© 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. 30934
    Prev
    Next
    INDEX
    Explanations

    regardless of

    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
    åĽºçĦ¶
    -0.10
     zwar
    -0.10
    ãģłãģijãģ§ãģªãģı
    -0.09
    ãģ¯ãĤĤãģ¡ãĤįãĤĵ
    -0.09
    ë¿IJë§Į
    -0.09
    è¦ģä¹Ī
    -0.09
    æĢİä¹ĪåĬŀ
    -0.08
     Ke
    -0.08
    iske
    -0.08
    ke
    -0.08
    POSITIVE LOGITS
    åıªè¦ģæĺ¯
    0.12
     ë¶Ī구íķĺê³ł
    0.12
    ä¸Ģå¾ĭ
    0.10
    OLT
    0.10
    mans
    0.09
    ä¸įå¿ĺ
    0.09
    ä¸ĢåĪ»
    0.09
    åıªè¦ģ
    0.09
    éĥ½ä¼ļæľī
    0.09
     ultimately
    0.09
    Activations Density 0.062%

    No Known Activations