© 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. 4266
    Prev
    Next
    INDEX
    Explanations

    University or CU followed by specific names

    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
    çĤľ
    -0.09
    主é¢ĺæķĻèĤ²
    -0.09
    Pont
    -0.09
    æĹ¶ä»£ä¸ŃåĽ½
    -0.09
    ous
    -0.09
    otte
    -0.08
    elden
    -0.08
    çĤ¹äºĨ
    -0.08
    ables
    -0.08
    æ¾İ
    -0.08
    POSITIVE LOGITS
    诰
    0.10
    å¤§åľ°
    0.09
    éĺħ
    0.09
    å£
    0.09
    ackle
    0.09
     unins
    0.08
    éĢļçŁ¥ä¹¦
    0.08
    èģĴ
    0.08
    iegel
    0.08
    ç¼ħ
    0.08
    Activations Density 0.008%

    No Known Activations