© 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. 56146
    Prev
    Next
    INDEX
    Explanations

    capsule endoscopy

    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
    liness
    -0.12
    fully
    -0.10
    minster
    -0.10
    est
    -0.09
     converse
    -0.09
    liest
    -0.09
    EMENT
    -0.09
    éĩijåįİ
    -0.08
    istar
    -0.08
    ãĤĩãģĨ
    -0.08
    POSITIVE LOGITS
    olini
    0.10
    arian
    0.09
    ary
    0.09
     tÆ°á»Łng
    0.09
    _Float
    0.09
    累累
    0.09
    isu
    0.08
    ãĥ³ãĤ°
    0.08
    riot
    0.08
     Americ
    0.08
    Activations Density 0.035%

    No Known Activations