© 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. 49507
    Prev
    Next
    INDEX
    Explanations

    brute force

    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
    ettes
    -0.10
    istant
    -0.09
     troll
    -0.08
    人éĥ½
    -0.08
    igg
    -0.08
    以æѤ
    -0.08
    fried
    -0.08
    æī¶
    -0.08
     LEG
    -0.08
    CEED
    -0.08
    POSITIVE LOGITS
    å¾Īå¤ļäºĭæĥħ
    0.10
    èĽ®
    0.10
    alem
    0.09
    èįIJ
    0.09
    罡
    0.09
    å¥Ķ
    0.09
    à¹ģละà¸Ħวาม
    0.08
    åıijæĶ¹
    0.08
    .inputs
    0.08
     HOL
    0.08
    Activations Density 0.030%

    No Known Activations