© 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. 61392
    Prev
    Next
    INDEX
    Explanations

    abstract / deciphering

    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
    protocols
    -0.09
    _IA
    -0.08
    aupt
    -0.08
    hin
    -0.08
    åģ¶çĦ¶
    -0.08
    GINE
    -0.08
    æľīä¸įå°ij
    -0.08
     jer
    -0.08
    è¹
    -0.08
    åĭĥåĭĥ
    -0.08
    POSITIVE LOGITS
    èĪĽ
    0.10
    éĻĤ
    0.09
    æīģ
    0.09
    人寿
    0.09
    åŁºçĿ£å¾Ĵ
    0.09
    è§Į
    0.08
    åIJ¸
    0.08
    éº
    0.08
    çĸ®
    0.08
    躺çĿĢ
    0.08
    Activations Density 0.002%

    No Known Activations