© 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. 4841
    Prev
    Next
    INDEX
    Explanations

    additional details

    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
    rup
    -0.10
    avr
    -0.10
    ungs
    -0.09
    çĨĻ
    -0.09
    ãģ¿ãģª
    -0.08
    bir
    -0.08
    æĺ¯æľī
    -0.08
    oration
    -0.08
     rumor
    -0.08
    rum
    -0.08
    POSITIVE LOGITS
    çĿ¡
    0.09
    åıij表äºĨ
    0.09
    nea
    0.09
    çĸ²æĥ«
    0.08
    çĥģ
    0.08
    iable
    0.08
    éĢĥ离
    0.08
    åıij表
    0.08
    æĥ³èµ·
    0.08
    åħ¨éķ¿
    0.08
    Activations Density 0.011%

    No Known Activations