© 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. 38663
    Prev
    Next
    INDEX
    Explanations

    things that can be seen

    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
    éĤij
    -0.10
    ãĤıãĤĬ
    -0.09
    terior
    -0.09
    oot
    -0.09
    æĹıèĩªæ²»åİ¿
    -0.09
    æľīä¿¡å¿ĥ
    -0.08
    uria
    -0.08
    orough
    -0.08
    оÑģÑĮ
    -0.08
    è¿ijè·Ŀ离
    -0.08
    POSITIVE LOGITS
    seeing
    0.17
    IFS
    0.11
    ä¸Ģçľ¼
    0.10
    INET
    0.10
    mares
    0.10
    ibilities
    0.10
    acular
    0.10
    less
    0.09
    ITES
    0.09
    andard
    0.09
    Activations Density 0.012%

    No Known Activations