© 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. 28134
    Prev
    Next
    INDEX
    Explanations

    developing applications and systems

    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
    æĭį
    -0.08
    éĢļè¿ĩ对
    -0.08
    shipment
    -0.08
     mer
    -0.08
    ä¸ºä½ł
    -0.08
    空æ°ĶåĩĢåĮĸ
    -0.08
    沦
    -0.08
    à¹īà¸ĩ
    -0.08
    axes
    -0.07
     lot
    -0.07
    POSITIVE LOGITS
     mirrored
    0.10
    ç»ħ
    0.09
    iens
    0.09
    æĹħ
    0.09
    ä¸įç»ı
    0.09
    isy
    0.09
    gger
    0.09
    NSE
    0.09
    åħħåĪĨä½ĵçİ°
    0.09
    elia
    0.09
    Activations Density 0.173%

    No Known Activations