© 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. 4919
    Prev
    Next
    INDEX
    Explanations

    need to be

    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
    åº
    -0.10
    äºĨä¸Ģ份
    -0.08
    Straight
    -0.08
    åģļäºĨ
    -0.08
    ledge
    -0.08
    roy
    -0.08
    sta
    -0.08
    äºĨä¸Ģ个
    -0.08
    第ä¸Ģ次
    -0.08
    ymous
    -0.07
    POSITIVE LOGITS
    æīįè¡Į
    0.25
    æīįèĥ½
    0.22
    è¿Ļæł·æīįèĥ½
    0.19
    æīįåı¯ä»¥
    0.19
    æīįèĥ½å¤Ł
    0.19
    æīįç®Ĺ
    0.15
    æīįä¼ļ
    0.15
    æĸ¹åı¯
    0.14
    æīįæľī
    0.14
    以便
    0.13
    Activations Density 0.277%

    No Known Activations