© 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. 54989
    Prev
    Next
    INDEX
    Explanations

    causes which

    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
     aging
    -0.07
    radio
    -0.07
    ä½İè°ĥ
    -0.07
    æĭħ忧
    -0.07
    oward
    -0.06
    Çİ
    -0.06
    typically
    -0.06
    ḿ
    -0.06
    æĭħå¿ĥ
    -0.06
    èĤĥ
    -0.06
    POSITIVE LOGITS
    åħ»æĪIJ
    0.09
    æ³¢çī¹
    0.08
    ä¸Ńä»ĭæľºæŀĦ
    0.08
    å¨ģåIJį
    0.08
    [P
    0.08
    äºĶåħŃ
    0.08
    .IsNullOr
    0.08
    äºĭæķħ
    0.08
    è¯ķè¡Į
    0.08
    _DAC
    0.08
    Activations Density 0.252%

    No Known Activations