© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21874
    Prev
    Next
    INDEX
    Explanations

    "I am" or "me"

    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
    è¤IJ
    -0.09
    eyer
    -0.09
    çľ¼åħī
    -0.08
    樱
    -0.08
    æķ¢
    -0.08
    举起
    -0.08
    飩
    -0.08
    нÑıÑĤÑĮ
    -0.08
    -appointed
    -0.08
    éĺľ
    -0.08
    POSITIVE LOGITS
    æ»ĵ
    0.09
    ç³ľ
    0.09
    åŃĺåľ¨äºİ
    0.09
    çģŃ亡
    0.09
    åĩºåľº
    0.09
    asthan
    0.08
    è´ŀ
    0.08
    èħIJ
    0.08
    _under
    0.08
    olph
    0.08
    Activations Density 0.120%

    No Known Activations