© 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. 27920
    Prev
    Next
    INDEX
    Explanations

    (source attribution)

    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
    æ®Ĭ
    -0.09
    ÑĤÑĥÑĢ
    -0.09
    омеÑĢ
    -0.08
    æ±Ģ
    -0.08
     attract
    -0.08
     proverb
    -0.08
    åİŁçIJĨ
    -0.07
    åı¸
    -0.07
    .googleapis
    -0.07
     genie
    -0.07
    POSITIVE LOGITS
    inta
    0.11
    .runner
    0.10
    æŃ»èĢħ
    0.09
    instr
    0.09
    åĤ§
    0.09
    åĿij
    0.09
    æIJħ
    0.09
    婧
    0.09
     reporter
    0.08
    .names
    0.08
    Activations Density 0.015%

    No Known Activations