© 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. 31207
    Prev
    Next
    INDEX
    Explanations

    race and prejudice

    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
    æĢ§åĪ«
    -0.18
     gender
    -0.15
    gender
    -0.12
    .gender
    -0.12
     Gender
    -0.11
    Gender
    -0.11
    çĶ·å¥³
    -0.11
    _gender
    -0.11
    æī©åħħ
    -0.11
    DED
    -0.10
    POSITIVE LOGITS
    æŀķ
    0.10
     Vent
    0.09
    â¾
    0.08
    åħĥ
    0.08
     morb
    0.08
    çīĪ
    0.08
    åį«çĶŁéĹ´
    0.07
    pbs
    0.07
     Mour
    0.07
    Vent
    0.07
    Activations Density 0.066%

    No Known Activations