© 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. 39484
    Prev
    Next
    INDEX
    Explanations

    chocolate, sugar, acid

    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
    碰æĴŀ
    -0.09
    oker
    -0.09
    è¾ħ
    -0.08
    åĨ·æ°´
    -0.08
     XPAR
    -0.08
     Prescott
    -0.08
    èĭįçĻ½
    -0.08
    çł´
    -0.08
    碰
    -0.08
    æĴŀ
    -0.08
    POSITIVE LOGITS
    æµĵéĥģ
    0.10
    åĩºèº«
    0.10
    çĭ±
    0.10
    lay
    0.09
    å°±å¾Ī
    0.09
    æµĵæµĵçļĦ
    0.09
    æķijæĬ¤
    0.09
    æ´Ħ
    0.09
    IRD
    0.09
    å°±éľĢè¦ģ
    0.09
    Activations Density 0.001%

    No Known Activations