© 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. 1535
    Prev
    Next
    INDEX
    Explanations

    Marvel Comics

    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
    çĻ¸
    -0.09
    æĦŁæĥħ
    -0.09
    伪
    -0.09
    ters
    -0.09
    éĩİå¤ĸ
    -0.09
    gart
    -0.08
    mates
    -0.08
    ê°ģ
    -0.08
    lle
    -0.08
    èī°èĭ¦
    -0.08
    POSITIVE LOGITS
    ously
    0.16
    ous
    0.15
    led
    0.11
    ably
    0.10
    able
    0.10
    lez
    0.10
    åİĨç¨ĭ
    0.09
    ä¸Ģ声
    0.09
    è°ij
    0.09
    GLOBALS
    0.09
    Activations Density 0.005%

    No Known Activations