© 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. 28077
    Prev
    Next
    INDEX
    Explanations

    condition, mediated, immunity, protector

    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
    åij³
    -0.10
    å£ħ
    -0.09
    è«
    -0.08
    erialize
    -0.08
    SOFTWARE
    -0.08
    饷
    -0.08
    stag
    -0.08
    LEM
    -0.08
    çīĮ
    -0.08
    pes
    -0.08
    POSITIVE LOGITS
    ä¼ļåijĺ
    0.10
    è£ģ
    0.10
    åıĹæįŁ
    0.09
    uard
    0.09
    ä¾Ľæ±Ĥ
    0.09
    æĪ¬
    0.09
    派驻
    0.09
    ç¿Ķ
    0.08
     protector
    0.08
    ingly
    0.08
    Activations Density 0.116%

    No Known Activations