© 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. 62784
    Prev
    Next
    INDEX
    Explanations

    "soup"

    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
    举
    -0.09
    uby
    -0.08
    obb
    -0.08
    åĭĥåĭĥ
    -0.08
    Ĩ
    -0.08
    ©
    -0.08
    esium
    -0.08
    轸
    -0.08
    neo
    -0.08
    é©°
    -0.08
    POSITIVE LOGITS
     Bowl
    0.10
    ier
    0.10
    lette
    0.10
    çĽİ
    0.10
     chuyá»ĩn
    0.09
    å¦ŀ
    0.09
    åĵĩ
    0.09
    è¡·
    0.09
    ftware
    0.09
    寮
    0.09
    Activations Density 0.009%

    No Known Activations