© 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. 3892
    Prev
    Next
    INDEX
    Explanations

    originating from

    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
    adin
    -0.09
    ç»ĵ
    -0.09
    asses
    -0.09
    лÑıÑĢ
    -0.08
    bia
    -0.08
    çµIJ
    -0.08
    æ¤ħ
    -0.08
    èµ°ä¸Ĭ
    -0.08
    ackle
    -0.08
    ç»Ŀä½³
    -0.08
    POSITIVE LOGITS
     elsewhere
    0.10
    éķ¿éķ¿
    0.10
    ä¹ħä¹ħ
    0.10
     previous
    0.10
    å¤ĸåľ°
    0.09
    å¤ĸæĿ¥
    0.09
    åij¨è½¬
    0.09
    åħ¶ä»ĸ
    0.09
    åĪ«çļĦ
    0.09
    æŁIJç§į
    0.09
    Activations Density 0.075%

    No Known Activations