© 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. 31004
    Prev
    Next
    INDEX
    Explanations

    earth, demons, evil, god

    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
    é¼
    -0.09
     versus
    -0.08
    ä¹ĭ称
    -0.08
    EXP
    -0.08
    isers
    -0.08
    æĸ¹æīį
    -0.08
    çķ
    -0.08
    æ´Ĺ涤
    -0.08
    coc
    -0.08
    为æŃ¢
    -0.08
    POSITIVE LOGITS
    å¿Ļç¢Į
    0.09
    [unit
    0.09
    [email
    0.09
    éĺŁåijĺ们
    0.08
    [char
    0.08
    ãĥIJãĥ¼
    0.08
    ä¸ĭæ²ī
    0.08
    åĩĨ
    0.08
    æĸ°äººçݩ家
    0.08
    .runner
    0.08
    Activations Density 0.161%

    No Known Activations