© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50514
    Prev
    Next
    INDEX
    Explanations

    wish to

    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
    ask
    -0.10
    åħļå§Ķ书记
    -0.09
    æĪijè§īå¾Ĺ
    -0.09
    aris
    -0.09
    vice
    -0.09
    atoes
    -0.09
    æľĪä¸ĭ
    -0.09
    /or
    -0.08
    à¹īว
    -0.08
     giác
    -0.08
    POSITIVE LOGITS
    fully
    0.17
    lessly
    0.14
    zelf
    0.11
    ernet
    0.10
    ãģĿãĤĮãģ§
    0.10
    ful
    0.10
    dam
    0.10
    olla
    0.10
    iness
    0.10
    fulness
    0.09
    Activations Density 0.010%

    No Known Activations