© 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. 1262
    Prev
    Next
    INDEX
    Explanations

    Spring + Security/framework/balance/mechanism/afternoon

    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
    aceous
    -0.10
    ·
    -0.10
    æŃī
    -0.10
    æ°ļ
    -0.09
    çŃIJ
    -0.09
    anos
    -0.09
    ulin
    -0.09
    asis
    -0.09
     Fist
    -0.09
    éķ¿è¿ľ
    -0.08
    POSITIVE LOGITS
    ripper
    0.11
    manship
    0.11
    èĩ¾
    0.10
    ularity
    0.10
    ioni
    0.10
    steen
    0.10
    roups
    0.10
    สà¸Ńà¸ļ
    0.10
    cloak
    0.09
    board
    0.09
    Activations Density 0.040%

    No Known Activations