0.08
    çļĦç¬ij容
    -0.08
    -bound
    -0.08
    POSITIVE LOGITS
    åIJ¬è¯´è¿ĩ
    0.23
    åIJ¬è¯´
    0.18
    éĹ»
    0.15
    åIJ¬è¿ĩ
    0.15
    èģŀ
    0.13
    ä¸ĢåIJ¬
    0.13
    èĢ³
    0.13
     heard
    0.11
    heard
    0.11
    åIJ¬è§ģ
    0.11
    Activations Density 0.074%

    No Known Activations