© 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. 56979
    Prev
    Next
    INDEX
    Explanations

    bright, brighter, brightest

    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
     tall
    -0.09
    æĶ¯æĮģåĬĽåº¦
    -0.09
    è´§æºIJ
    -0.09
    çĮĹ
    -0.09
    dra
    -0.08
    agu
    -0.08
    é«ĺå°ļ
    -0.08
    abouts
    -0.08
    iph
    -0.08
    ese
    -0.08
    POSITIVE LOGITS
    亮
    0.17
    èĢĢçľ¼
    0.15
    çģ¿çĥĤ
    0.14
     bright
    0.13
    -eyed
    0.13
    çħ§äº®
    0.12
    çĥģ
    0.12
     brightest
    0.12
    bright
    0.12
     Bright
    0.12
    Activations Density 0.040%

    No Known Activations