© 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. 17809
    Prev
    Next
    INDEX
    Explanations

    oxidation processes

    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
     squeez
    -0.09
    upon
    -0.09
    .FLAG
    -0.09
    bury
    -0.09
    æµģåĬ¨æĢ§
    -0.09
    lessly
    -0.08
    OnError
    -0.08
    èģĮä¸ļéģĵå¾·
    -0.08
    reh
    -0.08
    said
    -0.08
    POSITIVE LOGITS
    åıijè¾¾
    0.10
    æħĭ
    0.10
    æĢģ
    0.09
    esti
    0.09
    åģ¾
    0.09
    pis
    0.09
    èĢ·
    0.08
    çĦ°
    0.08
    eh
    0.08
    aina
    0.08
    Activations Density 0.145%

    No Known Activations