© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 6979
    Prev
    Next
    INDEX
    Explanations

    scorpion or scorch or endometriosis

    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
    ings
    -0.10
    è¿ĩ硬
    -0.09
    Licensed
    -0.09
    éģij
    -0.09
    棹
    -0.09
    âĻŃ
    -0.09
    æĴŀ
    -0.08
     облаÑģÑĤи
    -0.08
    thed
    -0.08
    rite
    -0.08
    POSITIVE LOGITS
    orp
    0.10
    пÑĥÑģ
    0.10
    HELL
    0.10
    pline
    0.09
    ate
    0.09
    å°¾
    0.09
    ective
    0.09
    èľĩ
    0.09
    ida
    0.08
    çIJĨ解åĴĮ
    0.08
    Activations Density 0.008%

    No Known Activations