© 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. 33829
    Prev
    Next
    INDEX
    Explanations

    parts of

    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
    hta
    -0.09
    à¸Ńร
    -0.09
     ty
    -0.09
    _partitions
    -0.08
     tm
    -0.08
    ẩm
    -0.08
    ady
    -0.08
    åħ¨éĿ¢
    -0.08
    WSC
    -0.08
    keywords
    -0.08
    POSITIVE LOGITS
    пÑĥ
    0.11
    鼶ç¢İ
    0.11
     пÑĥ
    0.10
    çīĩ段
    0.10
     à¹Ĩ
    0.10
    们
    0.09
    触åĬ¨
    0.09
    ç®ĹäºĨ
    0.09
    çݯèĬĤ
    0.09
    amount
    0.09
    Activations Density 0.135%

    No Known Activations