© 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. 2082
    Prev
    Next
    INDEX
    Explanations

    legendary

    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
    isoft
    -0.09
    ãģĤãģ£ãģ¦
    -0.09
    åľŃ
    -0.08
    lessly
    -0.08
    eds
    -0.08
    step
    -0.08
     readily
    -0.07
     vicinity
    -0.07
    mates
    -0.07
     specialty
    -0.07
    POSITIVE LOGITS
    èĪ¬çļĦ
    0.10
    ry
    0.10
    ivel
    0.09
     feats
    0.09
    åŀĵ
    0.09
    auge
    0.09
    峡
    0.09
    apis
    0.09
    rarian
    0.09
    令
    0.09
    Activations Density 0.007%

    No Known Activations