© 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. 18097
    Prev
    Next
    INDEX
    Explanations

    second half 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
    ç¼²
    -0.11
    rated
    -0.09
    ered
    -0.08
    åķ¬
    -0.08
     particular
    -0.08
     handing
    -0.08
    ä¸įåĩºæĿ¥
    -0.08
     Hansen
    -0.08
    lessly
    -0.08
     Trot
    -0.08
    POSITIVE LOGITS
    å·¥ä½ľæĢ»ç»ĵ
    0.09
     spent
    0.09
    å°¤åħ¶æĺ¯
    0.09
    缮åħī
    0.09
    åīĸæŀIJ
    0.09
    æĽĻåħī
    0.08
    lington
    0.08
    æľĭåıĭåľĪ
    0.08
    Chance
    0.08
    å®Ĺ
    0.08
    Activations Density 0.020%

    No Known Activations