© 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. 32642
    Prev
    Next
    INDEX
    Explanations

    absorb things

    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
    held
    -0.11
    uth
    -0.10
    ä¼¼
    -0.10
    ziel
    -0.09
    inen
    -0.09
    à¸ļรร
    -0.09
    odor
    -0.09
    unity
    -0.09
     held
    -0.08
    actory
    -0.08
    POSITIVE LOGITS
    åIJ¸æĶ¶
    0.13
     absorb
    0.11
    æİīäºĨ
    0.11
     absor
    0.10
     absorbs
    0.10
     absorbing
    0.10
    -abs
    0.10
     absorbed
    0.10
    ç²¾åįİ
    0.10
    æİī
    0.09
    Activations Density 0.184%

    No Known Activations