© 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. 62973
    Prev
    Next
    INDEX
    Explanations

    application or request

    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
    ä¸ĢéŨ
    -0.09
    uddle
    -0.09
    ç±¼
    -0.08
    ä¾µ
    -0.08
    Utf
    -0.08
     budd
    -0.08
    agara
    -0.08
    aje
    -0.07
    èĩªå·±çļĦ
    -0.07
    ä¸Ģåı°
    -0.07
    POSITIVE LOGITS
    æŸ
    0.10
    steller
    0.10
    èµĦ产管çIJĨ
    0.10
    éģ®
    0.10
    空åīį
    0.09
    å¹²éĥ¨éĺŁä¼į
    0.09
    å½ĸ
    0.09
    åĪ¶åº¦æĶ¹éĿ©
    0.09
    çĽĹçªĥ
    0.09
    nde
    0.08
    Activations Density 0.195%

    No Known Activations