© 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. 23232
    Prev
    Next
    INDEX
    Explanations

    script for context

    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
    -Jul
    -0.10
    WER
    -0.09
    ï½·
    -0.09
    ense
    -0.09
    heart
    -0.08
    Knight
    -0.08
    æĮĩ令
    -0.08
    éľĢè¦ģ注æĦı
    -0.08
    umm
    -0.08
    å°Ħ
    -0.08
    POSITIVE LOGITS
    ableObject
    0.11
    ellen
    0.10
    urally
    0.10
    èĥĸåŃIJ
    0.10
    edes
    0.09
    åı°è¯į
    0.09
    ural
    0.09
     kidd
    0.09
    æĢ§å¼º
    0.09
    ended
    0.09
    Activations Density 0.197%

    No Known Activations