© 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. 56386
    Prev
    Next
    INDEX
    Explanations

    speaking out or forward

    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
    alm
    -0.10
    æĺİçĻ½äºĨ
    -0.08
     sher
    -0.08
    ä¸Ńå°ı
    -0.08
     paul
    -0.08
    DS
    -0.08
    ÑĮеÑĢ
    -0.08
     Morr
    -0.08
    æĶ¶æĭ¾
    -0.08
    mor
    -0.08
    POSITIVE LOGITS
    éĹº
    0.11
    /rfc
    0.10
     against
    0.09
    åĻ¨å®ĺ
    0.09
    çĤ³
    0.09
    æ§Ľ
    0.09
    滥
    0.09
    acies
    0.08
     gros
    0.08
    ç¼ł
    0.08
    Activations Density 0.010%

    No Known Activations