© 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. 34118
    Prev
    Next
    INDEX
    Explanations

    free some original actual

    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
    åħ¸ç¤¼
    -0.11
    åŁ¹è®ŃçıŃ
    -0.10
    èķĪ
    -0.09
    ghan
    -0.09
    ä¸Ģ个æĺ¯
    -0.09
     водо
    -0.09
    reau
    -0.09
    çݩ家们
    -0.09
    åı®
    -0.09
    èĤ¡æ°ij
    -0.09
    POSITIVE LOGITS
     stuff
    0.14
     footage
    0.13
    stuff
    0.12
     legislation
    0.12
     territory
    0.11
     evidence
    0.11
    Stuff
    0.11
    agi
    0.10
    illi
    0.10
     terrain
    0.10
    Activations Density 0.272%

    No Known Activations