© 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. 58706
    Prev
    Next
    INDEX
    Explanations

    architectural style and features

    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
    篦
    -0.08
    æ¯ħ
    -0.08
    _wp
    -0.08
    å·¥åħ·
    -0.08
    DM
    -0.08
    ä»ħä¾ĽåıĤèĢĥ
    -0.08
     notorious
    -0.08
     Hen
    -0.08
    客
    -0.08
    isser
    -0.07
    POSITIVE LOGITS
    ä¸ĢéŨ
    0.11
    ä¸ĬéŨ
    0.11
    éŨæĪ·
    0.10
    .nlm
    0.10
    è£ħ饰
    0.09
    ç«ĭéĿ¢
    0.09
    ä¸Ģ座
    0.09
    éĶ¦æłĩ
    0.09
     orn
    0.09
    æĭ±
    0.09
    Activations Density 0.079%

    No Known Activations