© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9342
    Prev
    Next
    INDEX
    Explanations

    brocade / brooke / broken / broiler / brochu

    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
    uate
    -0.13
    ãģ¹ãģį
    -0.12
    ities
    -0.10
    oral
    -0.10
    --------------------------------
    -0.10
    nette
    -0.10
    _DIP
    -0.09
    emes
    -0.09
    lico
    -0.09
    frica
    -0.09
    POSITIVE LOGITS
    ðŁı»
    0.12
    ../../../
    0.12
    ccoli
    0.11
    posal
    0.11
    å§Ĭ妹
    0.10
    lier
    0.10
    å³Ļ
    0.10
    oklyn
    0.10
    ument
    0.10
    er
    0.09
    Activations Density 0.029%

    No Known Activations