© 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. 7356
    Prev
    Next
    INDEX
    Explanations

    calculating totals and averages

    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
    emacs
    -0.09
    /Instruction
    -0.08
    presence
    -0.08
    eln
    -0.08
    OKIE
    -0.08
    etal
    -0.08
    _quad
    -0.08
    ergy
    -0.08
    .fromLTRB
    -0.08
    è¿ĩæĿ¥
    -0.08
    POSITIVE LOGITS
    utt
    0.11
    çļĦæķ´ä½ĵ
    0.10
    åıĹå½±åĵį
    0.09
    atori
    0.09
    æķ´ä½ĵ
    0.09
    æŃ£åĵģ
    0.09
    çļĦéĤ£ä¸ª
    0.09
    èģ¿
    0.09
    lix
    0.08
    被æīĵ
    0.08
    Activations Density 0.064%

    No Known Activations