© 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. 9587
    Prev
    Next
    INDEX
    Explanations

    litigation or liturgical

    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
    ammen
    -0.09
    mitter
    -0.09
    ascript
    -0.09
    ei
    -0.09
    ðŁı»
    -0.08
    å·¥æľŁ
    -0.08
    sWith
    -0.08
    stag
    -0.08
    æĢ»çIJĨ
    -0.08
     brittle
    -0.08
    POSITIVE LOGITS
    urgy
    0.13
    urgical
    0.12
    ary
    0.09
    retch
    0.09
    ting
    0.09
    rary
    0.09
    mus
    0.09
    ally
    0.09
    ysis
    0.09
    owski
    0.09
    Activations Density 0.038%

    No Known Activations