© 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. 4171
    Prev
    Next
    INDEX
    Explanations

    audacity and presumption

    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
     medicine
    -0.09
    boo
    -0.09
    medicine
    -0.09
    ä¸ĭåįĬ
    -0.09
    æĥ§
    -0.09
    åĪ©çī©
    -0.08
    orthand
    -0.08
    ORAGE
    -0.08
    åĥ§
    -0.08
    gabe
    -0.08
    POSITIVE LOGITS
    åĵī
    0.09
    æµ·åĨħå¤ĸ
    0.09
     aud
    0.09
     Pig
    0.09
    æ´ĭæ´ĭ
    0.09
    大æ°Ķ
    0.08
    2
    0.08
    ç«ŁçĦ¶
    0.08
    å±ħçĦ¶
    0.08
    åĪĬçĻ»
    0.08
    Activations Density 0.010%

    No Known Activations