© 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. 12052
    Prev
    Next
    INDEX
    Explanations

    as represented by

    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
    ight
    -0.10
    หลวà¸ĩ
    -0.09
    ä¿£
    -0.09
    adic
    -0.09
    æĹ©æĹ©
    -0.08
    ä¿Ł
    -0.08
    umn
    -0.08
    ometer
    -0.08
    ough
    -0.08
    æĿ¥äºĨ
    -0.08
    POSITIVE LOGITS
     Ñģобой
    0.19
    atively
    0.12
    ertoire
    0.11
    OND
    0.11
    个人è§ĤçĤ¹
    0.10
    ational
    0.10
    ivec
    0.10
    éºĪ
    0.09
    nel
    0.09
    ond
    0.09
    Activations Density 0.052%

    No Known Activations