© 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. 1358
    Prev
    Next
    INDEX
    Explanations

    flanked 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
    棹
    -0.08
    gage
    -0.08
     Idea
    -0.08
    廨
    -0.08
    erate
    -0.08
    osity
    -0.08
    opic
    -0.08
    ECT
    -0.08
    uffs
    -0.08
    gens
    -0.08
    POSITIVE LOGITS
    两端
    0.10
    çĩ§
    0.10
    两边
    0.10
    sters
    0.10
    -mounted
    0.10
    ä¹ĭä½ľ
    0.09
    äºĶæĺŁ
    0.09
    RTL
    0.09
    ings
    0.09
     Rossi
    0.09
    Activations Density 0.016%

    No Known Activations