© 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. 53126
    Prev
    Next
    INDEX
    Explanations

    paper products

    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
    transparent
    -0.08
    å¥Ķ
    -0.08
     Flesh
    -0.08
    Specifier
    -0.08
    rong
    -0.08
    apos
    -0.08
     privé
    -0.08
     transparent
    -0.08
    es
    -0.08
    bsp
    -0.08
    POSITIVE LOGITS
    å±ij
    0.11
    鸢
    0.10
    stick
    0.09
    matic
    0.09
    æŁ¬
    0.09
    maid
    0.09
    wort
    0.08
    åĪ¶åĵģ
    0.08
    .CheckBox
    0.08
    èģĶéĤ¦
    0.08
    Activations Density 0.014%

    No Known Activations