© 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. 62014
    Prev
    Next
    INDEX
    Explanations

    ghost stories, lights, crabs

    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
    ookie
    -0.10
    aurus
    -0.09
    uba
    -0.09
    好çļĦ
    -0.09
    .Selenium
    -0.09
     therap
    -0.08
    ÑģÑģ
    -0.08
    èŀį
    -0.08
    转åıij
    -0.08
     ÅŁek
    -0.08
    POSITIVE LOGITS
    busters
    0.13
    nier
    0.10
    æIJŀ好
    0.09
    (er
    0.09
    ä¸įæŃ»
    0.09
    çĮ·
    0.09
    fulness
    0.08
    èģĶç³»æĪij们
    0.08
    çĻ»éĻĨ
    0.08
    yun
    0.08
    Activations Density 0.010%

    No Known Activations